Your SlideShare is downloading. ×
0
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Scot baxtor cdma
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×
Saving this for later? Get the SlideShare app to save on your phone or tablet. Read anywhere, anytime – even offline.
Text the download link to your phone
Standard text messaging rates apply

Scot baxtor cdma

3,433

Published on

0 Comments
3 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total Views
3,433
On Slideshare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
158
Comments
0
Likes
3
Embeds 0
No embeds

Report content
Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
No notes for slide

Transcript

  • 1. Course RF200 Wireless CDMA RF Wireless CDMA RF Performance Optimization Performance OptimizationNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 1
  • 2. Contents Chapter Slide #1. Introduction 12. Foundation Topics Layer-3 Messaging 9 Call Processing 14 Performance Indicators and Problem Signatures 95 PN Planning and Search Windows 1162. Analyzing System Performance 138 System Data and Analysis Techniques 1413. Mobile Field Tools and Data Analysis 191 Autonomous Mobile Data Collection 196 Conventional Field Data Collection Tools 2014. Multiple Carrier Systems: Operating Principles and Analysis 2625. Applied Optimization 2926. 1xRTT Optimization Issues 334Appendix I. Cell Loading Example 405Appendix II. CDMA/3g1x Books, Publications, Web Resources 419November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 2
  • 3. Course RF200 Introduction to Performance Introduction to Performance Optimization OptimizationNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 3
  • 4. Welcome to Course RF200 Course RF100 is an introduction to RF and CDMA principles. After completing it, you should be familiar with: • General RF system design principles • CDMA technology - principles, channels, network basics • key fundamentals of Messaging and Call Processing Course RF200 covers how to recognize and deal with system performance problems • key performance indicators and what they mean • what tools are available for discovering and analyzing problems • mechanisms and situations that cause trouble • how to solve many of the problems you’ll seeNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 4
  • 5. Good Performance is so Simple!! One, Two, or Three good signals in handoffBTS BTS • Composite Ec/Io > -10 db Enough capacity • No resource problems – I’ve got what IBTS need Ec/Io BTS A BTS B BTS C -10 available FORWARD power Traffic LINK Channels In use Paging Sync Pilot November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 5
  • 6. Bad Performance Pilot PollutionBTS BTS Weak Signal Scarce Resources available power • BTS forward power TrafficBTS Channels • BTS receive power In use BTS Rx Pwr Paging • channel elements Sync Pilot • packet pipes Poor System Statistics • High Dropped Calls • High Access Failures Percent Total Drop Call Percentage 5.0% 4.5% %Drops 4.0% 3.5% 3.0% 2.5% 2.0% 1.5% 1.0% 0.5% 0.0% Date November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 6
  • 7. What is Performance Optimization? The words “performance optimization” mean different things to different people, viewed from the perspective of their own jobs System Performance Optimization includes many different smaller processes at many points during a system’s life • recognizing and resolving system-design-related issues (can’t build a crucial site, too much overlap/soft handoff, coverage holes, etc.) • “cluster testing” and “cell integration” to ensure that new base station hardware works and that call processing is normal • “fine-tuning” system parameters to wring out the best possible call performance • identifying causes of specific problems and customer complaints, and fixing them • carefully watching system traffic growth and the problems it causes - implementing short-term fixes to ease “hot spots”, and recognizing problems before they become criticalNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 7
  • 8. Performance Optimization Phases/Activities Phase Drivers/Objectives Activities Main Tools Success Indicators Cover desired area; Plan cells to effectively cover Prop. Models,RF Design and have capacity for as needed and divide traffic Test Transmitters, Model resultsCell Planning anticipated traffic load appropriately planning tools New Cluster Ensure cells properly Drive-test: coverage, all Drive-test tools; All handoffs occur; constructed and Testing and handoff boundaries, all call cell diagnostics and all test cases configured to giveCell Integration events and scenarios hardware test verified normal performanceSolve Specific Identify problems Drive-test tools, Identified Detect, Investigate, Resolve Performance from complaints or system stats, problems are performance problems Problems statistics; fix them! customer reports resolved Well-System Ensure present ‘plant’ Watch stats: Drops, Blocks, Acceptable levels Performance is giving best possible Access Failures; identify/fix hot System statistics and good trends Management performance spots for all indicators Manage congested Watch capacity indicators; Smart optimization Stats-Derived Capacity areas for most identify problem areas, tune of parameters; indicators; carried Optimization effective performance parameters & configuration system statistics traffic levels Sectors are Growth expanded soon hello Management: Overall traffic increases and Predict sector and area Traffic analysis and trending tools; after first signs ofOptimizing both exhaustion: plan and validate congestion; congestion; prop. models for Performance effective growth plan, avoid capital budget competition for capital cell spliiting; carrier and Capital integration impact remains within during tight times additions Effectiveness comfortable bounds November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 8
  • 9. Course RF200 CDMA2000 Layer 3 Messages CDMA2000 Layer 3 MessagesNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 9
  • 10. Messages in CDMA In CDMA, most call processing events are driven by messages Some CDMA channels exist for the sole purpose of carrying messages; they never carry user’s voice traffic • Sync Channel (a forward channel) • Paging Channel (a forward channel) • Access Channel (a reverse channel) • Forward or Reverse Dedicated Control Channels • On these channels, there are only messages, not voice or data Some CDMA channels exist just to carry user traffic • Forward Fundamental and Supplemental Channels • Reverse Fundamental and Supplemental Channels • On these channels, most of the time is filled with traffic and messages are sent only when there is something to do All CDMA messages have very similar structure, regardless of the channel on which they are sentNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 10
  • 11. The Basic Format of CDMA Messages EXAMPLE:CDMA messages on both forward A POWER MEASUREMENTand reverse traffic channels arenormally sent via dim-and-burst REPORT MESSAGEMessages include many fields of Field Lengthbinary data (in bits)The first byte of each message MSG_TYPE (‘00000110’) 8identifies message type: this allows ACK_SEQ 3the recipient to parse the contents MSG_SEQ 3To ensure no messages are ACK_REQ 1missed, all CDMA messages bearserial numbers and important ENCRYPTION 2messages contain a bit requesting ERRORS_DETECTED 5acknowledgment POWER_MEAS_FRAMES 10Messages not promptlyacknowledged are retransmitted LAST_HDM_SEQ 2several times. If not acknowledged, NUM_PILOTS 4the sender may release the call NUM_PILOTS occurrences of this field:Field data processing tools captureand display the messages for study PILOT_STRENGTH 6 t RESERVED (‘0’s) 0-7November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 11
  • 12. Message Vocabulary: Acquisition & Idle States Pilot Channel Sync Channel No Messages Sync Channel Msg Access Channel Paging Channel BTS Registration Msg Access Parameters Msg General Page Msg Order Msg System Parameters Msg Order Msg • Mobile Station Acknowldgment •Base Station Acknowledgment •Lock until Power-Cycled • Long Code Transition Request • Maintenance required • SSD Update Confirmation CDMA Channel List Msg many others….. many others….. Extended System Channel Assignment Origination Msg Parameters Msg Msg Extended Neighbor Page Response Msg List Msg Feature Notification Msg Authentication Challenge Global Service Authentication Response Msg Redirection Msg Challenge Msg Status Response Msg Service Redirection Msg Status Request Msg TMSI Assignment SSD Update Msg TMSI Assignment Msg Completion Message Data Burst Msg Null Msg Data Burst Msg November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 12
  • 13. Message Vocabulary: Conversation State Forward Traffic Channel Order Msg Alert With Reverse Traffic Channel• Base Station Acknowledgment Information Msg • Base Station Challenge Confirmation Service Request Msg Service Request Msg Origination • Message Encryption Mode Continuation Msg Authentication Service Response Msg Service Response Msg Authentication Challenge Challenge Msg Response Msg TMSI Assignment Msg Service Connect Msg Service Connect TMSI Assignment Completion Message Completion Message Send Burst DTMF Msg Service Option Service Option Control Send Burst DTMF Msg Control Msg Message Set Parameters Msg Status Request Msg Status Response Msg Parameters Response Message Power Control Flash With Flash With Power Measurement Parameters Msg. Information Msg Information Msg Report MsgRetrieve Parameters Msg Data Burst Msg Data Burst Message Order Message • Mobile Sta. Acknowledgment Analog Handoff Extended Handoff Pilot Strength •Long Code Transition Direction Msg Direction Msg Measurement Msg Request • SSD Update Confirmation SSD Update Msg Neighbor List Handoff Completion Msg • Connect Update Msg Mobile Station In-Traffic System Registered Msg Parameters Msg November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 13
  • 14. Course RF200 CDMA Call Processing Basics CDMA Call Processing BasicsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 14
  • 15. Troubleshooting Call Processing CDMA call processing is complex! • Calls are a relationship between mobile and system – the events driven by messaging – the channels supported by RF transmission • Multiple codes and channels available for use • Multiple possible problems - physical, configuration, software • Multiple concurrent processes in the mobile and the system Troubleshooting focuses on the desired call events • What is the desired sequence of events? • Compare the actual sequence of events. – What’s missing or wrong? Why did it happen? Messaging is a major blow-by-blow troubleshooting tool RF indications reveal the transmission risks and the channel configurationsBottom Line: To troubleshoot effectively, you’ve got to know call processing steps and details AND the RF basis of the transmissionNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 15
  • 16. Course RF200 Lets Acquire The System! Lets Acquire The System!November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 16
  • 17. What’s In a Handset? How does it work? Digital Rake Receiver Symbols Chips Traffic Correlator summing PN xxx Walsh xx bits Traffic Correlator PN xxx Walsh xx Σ Symbols control Receiver Traffic Correlator ∆t Viterbi Decoder, time-aligned RF Section Convl. Decoder, PN xxx Walsh xx Demultiplexer powerIF, Detector Traffic Correlator Packets AGC PN xxx Walsh xx RF Audio Open Loop MessagesDuplexer Pilot Searcher CPU Vocoder PN xxx Walsh 0 RF Transmit Gain Adjust Audio Messages Transmitter Transmitter Digital Section RF Section Long Code Gen.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 17
  • 18. The Task of Finding the Right System Reverse Link Frequencies Forward Link Frequencies (Mobile Transmit) (Base Station Transmit) 800 MHz. Cellular Spectrum824 MHz. 835 845 849 870 880 890 894 A B Paging, ESMR, etc. A B 825 846.5 869 891.5 1900 MHz. PCS Spectrum unlic. unlic. A D B E F C data voice A D B E F C1850MHz. 1910MHz. 1930MHz. 1990 MHz. FREQUENCY LISTS:Mobile scans forward link frequencies: HISTORY PREFERRED (Cellular or PCS, depending on model) LIST/MRU ROAMING History List (MRU) LIST/PRL Last-used: Preferred Roaming List (PRL) Freq System1 Freq System2 until a CDMA signal is found. Freq System3 Use PRL to find best signal in area. Freq System4 Freq System5 NO CDMA? Try AMPS. No AMPS? Standby etc. etc. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 18
  • 19. The System Determination AlgorithmAt turnon, Idle mobiles use proprietary System Determination Algorithms(SDA) to find the initial CDMA carrier intended for them to useThe mobile finally acquires a CDMA signal and reads the Sync channel • Find the SID & NID in the PRL (Preferred Roaming List) • Check: is there a more-preferred system in the PRL? What Freq(s)? • Go look for the better system Start Preferred MRU Only Bit 0 PRL Acq Idx Yes Go to last Strongest Is better Is SID frequency PN, read SID permitted? from MRU Sync available? No Signal No Denied SID Read Last Resort: Paging Best System Found! GEO escape Channel Begin Normal Paging Channel Operation Or Analog Legend Steps from Steps from Proprietary Typical Mobile the CDMA proprietary SDA standards SDAs databases System Determination Algorithm November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 19
  • 20. 1xRTT Acquisition On the Current Frequency: Find Strongest Pilot, Read Sync Channel All PN Offsets 0 Ec/Io 1. Pilot Searcher Scans the Entire Range of PNs -20 Chips 0 32K PN 0 512 SYNC CHANNEL MESSAGE 2. Put Rake finger(s) on strongest MSG_LENGTH, 28, 28 octets MSG_TYPE, 1, Sync Channel Message available PN, decode Walsh 32, P_REV, 6, IS-2000 Revision 0 MIN_P_REV, 1, J-STD-008 and read Sync Channel Message SID 995, Is this the right system to use? NID 3, PILOT_PN 240 Check the PRL! Active Pilot LC_STATE, 0x00 25 93 12 7C FA, SYS_TIME, 0x02 20 34 B7 53, 10/23/2001 11:02:54Handset Rake Receiver Rake Fingers LP_SEC, 13, LTM_OFF, 54, -660 minutes F1 PN168 W32 DAYLT, 1, Yes F2 PN168 W32 PRAT, 1, 4800 bps RF CDMA_FREQ, 274 (IS-95) ≈ x ≈ F3 PN168 W32 EXT_CDMA_FREQ, 274 (1xRTT) LO Srch PN??? W0 SR1_BCCH_SUPPORTED, 0 SR3_INCL, 0, No Reference PN RESERVED, 0, November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 20
  • 21. PRL Database Guides System Determination Handsets can be programmed with their Preferred Only bit set to True or TRUE False. If True, the handset can only used preferred systems. If False, the Preferred Only Bit FALSE handset can use non-preferred systems, but will prefer preferred systems when available. Acquisition Index There are 29 Acq Indexes in the current PRL. It is normal for some to contain duplicate channels. 0 CDMA channels 350,400 1 CDMA channels 50, 100 2 Analog Block A When the phone Every three minutes idle 3 Analog Block B loses service, it phones rescan for any more- scans the list of preferred signals in the current channels in its Geo Group. This is called current GEO group. “climbing the GEO group”. System Records SID NID PREF GEO Priority Index Roam Indicator 4139 65535 Pref New More 0 Off 59 65535 Pref Same More 2 On 52 65535 Pref Same More 3 Flash Some records are merely analog “Guideposts” to allow the phone to 67 65535 Neg Same Same 3 Short-short-long recognize where it is and position into the 4412 65535 Pref New More 1 Off proper GEO group “GEO confinement”. : : : : : : : 61737 226 Neg New More 0 Off The last system record is not a real system. It merely contains the version65535 is a “wildcard” NID. Preferred “more” number of the PRl and is used by someThe phone is to accept any than the following phones to allow displaying the version.NID it sees on this system. record.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 21
  • 22. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 22
  • 23. Climbing the GEO Group SYSTEM TABLE ACQUISITION TABLE ROAMING LIST NEG/ ACQ ROAM INDEX ACQ TYPE CH1 CH2 CH3 CH4 CH5 CH6 CH7 CH8 CH9 0 6 500 425 825 575 850 325 625 INDEX SID NID PREF GEO PRI INDEX IND 1 6 575 625 500 425 Roaming List Type: IS-683A 2 6 50 100 75 475 825 850 175 250 296 4144 65535 Pref NEW SAME 13 1 Preferred Only: FALSE 297 4812 65535 Pref SAME MORE 21 1 3 6 25 200 350 375 725 50 475 175 250 a GEO GROUP 4 1 Both Default Roaming Indicator: 0 298 205 65535 Pref SAME SAME 4 0 5 6 450 500 350 575 650 Climb! 299 208 65535 Pref SAME MORE 37 0 6 6 675 500 600 575 475 Preferred List ID: 10018 7 6 250 50 175 300 208 65535 Pref SAME SAME 4 0 8 6 550 375 425 625 301 342 65535 Pref SAME MORE 37 0 9 6 75 50 175 250 302 342 65535 Pref SAME SAME 4 0 10 6 200 250 175 50When traveling the first signal 11 6 425 500 575 25 325 650 303 478 65535 Pref SAME SAME 4 0 12 6 500 575 475 25 675 304 1038 65535 Pref SAME SAME 4 0 13 6 500 625 350 50 375 775 575 725 425found is usually not the best 305 306 1050 1058 65535 Pref 65535 Pref SAME SAME SAME SAME 4 4 0 0 14 15 16 6 6 6 650 25 425 500 50 550 675 375 225 25 350 725 75 250 750 425 50 575 175 775one to use 307 308 1375 1385 65535 Pref 65535 Pref SAME SAME SAME MORE 4 4 0 0 17 18 6 6 200 825 50 850 175 925 375 250 19 6 350 325 375 675 25 1175 725 600 100 309 143 65535 Pref SAME MORE 37 0When the SID and NID are 310 143 65535 Pref SAME MORE 4 0 20 21 22 6 6 6 750 325 1150 725 725 1175 775 350 750 375 775 425 575 625looked up in the PRL, they 311 312 4103 4157 65535 Pref 65535 Pref NEW SAME SAME MORE 3 2 1 1 23 24 6 6 350 25 875 1175 325 825 375 1175 200 75 175 250are far down the list of 313 312 65535 Pref SAME SAME 4 0 25 6 50 200 25 100 250 75 a GEO GROUP 26 6 500 1075 850 825 314 444 65535 Pref SAME MORE 37 0 27 1 Aavailable choices 315 316 444 1008 65535 Pref 65535 Pref SAME SAME SAME SAME 4 4 0 0 28 29 30 1 5 5 B A B 317 1012 65535 Pref SAME SAME 4 0The starts at the top of the 31 5 C 318 1014 65535 Pref SAME SAME 4 0 32 5 D 33 5 E 319 1688 65535 Pref SAME MORE 4 0GEO group and works down 320 321 113 113 65535 Pref 65535 Pref SAME SAME MORE 37 SAME 4 0 0 34 35 36 5 4 4 F A Bto the first (most preferred) 322 179 65535 Pref SAME MORE 37 0 37 38 4 6 Both 350 825system it can find 323 324 179 465 65535 Pref 65535 Pref SAME SAME SAME SAME 4 4 0 0 39 40 41 6 6 6 25 675 850 100 600 750 850 1175 775 325 2119 65535 Pref SAME MORE 4 0 42 6 650 • the Acquisition Table is 326 327 2094 1005 65535 Pref 65535 Pref SAME SAME MORE SAME 4 4 0 0 43 44 6 6 450 325 475 350 375 1025 1050 1075 the list of frequencies 328 1013 65535 Pref SAME SAME 4 0 45 46 6 6 150 1025 475 625 675 1050 1075 used by the various PRL: Preferred Roaming List systems, so the mobile Programmed into each phone by the system knows where to search operator; can be updated over the air.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 23
  • 24. Found it! Now we’re on the Right System All PN Offsets 0 Ec/Io 1. Pilot Searcher Scans the Entire Range of PNs -20 Chips 0 32K PN 0 512 SYNC CHANNEL MESSAGE 2. Put Rake finger(s) on strongest 98/05/24 23:14:09.817 [SCH] available PN, decode Walsh 32, MSG_LENGTH = 208 bits MSG_TYPE = Sync Channel Message and read Sync Channel Message P_REV = 3 MIN_P_REV = 2 If PRL shows: Go to the SID = 179 This is the Best Paging Active Pilot NID = 0 Available System! Channel! PILOT_PN = 168 Rake Fingers Offset IndexHandset Rake Receiver LC_STATE = 0x0348D60E013 F1 PN168 W32 SYS_TIME = 98/05/24 23:14:10.160 LP_SEC = 12 F2 PN168 W32 RF LTM_OFF = -300 minutes ≈ x ≈ F3 PN168 W32 DAYLT = 0 LO Srch PN??? W0 PRAT = 9600 bps Ref. RESERVED = 1 PN November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 24
  • 25. Course RF200 After finding the right system: After finding the right system: Normal Paging Channel Operation Normal Paging Channel OperationNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 25
  • 26. The Configuration Messages After reading the Sync Channel, the mobile is now capable of reading the Paging Channel, which it now monitors constantly Before it is allowed to transmit or operate on this system, the mobile must collect a complete set of configuration messages In IS-95, the configuration messages are sent on the Paging Channel, repeated every 1.28 seconds In CDMA2000 systems, the configuration messages may be sent on the separate F-BCH channel • This would be indicated as SR1_BCCH_SUPPORTED = 1 There are six possible types of configuration messages; some are optional; and they may happen in any order The configuration messages contain sequence numbers so the mobile can recognize if any of the messages have been freshly updated as it continues to monitor the paging channel • Access parameters message sequence number • Configuration message sequence number • If a mobile notices a changed sequence number, or if 600 seconds passes since the last time these messages were read, the mobile reads all of them againNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 26
  • 27. Reading the Configuration Messages All PN Offsets 0 Ec/Io -20 Chips 0 32K PN 0 Read the 512 Configuration Messages Access Parameters Msg Keep Rake finger(s) on strongest available PN, monitor Walsh 1, System Parameters Msg the Paging Channel CDMA Channel List Msg Active Pilot Extended System Parameters Msg (*opt.)Handset Rake Receiver Rake Fingers (Extended*) Neighbor List Msg F1 PN168 W01 Global Service F2 PN168 W01 RF Redirection Msg (*opt.) ≈ x ≈ F3 PN168 W01 LO Srch PN??? W0 Now we’re ready to operate!! Reference PN November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 27
  • 28. 1xRTT Access Parameters Message ACCESS PARAMETERS MESSAGE Basic Access Procedure 000035, Time 15:28:37.709, Record 6408, QcpCdmaLogMsgPagingChan Any Access Msg PD: P_REV_IN_USE < 6 MSG_TYPE: Access Parameters Message Success! PILOT_PN: 36 ACC_MSG_SEQ: 2 BTS MS ACC_CHAN: 1 Access Channel(s) Probing NOM_PWR: 3 dB INIT_PWR: -13 dB an Access Probe a Probe Sequence PWR_STEP: 5 dB NUM_STEP: 4 Probe(s) an Access Attempt MAX_CAP_SZ: 6 ACH Frames PAM_SZ: 3 ACH Frame(s) PSIST(0-9): 0 The Access Parameters message PSIST(10): 0 PSIST(11): 0 controls all the steps mobiles must PSIST(12): 0 perform when they transmit on the PSIST(13): 0 Access Channel PSIST(14): 0 PSIST(15): 0 Mobiles perform a trial-and-error MSG_PSIST: 1.00 process called “Probing” to get their REG_PSIST: 1.00 PROBE_PN_RAN: 0 PN chip(s) messages through ACC_TMO: 240 ms PROBE_BKOFF: 1 Slot(s) BKOFF: 1 Slot(s) MAX_REQ_SEQ: 3 MAX_RSP_SEQ: 3 AUTH_MODE: 0 NOM_PWR_EXT: -8 to 7 dB inclusive PSIST_EMG_INCL: No RESERVED: 0November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 28
  • 29. Phone Operation on the Access Channel Successful Basic Access AttemptA sector’s Paging Channel announces 1(typ) to 32 (max) Access Channels: PN Origination Msg ACCESSLong Code offsets for mobiles to use if Success!accessing the system. BTS MS • For mobiles sending Registration, Probing Origination, Page Responses an Access Probe • Base Station always listening! a Probe Sequence an Access AttemptOn the access channel, phones are notyet under BTS closed-loop power control! PAGING Base Sta. Acknlgmt. OrderPhones access the BTS by “probing” at FW TFC TFC frames of 000spower levels determined by receive powerand an open loop formula PAGING Channel Assnmt. Msg. • If “probe” not acknowledged by BTS within ACC_TMO (~400 mS.), phone TFC preamble of 000s RV TFC will wait a random time (~200 mS) FW FC Base Sta. Acknlgmt. Order then probe again, stronger by PI db. • There can be 15 max. (typ. 5) probes Mobile Sta. Ackngmt. Order RV TFC in a sequence and 15 max. (typ. 2) sequences in an access attempt FW TFC Service Connect Msg. • most attempts succeed on first probe! Svc. Connect Complete Msg RV TFCThe Access Parameters message on thepaging channel announces values of all FW TFC Base Sta. Acknlgmt. Orderrelated parameters Call is Established!November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 29
  • 30. 1xRTT System Parameters Message SYSTEM PARAMETERS MESSAGE 000029, Time 15:28:37.607, Record 6330, QcpCdmaLogMsgPagingChan PD: P_REV_IN_USE < 6 MSG_TYPE: System Parameters Message PILOT_PN: 36 CONFIG_MSG_SEQ: 1 SID: 4379 NID: 15 REG_ZONE: 6 TOTAL_ZONES: 3 ZONE_TIMER: 1 min MULT_SIDS: No MULT_NIDS: No BASE_ID: 2155 BASE_CLASS: Public PCS System # Paging Channels, Slotted Mode period PAGE_CHAN: 1 MAX_SLOT_CYCLE_INDEX: 1 HOME_REG: Yes FOR_SID_REG: Yes FOR_NID_REG: Yes POWER_UP_REG: Yes POWER_DOWN_REG: Yes Who Registers? PARAMETER_REG: No REG_PRD: 30.89 min Why & When? BASE_LAT: 37D1835.00N BASE_LONG: 079D1519.00W REG_DIST: 0 SRCH_WIN_A: 60 chips Search Window SRCH_WIN_N: 60 chips SRCH_WIN_R: 80 chips NGHBR_MAX_AGE: 0 Widths PWR_REP_THRESH: 2 Bad Frame(s) PWR_REP_FRAMES: 113 frame(s) PWR_THRESH_ENABLE: Yes Handoff Thresholds PWR_PERIOD_ENABLE: No PWR_REP_DELAY: 4 frames RESCAN: No T_ADD: -14.0 dB T_DROP: -16.0 dB T_COMP: 4.0 T_TDROP: 4 sec EXT_SYS_PARAMETER: Yes EXT_NGHBR_LIST: Yes GEN_NGHBR_LIST: No GLOBAL_REDIRECT: Yes PRI_NGHBR_LIST: No USER_ZONE_ID: No What other optional EXT_GLOBAL_REDIRECT: No EXT_CHAN_LIST: Yes configuration messages RESERVED: 0 exist?November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 30
  • 31. 1xRTT Extended System Parameters Message EXTENDED SYSTEM PARAMETERS 000021, Time 15:28:37.421, Record 6188, QcpCdmaLogMsgPagingChan PD: P_REV_IN_USE < 6 One main job of this message is to MSG_TYPE: Extended System Parameters Message tell mobiles how to report their PILOT_PN: 36 CONFIG_MSG_SEQ: 1 identities when they transmit on the DELETE_FOR_TMSI: No Access Channel USE_TMSI: No PREF_MSID_TYPE: IMSI and ESN • IMSI - International Mobile MCC: 1134 IMSI_11_12: 813 Subscriber Identity TMSI_ZONE_LEN: 1 octet TMSI_ZONE: 0 – The “world” phone number BCAST_INDEX: Disable Periodic Broadcast Paging of the mobile IMSI_T_SUPPORTED: No P_REV: IS-2000 Revision 0 • ESN - Electronic Serial Number MIN_P_REV: J-STD-008 SOFT_SLOPE: 18 Different Networks may request ADD_INTERCEPT: 6 dB different identification modes; the DROP_INTERCEPT: 6 dB PACKET_ZONE_ID: Base Station Does Not Support A phones simply comply Packet Data Service Zone MAX_NUM_ALT_SO: 0 • IMSI and ESN RESELECT_INCLUDED: No PILOT_REPORT: No • IMSI only NGHBR_SET_ENTRY_INFO: No NGHBR_SET_ACCESS_INFO: No • ESN only BROADCAST_GPS_ASST: No QPCH_SUPPORTED: No Intelligent soft handoff parameters SDB_SUPPORTED: No are also included RLGAIN_TRAFFIC_PILOT: 0.000000 dB REV_PWR_CNTL_DELAY_INCL: No AUTO_MSG_SUPPORTED: No RESERVED: 0November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 31
  • 32. The Neighbor List Message EXTENDED NEIGHBOR LIST The Neighbor List Message gives the 000017, Time 15:28:37.381, Record 6158, mobile up to 20 PN offsets of sectors it QcpCdmaLogMsgPagingChan may soon need in handoff PD: P_REV_IN_USE < 6 MSG_TYPE: Extended Neighbor List Message • This enables the mobile to search PILOT_PN: 36 CONFIG_MSG_SEQ: 1 smarter and faster PILOT_INC: 4 NGHBR_CONFIG: 0, NGHBR_PN: 32 On the paging channel, Enhanced or SEARCH_PRIORITY: Very High, FREQ_INCL: No Extended neighbor lists may also include NGHBR_CONFIG: 0 NGHBR_PN: 28 SEARCH_PRIORITY: Very High FREQ_INCL: No neighbors on different frequencies NGHBR_CONFIG: 0 NGHBR_PN: 308 SEARCH_PRIORITY: Very High FREQ_INCL: No • Slotted mode mobiles can jump to NGHBR_CONFIG: 0 NGHBR_PN: 432 other frequencies in their “sleep” SEARCH_PRIORITY: Very High FREQ_INCL: No NGHBR_CONFIG: 0 NGHBR_PN: 20 time to check pilots SEARCH_PRIORITY: Very High FREQ_INCL: No NGHBR_CONFIG: 0 NGHBR_PN: 24 • This is useful at system boundaries SEARCH_PRIORITY: Very High FREQ_INCL: No NGHBR_CONFIG: 0 NGHBR_PN: 260 During a call, a mobile first uses the SEARCH_PRIORITY: Very High FREQ_INCL: No neighbor list remembered from idle mode NGHBR_CONFIG: 0 NGHBR_PN: 196 SEARCH_PRIORITY: Very High FREQ_INCL: No • After each handoff, a new Neighbor NGHBR_CONFIG: 0 NGHBR_PN: 392 List Update message is sent to the SEARCH_PRIORITY: Very High FREQ_INCL: No NGHBR_CONFIG: 0 NGHBR_PN: 312 mobile on the Forward Traffic SEARCH_PRIORITY: Very High FREQ_INCL: No Channel NGHBR_CONFIG: 0 NGHBR_PN: 316 SEARCH_PRIORITY: Very High FREQ_INCL: No Each neighbor list received by the mobile RESERVED: 0 overwrites and replaces the previous neighbor listNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 32
  • 33. The CDMA Channel List Message EXTENDED CDMA CHANNEL LIST MESSAGE If a mobile sees a CDMA 000005, Time 15:28:37.056, Record 5910, Channel List Message, it notices QcpCdmaLogMsgPagingChan PD: P_REV_IN_USE < 6 the list of channels included in the MSG_TYPE: Extended CDMA Channel List Message message PILOT_PN: 36 CONFIG_MSG_SEQ: 1 • There may be one, two, NUM_FREQ: 1 CDMA_FREQ: 600 three, or more channels listed RC_QPCH_SEL_INCL: No TD_SEL_INCL: No The mobile immediately uses a RESERVED: 0 random selection process called “hashing” to select one of the listed channels • The outcome of hashing depends only on the mobile’s F3 F2 IMSI CDMA Ch HASH using F1 Fnow List Message IMSI • Both the system and the mobile know which carrier the mobile will choose The message also includes an indicator to show if the QPCH is in use, and for what radio configurationsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 33
  • 34. How Hashing Works If a mobile sees a CDMA Channel List Message, it notices the list of channels included in the message • There may be one, two, three, or more channels listed Whenever a phone encounters multiple announced resources, it uses its number (IMSI, International Mobile Subscriber Identity) and a randomized process called “hashing” to determine which resource it should use. This is how mobiles select: • Carrier Frequencies in idle mode • Preferred Paging Channel • Preferred Access Channel • Paging Time Slot in Slotted Mode Optimization personnel may wish to carry a phone for each carrier frequency, or use the multiple NAM capability of some handsets to operate on different numbers so as to prefer different frequenciesNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 34
  • 35. Hashing Examples Try your own phone in the spreadsheet Hashing.xls (in utilities folder)Hashing Examples Time between active slots, seconds:v2. 1-28-2000 1.28 2.56 5.12 10.24 20.48 40.96 81.92 163.84 Number of Slots in Mobiles Cycle: 16 32 64 128 256 512 1024 2048 How Many How Many PagingKey in red-shaded Frequencies? Channels? Slot Cycle Index: values 2 1 0 1 2 3 4 5 6 7 10 Digit IMSI Use Freq. # Use PCH # Slot# Slot# Slot# Slot# Slot# Slot# Slot# Slot# 6153000124 1 1 15 31 63 127 127 383 895 895 6153000125 1 1 11 27 27 27 27 27 539 1563 6153000126 1 1 5 5 5 69 69 69 69 69 6153000127 1 1 3 3 3 67 195 451 451 1475 6153000128 2 1 8 24 24 24 152 152 152 1176 6153000129 2 1 9 25 25 25 25 25 25 25 6153000130 1 1 11 27 27 27 27 27 539 1563 6153000131 2 1 1 1 33 97 225 225 737 737 6153000132 1 1 8 8 40 40 40 40 552 552 6153000133 1 1 3 19 51 115 243 243 755 755 November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 35
  • 36. The Global Service Redirection Message GLOBAL SERVICE REDIRECTION 000011, Time 15:28:37.118, Record 5957, QcpCdmaLogMsgPagingChan PD: P_REV_IN_USE < 6 The GSRM was originally MSG_TYPE: Global Service Redirection Message intended as a way to PILOT_PN: 36 CONFIG_MSG_SEQ: 1 solve system and REDIRECT_ACCOLC (ACCOLC_0): No multicarrier border REDIRECT_ACCOLC (ACCOLC_1): No REDIRECT_ACCOLC (ACCOLC_2): No problems REDIRECT_ACCOLC (ACCOLC_3): No REDIRECT_ACCOLC (ACCOLC_4): No • Outermost F2 cells REDIRECT_ACCOLC (ACCOLC_5): No REDIRECT_ACCOLC (ACCOLC_6): No transmit GSRM, REDIRECT_ACCOLC (ACCOLC_7): No REDIRECT_ACCOLC (ACCOLC_8): No sending distant F2 REDIRECT_ACCOLC (ACCOLC_9): No mobiles to F1 REDIRECT_ACCOLC (ACCOLC_10): No REDIRECT_ACCOLC (ACCOLC_11): No The GSRM can also be REDIRECT_ACCOLC (ACCOLC_12): No REDIRECT_ACCOLC (ACCOLC_13): No used to manually REDIRECT_ACCOLC (ACCOLC_14): No REDIRECT_ACCOLC (ACCOLC_15): No distribute idle mobiles to RETURN_IF_FAIL: No different frequencies DELETE_TMSI: No EXCL_P_REV_MS: No • A GSRM applies only RECORD_TYPE: Redirection to An Analog System RECORD_LEN: 3 octets to phones of Access EXPECTED_SID: 0 IGNORE_CDMA: No Overload Classes SYS_ORDERING: Attempt To Obtain Service On Either System A specified in the Or System B. If Unsuccessful, Attempt Alternate System MAX_REDIRECT_DELAY: 0 sec messageNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 36
  • 37. Summary: How Idle Mobiles Choose CDMA CarriersAt turnon, Idle mobiles use proprietary System Determination Algorithms(SDA) to find the initial CDMA carrier intended for them to useOn the paging channel of the idle mobile’s newly-found home signal, themobile might be sent to a different frequency if it hears • CDMA Channel List Message • Global Service Redirection Message (GSRM) Start System Determination Algorithm Preferred MRU Only Bit 0 PRL Acq Idx Go to last Strongest Is better Yes Idle Mode Carrier Selection Is SID frequency PN, read SID permitted? from MRU Sync available? F3 No Signal Denied SID No CDMA Ch HASH using F2 Config List Message IMSI F1 Messages: Read remain Last Resort: Paging GEO escape Channel Global Svc my ACCOLC? Or Analog Redir Msg redirect to another CDMA frequency or system Legend to Analog Steps from Steps from Proprietary the CDMA proprietary SDA standards SDAs databases November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 37
  • 38. Course RF200 Let’s Do An Idle Mode Let’s Do An Idle Mode Handoff! Handoff!November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 38
  • 39. Idle Mode Handoff An idle mobile always demodulates the best available signal • In idle mode, it isn’t possible to do soft handoff and listen to multiple sectors or base stations at the same time -- the paging channel information stream is different on each sector, not synchronous -- just like ABC, NBC, CBS, and CNN TV news programs aren’t in word-sync for simultaneous viewing • Since a mobile can’t combine signals, the mobile must switch quickly, always enjoying the best available signal The mobile’s pilot searcher is constantly checking neighbor pilots If the searcher notices another signal at least 3 db better than the present one, and it remains so for 5 seconds, the mobile starts listening to it at the beginning of the next paging slot. • The mobile doesn’t automatically say anything to the system, so system doesn’t know about the idle mode handoff On the new paging channel, if the mobile learns that registration is required, it re-registers on the new sectorNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 39
  • 40. Idle Mode on the Paging Channel: Meet the Neighbors, track the Strongest Pilot All PN Offsets 0 Ec/Io -20Chips 0 SRCH_WIN_A Mobile Rake RX 32K PN 0 F1 PN168 W01 512 Active Pilot F2 PN168 W01 Rake Fingers F3 PN168 W01 Srch PN??? W0 SRCH_WIN_N The phone’s pilot searcher constantly checks the pilots listed in the Neighbor List Message Reference PN Neighbor Set If the searcher ever notices a neighbor pilot substantially stronger than the current reference pilot, it becomes the new reference pilot and the phone switches over to its paging channel on the next superframe. This is called an idle mode handoff. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 40
  • 41. Course RF200 Let’s Register! Let’s Register!November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 41
  • 42. Registration Registration is the process by which an idle mobile lets the system know it’s awake and available for incoming calls • this allows the system to inform the mobile’s home switch of the mobile’s current location, so that incoming calls can be delivered • registration also allows the system to intelligently page the mobile only in the area where the mobile is currently located, thereby eliminating useless congestion on the paging channels in other areas of the system There are many different conditions that could trigger an obligation for the mobile to register • there are flags in the System Parameters Message which tell the mobile when it must register on the current systemNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 42
  • 43. Registration Registration Message (by PROBING) BTS Base Station Acknowledgment OrderPaging AccessChannel Channel November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 43
  • 44. An Actual 1xRTT Registration SYSTEM PARAMETERS MESSAGEIS-95 Message Type: System ParametersPN Offset: 44 CONFIG_MSG_SEQ 0 SID 1121 NID 1REG_ZONE: 0 TOTAL_ZONES: 0 Zone timer length (min): 1 The System Parameters Message tellsMULT_SIDS: 0 MULT_NIDS: 0BASE_ID: 5586 BASE_CLASS: Public Macrocellular System all mobiles when they should register.PAG_CHAN: 1 MAX_SLOT_CYCLE_INDEX: 2 This mobile notices that it is obligated toHOME_REG: 1 FOR_SID_REG: 1 FOR_NID_REG: 1,POWER_UP_REG: 1 POWER_DOWN_REG: 1 register, so it transmits a RegistrationPARAMETER_REG: 1 Registration period (sec): 1853.60Base station 0°00´00.00¨ Lon., 0°00´00.00° Lat. REG_DIST: 0 Message.SRCH_WIN_A: 20ch SRCH_WIN_N: 100ch SRCH_WIN_R: 320chNGHBR_MAX_AGE: 0 PWR_REP_THRESH: 2 REGISTRATION MESSAGEPWR_REP_FRAMES (frames): 905 PWR_THRESH_ENABLE: 1 IS-95 Message Type: RegistrationPWR_PERIOD_ENABLE: 0, PWR_REP_DELAY: 1 (0 frames) ACK_SEQ: 7 MSG_SEQ: 5 ACK_REQ: 1 VALID_ACK: 0Re-Init and Re-acquire After This Message?: No ESN (Electronic Serial Number):0xB38092BCT_ADD: -14dB T_DROP: -16dB T_COMP: 1 DB, T_TDROP: 4s IMSI Class: 0 IMSI Class 0 Type: IMSI_S onlySending Extended System Parameters Messages?: Yes IMSI_S: 694 582 9500Are Extended Neighbor List Messages Being Sent?: No Pilot Strength: -8.0 dBAre General Neighbor List Messages Being Sent?: No Active pilot is first one probed?: YesUsing Global Redirect Messages?: No Original pilot is same as pilot in previous probe?: NoAre Private Neighbor List Messages Being Sent?: No Number of additional pilots: 0Are User Zone ID Messages Being Sent?: No Registration Type: Timer-based Slot Cycle Index: 2Are Extended Global Redirection Messages Being Sent?: No Mobile Protocol Revision Level: 6Are Extended Channel List Messages Being Sent?: Yes Station Class Mark: Dual Mode, Slotted, Discontinuous Xmit, Power Class 3 Mobile-Terminated Calls Acceptable?: Yes BASE STATION ACKNOWLEDGMENT IS-95 Message Type: Order ACK_SEQ: 5 MSG_SEQ: 2 ACK_REQ: 0 VALID_ACK: 1 The base station confirms that the Address Type: IMSI IMSI Class: 0 mobile’s registration message was IMSI Class 0 Type: IMSI_S, IMSI_11_12, and MCC Mobile Country Code (MCC): 310 IMSI 11th+12th Digits: 00 received. We’re officially registered! IMSI_S: 694 582 9500 Order Message Type: Base ACK November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 44
  • 45. Example 4 Let’s Receive an Incoming Let’s Receive an Incoming Call! Call!November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 45
  • 46. Receiving an Incoming Call All idle mobiles monitor the paging channel to receive incoming calls. When an incoming call appears, the paging channel notifies the mobile in a General Page Message. A mobile which has been paged sends a Page Response Message on the access channel. The system sets up a traffic channel for the call, then notifies the mobile to use it with a Channel Assignment Message. The mobile and the base station notice each other’s traffic channel signals and confirm their presence by exchanging acknowledgment messages. The base station and the mobile negotiate what type of call this will be -- I.e., 13k voice, etc. The mobile is told to ring and given a “calling line ID” to display. When the human user presses the send button, the audio path is completed and the call proceeds.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 46
  • 47. Incoming Call Delivery Scenario General Page Message Page Response Message (by PROBING) BTS Base Station Acknowledgment OrderPaging Channel Assignment Message AccessChannel Channel Continuous frames of all 000’s Traffic Channel Preamble: Frames of 000’s Base Station Acknowledgment OrderForward Reverse Traffic TrafficChannel Mobile Station Acknowledgment Order Channel Service Connect Message Service Connect Complete Message The Call is now officially Established! November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 47
  • 48. An Actual Page and Page Response GENERAL PAGE MESSAGE98/05/24 23:14:46.127 [PCH] General Page MessageMSG_LENGTH = 128 bits The system pages the mobile,MSG_TYPE = General Page Message 615-330-0644.CONFIG_MSG_SEQ = 1 ACC_MSG_SEQ = 20CLASS_0_DONE = 1CLASS_1_DONE = 1 RESERVED = 0 PAGE RESPONSE MESSAGEBROADCAST_DONE = 1 RESERVED = 0ADD_LENGTH = 0 bits ADD_PFIELD = Field Omitted 98/05/24 23:14:46.425 [ACH] Page Response MessagePAGE_CLASS = 0 PAGE_SUBCLASS = 0 MSG_LENGTH = 216 bitsMSG_SEQ = 1 MSG_TYPE = Page Response MessageIMSI_S = 6153300644 ACK_SEQ = 1 MSG_SEQ = 2 ACK_REQ = 1SPECIAL_SERVICE = 1 VALID_ACK = 1 ACK_TYPE = 2SERVICE_OPTION = 32768 MSID_TYPE = IMSI and ESN MSID_LEN = 9 octetsRESERVED = Field Omitted ESN = 0xD30E415C IMSI_CLASS = 0 IMSI_CLASS_0_TYPE = 0 RESERVED = 0 IMSI_S = 6153300644 AUTH_MODE = 1 The mobile responds to the page. AUTHR = 0x307B5 RANDC = 0xC6 COUNT = 0 MOB_TERM = 1 SLOT_CYCLE_INDEX = 0 MOB_P_REV = 3 SCM = 106BASE STATION ACKNOWLEDGMENT REQUEST_MODE = Either Wide Analog or CDMA Only SERVICE_OPTION = 32768 PM = 098/05/24 23:14:46.768 [PCH] Order Message NAR_AN_CAP = 0 RESERVED = 0MSG_LENGTH = 112 bitsMSG_TYPE = Order MessageACK_SEQ = 2 MSG_SEQ = 0 ACK_REQ = 0VALID_ACK = 1ADDR_TYPE = IMSI ADDR_LEN = 40 bitsIMSI_CLASS = 0 IMSI_CLASS_0_TYPE = 0 RESERVED = 0 The base station confirms that the mobile’sIMSI_S = 6153300644ORDER = Base Station Acknowledgement Order page response was received. Now theADD_RECORD_LEN = 0 bits mobile is waiting for channel assignment,Order-Specific Fields = Field Omitted RESERVED = 0 expecting a response within 12 seconds. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 48
  • 49. Channel Assignment and Traffic Channel Confirmation CHANNEL ASSIGNMENT MESSAGE18:14:47.027 Paging Channel: Channel Assignment Only about 400 ms. after the base stationACK_SEQ: 2 MSG_SEQ: 1 ACK_REQ: 0 VALID_ACK: 1MSID_TYPE: 2 IMSI: (Class: 0, Class_0_type: 0) acknowledgment order, the mobile receives[0x 01 f8 39 6a 15] 615-330-0644ASSIGN_MODE: Traffic Channel Assignment the channel assignment message.ADD_RECORD_LEN: 5 FREQ_INCL: 1 GRANTED_MODE: 2CODE_CHAN: 43 FRAME_OFFSET: 2ENCRYPT_MODE: Encryption disabledBAND_CLASS: 800 MHz cellular bandCDMA_FREQ: 283 The mobile sees at least twoThe base station is already good blank frames in a row on sending blank frames on the forward channel, andthe forward channel,using concludes this is the right trafficthe assigned Walsh code. channel. It sends a preamble of two blank frames of its own on the reverse traffic channel.BASE STATION ACKNOWLEDGMENT MOBILE STATION ACKNOWLEDGMENT18:14:47.581 Forward Traffic Channel: OrderACK_SEQ: 7 MSG_SEQ: 0 ACK_REQ: 1 18:14:47.598 Reverse Traffic Channel: OrderENCRYPTION: 0 USE_TIME: 0 ACTION_TIME: 0 ACK_SEQ: 0 MSG_SEQ: 0 ACK_REQ: 0Base Station Acknowledgement Order ENCRYPTION: 0 Mobile Station Acknowledgement Order The base station acknowledges The mobile station acknowledges the receiving the mobile’s preamble. base station’s acknowledgment. Everybody is ready! November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 49
  • 50. Service Negotiation and Mobile Alert SERVICE CONNECT MESSAGE 18:14:47.760 Forward Traffic Channel: Service Connect Now that both sides have arrived on the ACK_SEQ: 0 MSG_SEQ: 1 ACK_REQ: 0 ENCRYPTION: 0 USE_TIME: 0 ACTION_TIME: 0 SERV_CON_SEQ: 0 traffic channel, the base station Service Configuration: supported Transmission: Forward Traffic Channel Rate (Set 2): 14400, 7200, 3600, 1800 bps proposes that the requested call Reverse Traffic Channel Rate (Set 2): 14400, 7200, 3600, 1800 bps actually begin. Service option: (6) Voice (13k) (0x8000) Forward Traffic Channel: Primary Traffic Reverse Traffic Channel: Primary Traffic SERVICE CONNECT COMPLETE MSG. 18:14:47.835 Reverse Traffic Channel: Service Connect Completion ACK_SEQ: 1 MSG_SEQ: 3 ACK_REQ: 1 ENCRYPTION: 0 SERV_CON_SEQ: 0 ALERT WITH INFORMATION MESSAGE 18:14:47.961 Forward Traffic Channel: The mobile agrees and Alert With Information says its ready to play. ACK_SEQ: 3 MSG_SEQ: 1 ACK_REQ: 1 ENCRYPTION: 0 SIGNAL_TYPE = IS-54B Alerting ALERT_PITCH = Medium Pitch (Standard Alert) SERVICE CONNECT COMPLETE is a SIGNAL = Long RESERVED = 0 major milestone in call processing. Up RECORD_TYPE = Calling Party Number until now, this was an access attempt. RECORD_LEN = 96 bits NUMBER_TYPE = National Number Now it is officially a call. NUMBER_PLAN = ISDN/Telephony Numbering Plan PI = Presentation Allowed SI = Network Provided 18:14:48.018 Reverse Traffic Channel: Order CHARi = 6153000124 RESERVED = 0 RESERVED = 0 ACK_SEQ: 1 MSG_SEQ: 4 ACK_REQ: 0 ENCRYPTION: 0The base station orders the mobile to ring, and Mobile Station Acknowledgement Order gives it the calling party’s number to display. The mobile says it’s ringing. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 50
  • 51. The Human Answers! Connect Order The mobile has been ringing for several seconds. The human user finally comes over and presses the send button to answer the call. CONNECT ORDER 18:14:54.758 Reverse Traffic Channel: Order ACK_SEQ: 6 MSG_SEQ: 0 ACK_REQ: 1 ENCRYPTION: 0 Connect Order BASE STATION ACKNOWLEDGMENT 18:14:54.920 Forward Traffic Channel: Order ACK_SEQ: 0 MSG_SEQ: 1 ACK_REQ: 0 ENCRYPTION: 0 USE_TIME: 0 ACTION_TIME: 0 Base Station Acknowledgement Order Now the switch completes the audio circuit and the two callers can talk!November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 51
  • 52. Course RF200 Let’s Make An Outgoing Call! Let’s Make An Outgoing Call!November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 52
  • 53. Placing an Outgoing Call The mobile user dials the desired digits, and presses SEND. Mobile transmits an Origination Message on the access channel. The system acknowledges receiving the origination by sending a base station acknowledgement on the paging channel. The system arranges the resources for the call and starts transmitting on the traffic channel. The system notifies the mobile in a Channel Assignment Message on the paging channel. The mobile arrives on the traffic channel. The mobile and the base station notice each other’s traffic channel signals and confirm their presence by exchanging acknowledgment messages. The base station and the mobile negotiate what type of call this will be -- I.e., 13k voice, etc. The audio circuit is completed and the mobile caller hears ringing. Supplemental channels can be requested for data bursts as neededNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 53
  • 54. Mobile-Originated Call Scenario Origination Message (by PROBING) BTS Base Station Acknowledgment OrderPaging Channel Assignment Message AccessChannel Channel Continuous frames of all 000’s Traffic Channel Preamble: Frames of 000’s Base Station Acknowledgment OrderForward Reverse Traffic TrafficChannel Mobile Station Acknowledgment Order Channel Service Connect Message Service Connect Complete Message The Call is now officially Established! November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 54
  • 55. 1xRTT Origination ORIGINATION MESSAGE MSG_LENGTH 36 octets PD, 1, P_REV_IN_USE >= 6 The mobile sends an MSG_ID, 4, Origination Message LAC_LENGTH 13 octets origination message ACK_SEQ 7 MSG_SEQ 0 ACK_REQ 1 VALID_ACK 0 ACK_TYP, 0 on the access MSID_TYPE, 3, IMSI and ESN MSID_LEN, 9, 9 octets ESN, 0x9F 5C BB F5, 2673654773 channel. IMSI_CLASS, 0, IMSI_CLASS_0_TYPE, 0, IMSI_S included RESERVED, 0, IMSI_S, 0x03 C8 87 79 AF, 9132209814 AUTH_MODE, 0, 0 LAC_PADDING, 0, ACTIVE_PILOT_STRENGTH, 5, -2.50 dB FIRST_IS_ACTIVE, 1, Yes MOB_TERM, 1, Yes SLOT_CYCLE_INDEX 2 512 MOB_P_REV 6 IS-2000 Rev 0 SCM 234 BandClass 1DualMode Slotted Continuous Class III REQUEST_MODE, 1, CDMA Only SPECIAL_SERVICE YesBASE STATION ACKNOWLEDGMENT SERVICE_OPTION, 33, Std: 144kbps PacketData, IP or ISO PM, 0, No DIGIT_MODE, 0, 4-bit DTMF CodesMSG_LENGTH, 16, 16 octets MORE_FIELDS, 0, No NUM_FIELDS, 4,PD, 0, P_REV_IN_USE < 6MSG_TYPE, 7, Order Message The base station Character, CHARi, 12, #, 7, 7, 7, 7 7, 7 NAR_AN_CAP, 0, NoACK_SEQ, 0, confirms that the PACA_REORIG, 0, User Directed OriginationMSG_SEQ, 0, origination RETURN_CAUSE, 0, Normal AccessACK_REQ, 0, No MORE_RECORDS, 0, PACA_SUPPORTED, 0, NoVALID_ACK, 1, YesADDR_TYPE, 2, IMSI message was NUM_ALT_SO, 0, DRS, 1, Yes UZID_INCL, 0, No CH_IND, 1, Fundamental Channel SR_ID, 1,ADDR_LEN, 7, 7 octets received. OTD_SUPPORTED, 0, No QPCH_SUPPORTED, 1, YesIMSI_CLASS, 0, ENHANCED_RC, 1, Yes FOR_RC_PREF, 3,IMSI_CLASS_0_TYPE, 3, IMSI_S, IMSI_11_12, and MCC REV_RC_PREF, 3, FCH_SUPPORTED, 1, Yesincluded FCH_FRAME_SIZE, 0, Supports only 20 ms Frame SizesRESERVED, 0, FOR_FCH_LEN, 2, F-RC1, 1, Yes F-RC2, 1, Yes F-RC3, 1,MCC, 209, 310 Yes F-RC4, 1, Yes F-RC5, 1, Yes F-RC6, 0, NoIMSI_11_12, 99, 00 REV_FCH_LEN, 2, R-RC1, 1, Yes R-RC2, 1, Yes R-RC3,IMSI_S, 0x03 C8 87 79 AF, 9132209814 1, Yes R-RC4, 1, Yes R-RC5, 0, NoORDER, 16, Base Station Acknowledgement Order DCCH_SUPPORTED, 0, No ENC_INFO_INCL, 0, NoADD_RECORD_LEN, 0, 0 octets SYNC_ID_INCL, 1, Yes SYNC_ID, error, error: 16 bit field, 6RESERVED, 0, bits available RESERVED, 0, November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 55
  • 56. Traffic Channel Assignment EXTENDED CHANNEL ASSIGNMENT MESSAGEMSG_LENGTH, 30, 30 octetsPD, 0, P_REV_IN_USE < 6MSG_TYPE, 21, Extended Channel Assignment MessageACK_SEQ, 0, MSG_SEQ, 1, ACK_REQ, 0, NoVALID_ACK, 1, Yes ADDR_TYPE, 2, IMSIADDR_LEN, 7, 7 octets IMSI_CLASS, 0,CLASS_0_TYPE, 3, IMSI_S, IMSI_11_12, and MCC includedRESERVED, 0, MCC, 209, 310 IMSI, IMSI_11_12, 99, 00IMSI_S, 0x03 C8 87 79 AF, 9132209814RESERVED_1, 0, ADD_RECORD_LEN, 14, 13 octetsASSIGN_MODE, 0, Traffic Channel AssignmentRESERVED_2, 0, FREQ_INCL, 1, YesDEFAULT_CONFIG, 4, ReservedBYPASS_ALERT_ANSWER, 0, NoRESERVED, 0, NUM_PILOTS, 0, 1 PilotsGRANTED_MODE, 0, FRAME_OFFSET, 15, 18.75 msENCRYPT_MODE, 0, Encryption DisabledBAND_CLASS, 1, 1.850 to 1.990 GHz BandCDMA_FREQ, 274, PILOT_PN, 240,PWR_COMB_IND, 0, No CODE_CHAN, 17,FOR_FCH_RC, 3, RC 3 REV_FCH_RC, 3, RC 3FPC_FCH_INIT_SETPT, 56, 7.000 dB The base station sends aFPC_FCH_FER, 0, 0.2%FPC_FCH_MIN_SETPT, 2, 0.250 dB Channel AssignmentFPC_FCH_MIN_SETPT, 4, 0.500 dBFPC_SUBCHAN_GAIN, 7, 0 dB Message and the mobileRLGAIN_ADJ, 8, 8 dB goes to the traffic channel.REV_FCH_GATING_MODE, 0, No RESERVED, 0, November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 56
  • 57. Traffic Channel Confirmation The mobile sees at least twoThe base station is already good blank frames in a row on sending blank frames on the forward channel, andthe forward channel,using concludes this is the right trafficthe assigned Walsh code. channel. It sends a preamble of two blank frames of its own on the reverse traffic channel.BASE STATION ACKNOWLEDGMENT MOBILE STATION ACKNOWLEDGMENTMSG_LENGTH, 8, 8 octets MSG_TYPE, 1, Order MessageACK_SEQ, 7, MSG_SEQ, 0, ACK_REQ, 1, Yes MSG_LENGTH, 8, 8 octets MSG_TYPE, 1, Order MessageENCRYPTION, 0, Encryption Disabled ACK_SEQ, 0, MSG_SEQ, 1, ACK_REQ, 1, YesUSE_TIME, 0, No ACTION_TIME, 0, 0 ms ENCRYPTION, 0, Encryption DisabledORDER, 16, Base Station Acknowledgement Order USE_TIME, 0, No ACTION_TIME, 0, 0 msADD_RECORD_LEN, 0, 0 octets RESERVED, 0, ORDER, 16, Mobile Station Acknowledgement Order ADD_RECORD_LEN, 0, 0 octets RESERVED, 0, The base station acknowledges The mobile station acknowledges the receiving the mobile’s preamble. base station’s acknowledgment. Everybody is ready! November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 57
  • 58. Status Request/Response STATUS REQUEST STATUS RESPONSE MESSAGEMSG_LENGTH, 9 octets MSG_TYPE, 16, Status Request Message MSG_LENGTH 44 MSG_TYPE, 16, Status Response MessageACK_SEQ, 7, MSG_SEQ, 1, ACK_REQ, 1, Yes ACK_SEQ, 1, MSG_SEQ, 0, ACK_REQ, 0 ENCRYPTION 0ENCRYPTION, 0, Encryption Disabled QUAL_INFO_TYPE, 0, None QUAL_INFO_LEN 0 octetsQUAL_INFO_TYPE, 0, None QUAL_INFO_LEN, 0, 0 octets RECORD_TYPE 27, Reserved RECORD_LEN 9 octetsNUM_FIELDS, 2, RECORD_TYPE, 27, Reserved OTD_SUPPORTED, 0, No FCH_SUPPORTED, 1, YesRECORD_TYPE, 28, Channel Configuration Capability Information FCH_FRAME_SIZE, 0, FOR_FCH_LEN 2 RC1:1 RC2:1 RC3:1 RC4:1 RC5:1 RC6: 0 REV_FCH_LEN 2 RC1:1RC2:1 RC3:1 RC4:1 RC5:0 RC6: 0 DCCH_SUPPORTED 0 No FOR_SCH_SUPPORTED 1 Yes FOR_SCH_LEN 2 RC1:0 RC2:0 RC3:1 RC4:1 RC5:0 RC6:0 FOR_SCH_NUM, 1, FOR_TURBO_SUPPORTED 0 FOR_CONV_SUPPORTED 1 FOR_MAX_CONV_BLOCK_SIZE 4 RS1: 3048, RS 2: 4584 FOR_FRAME_40_SUPPORTED 0 FOR_FRAME_80_SUPPORTED 0 FOR_MAX_RATE, 0, 9.6 kbps or 14.4 kbps REV_SCH_SUPPORTED, 1, Yes REV_SCH_LEN 1 RC1:0 RC2:0 RC3:1 REV_SCH_NUM 1 REV_TURBO_SUPPORTED 0 REV_CONV_SUPPORTED 1 REV_MAX_CONV_BLOCK_SIZE 4 RS1: 3048, RS2: 4584 REV_FRAME_40_SUPPORTED 0 REV_FRAME_80_SUPPORTED 0 REV_MAX_RATE, 0, 9.6 kbps or 14.4 kbps NONOCTET_ALIGNED_DATA 0 OCTET_ALIGNED_DATA 0 STS_SUPPORTED, 0, No 3X_CCH_SUPPORTED, 0, No RECORD_TYPE 14 Band Class Information RECORD_LEN 12 BAND_CLASS_0:0 BAND_CLASS_1: 0 BAND_CLASS_2: 0 BAND_CLASS_3: 1 BAND_CLASS_4: 0 BAND_CLASS_5: 0 BAND_CLASS_6: 0 BAND_CLASS_7: 0 RECORD_TYPE, 0, Reserved RECORD_LEN, 15, 15 octets BASE STATION ACKNOWLEDGMENT RESERVED128 RESERVED 0 RESERVED 23 RESERVED 129 MSG_LENGTH, 7, 7 octets MSG_TYPE, 1, Order Message RESERVED, 0, RESERVED, 0, RESERVED, 248, RESERVED, 0, ACK_SEQ 3, MSG_SEQ 4, ACK_REQ 0, ENCRYPTION 0 RESERVED, 1, RESERVED, 120, RESERVED, 0, RESERVED, 16, ORDER, 16, Base Station Acknowledgement Order RESERVED, 36, RESERVED, 132, RESERVED, 16, RESERVED, 66, ADD_RECORD_LEN, 0 CON_REF_INCL, 0, No RESERVED, 0, RESERVED, 64, RESERVED, 145, RESERVED, 16, RESERVED, 64, RESERVED, 136, RESERVED, 0, November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 58
  • 59. Status Response Message STATUS RESPONSE MESSAGE MSG_LENGTH 66 MSG_TYPE, 16, Status Response Message ACK_SEQ 2, MSG_SEQ 3, ACK_REQ 0, ENCRYPTION 0 QUAL_INFO_TYPE 2, BAND_CLASS+OP_MODE QUAL_INFO_LEN 2 BAND_CLASS, 0, 800 MHz Cellular Band OP_MODE, 1, TIA/EIA/IS-95 CDMA Mode In Band Class 0 RESERVED, 0, RECORD_TYPE, 8, Terminal Info RECORD_LEN 55 MOB_P_REV, 6, IS-2000 Revision 0 MOB_MFG_CODE, 159, MOB_MODEL, 69, MOB_FIRM_REV, 783, SCM, 106, Band Class 0, Dual Mode, Slotted, Continuous, Class III LOCAL_CTRL, 0, No SLOT_CYCLE_INDEX, 2, 5.12 SERVICE_OPTION, 32770, QUALCOMM: 8K Markov Old SERVICE_OPTION, 32796, QUALCOMM: 13K Markov Old SERVICE_OPTION, 32798, QUALCOMM: 8K Markov New SERVICE_OPTION, 32799, QUALCOMM: 13K Markov New SERVICE_OPTION, 2, Standard: 8K Loopback SERVICE_OPTION, 9, Standard: 13K Loopback SERVICE_OPTION, 6, Standard: Short Message Services (Rate Set1) SERVICE_OPTION, 14, Standard: Short Message Services (Rate Set2) SERVICE_OPTION, 18, Standard: OverTheAir Param Admin (Rate Set1) SERVICE_OPTION, 19, Standard: OverTheAir Param Admin (Rate Set2) SERVICE_OPTION, 32768, QUALCOMM: Voice 13K SERVICE_OPTION, 17, Standard: High Rate Voice (13 kbps) SERVICE_OPTION, 3, Standard: EVRC (8 kbps) SERVICE_OPTION, 32776, AT&T: Unknown SERVICE_OPTION, 32, Standard: Test Data Service Option (TDSO) SERVICE_OPTION, 33, Standard: 144kbps PacketData, Internet or ISO Protocol SERVICE_OPTION, 25, Std: HighSpeedPacketData:Inet or ISO (RS2 Fwd, RS2 Rev) SERVICE_OPTION, 22, Std: HighSpeedPacketData:Inet or ISO (RS1 Fwd, RS1 Rev) SERVICE_OPTION, 15, Standard: 14.4kbps PDS Internet or ISO Protocol Stack SERVICE_OPTION, 7, Standard: PDS Internet or ISO Protocol Stack SERVICE_OPTION, 4, Standard: Asynchronous Data Service SERVICE_OPTION, 12, Standard: Data SERVICE_OPTION, 5, Standard: Group 3 Facsimile (9.6kbps) SERVICE_OPTION, 13, Standard: Group 3 Facsimile (14.4 or 9.6kbps) RESERVED, 0, RESERVED, 0,November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 59
  • 60. Service Request SERVICE REQUEST MSG. MSG_LENGTH 38 MSG_TYPE, 12, Service Request Message ACK_SEQ 0 MSG_SEQ 0 ACK_REQ 1 ENCRYPTION 0 SERV_REQ_SEQ 0 REQ_PURPOSE 2 Propose A Service Configuration RECORD_TYPE 19 Service Configuration Information RECORD_LEN 30 FOR_MUX_OPTION 1, REV_MUX_OPTION, 1, FOR_NUM_BITS, 240, REV_NUM_BITS, 240, NUM_CON_REC, 1, RECORD_LEN, 12, 12 octets CON_REF, 0, SERVICE_OPTION, 33, Std: 144kbps PacketData, Internet or ISO Protocol FOR_TRAFFIC, 1, SO Uses Primary Traffic On FTC REV_TRAFFIC, 1, SO Uses Primary Traffic On RTC UI_ENCRYPT_MODE, 0, User Information Encryption Disabled SR_ID, 1, RLP_INFO_INCL, 1, Yes RLP_BLOB_LEN, 5, 5 octets RLP_BLOB, 46, RLP_BLOB, 219, RLP_BLOB, 101, RLP_BLOB, 50, RLP_BLOB, 152, QOS_PARMS_INCL, 0, No RESERVED, 0, FCH_CC_INCL, 1, Yes FCH_FRAME_SIZE, 0, Supports only 20 ms FOR_FCH_RC, 3, RC 3 REV_FCH_RC, 3, RC 3 DCCH_CC_INCL 0 FOR_SCH_CC_INCL 1 NUM_FOR_SCH, 1 FOR_SCH_ID, 0, FOR_SCH_MUX, 2337, SCH_REC_LEN, 2, 2 octets SCH_RC 3 SprdRate=1; 1200,1350,1500,2400,2700,4800,9600,19200, 38400,76800,153600 bps data R=1/4, QPSK pre-sprdng symb TD allowed Coding Type, CODING, 0, Convolutional Coding FRAME_40_USED, 0, No FRAME_80_USED, 0, No MAX_RATE, 0, 9.6 kbps or 14.4 kbps REV_SCH_CC_INCL, 1, Yes NUM_REV_SCH, 1, REV_SCH_ID, 0, REV_SCH_MUX, 2321, SCH_REC_LEN, 2, 2 octets SCH_RC, 3, SprdRate=1; 1200,1350,1500,2400,2700,4800,9600,19200, 38400,76800,153600 bps data rates R=1/4, 307200 bps data rate R=1/2; BPSK modulation with a pilot Coding Type, CODING, 0, Convolutional Coding FRAME_40_USED, 0, No FRAME_80_USED, 0, No MAX_RATE, 0, 9.6 kbps or 14.4 kbps RESERVED, 0,November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 60
  • 61. Service Connect/Service Connect Complete SERVICE CONNECT MESSAGEMSG_LENGTH 42 MSG_TYPE, 20, Service Connect MessageACK_SEQ 0 MSG_SEQ 4 ACK_REQ 1 ENCRYPTION 0USE_TIME 0 ACTION_TIME 0ms SERV_CON_SEQ 0RESERVED, 0, USE_OLD_SERV_CONFIG, 0,RECORD_TYPE, 7, Service Configuration RECORD_LEN 30FOR_MUX_OPTION, 1, REV_MUX_OPTION, 1,RS1_9600_FOR 1 RS1_4800_FOR 1 RS1_2400_FOR 1 RS1_1200_FOR 1 RSV 0 0RS1_9600_REV 1 RS1_4800_REV 1 RS1_2400_REV 1 RS1_1200_REV 1RESERVED 0 NUM_CON_REC 1 RECORD_LEN 12 CON_REF, 1,SERVICE_OPTION 33 Std:144kbps PacketData Internet or ISO ProtocolFOR_TRAFFIC, 1, SO Uses Primary Traffic On FTCREV_TRAFFIC, 1, SO Uses Primary Traffic On RTCUI_ENCRYPT_MODE, 0, User Information Encryption DisabledSR_ID, 1, RLP_INFO_INCL, 1, Yes RLP_BLOB_LEN, 5, 5 octets SERVICE CONNECT COMPLETERLP_BLOB, 0x2C 0D 94 CA 60, QOS_PARMS_INCL, 0, No MSG_LENGTH 6 MSG_TYPE 14 Service Connect Completion MessageRESERVED, 0, FCH_CC_INCL, 1, Yes FCH_FRAME_SIZE, 0, No ACK_SEQ 4, MSG_SEQ 1, ACK_REQ 1, ENCRYPTION 0FOR_FCH_RC 3 RC 3 REV_FCH_RC 3 RC 3 DCCH_CC_INCL, 0,No RESERVED, 0, SERV_CON_SEQ, 0, RESERVED, 0,FOR_SCH_CC_INCL, 1, Yes NUM_FOR_SCH, 1, FOR_SCH_ID, 0,FOR_SCH_MUX, 2321, SCH_REC_LEN, 2, 2 octetsSCH_RC 3 SprdRate=1; 1200,1350,1500,2400,2700,4800, 9600, 19200,38400,76800,153600 bps data; R=1/4 QPSK pre-sprdg symbols, TD allowedCODING, 0, Convolutional CodingFRAME_40_USED, 0, No FRAME_80_USED, 0, NoMAX_RATE, 0, 9.6 kbps or 14.4 kbpsREV_SCH_CC_INCL, 1, Yes NUM_REV_SCH, 1,REV_SCH_ID, 0, REV_SCH_MUX, 2321, SCH_REC_LEN, 2, 2 octetsSCH_RC 3 SprdRate=1; 1200,1350,1500,2400,2700,4800,9600,19200,38400,76800,153600 bps R=1/4, 307200 bps R=1/2; BPSK modulation with a pilotCoding Type, CODING, 0, Convolutional CodingFRAME_40_USED, 0, No FRAME_80_USED, 0, NoMAX_RATE, 0, 9.6 kbps or 14.4 kbps RESERVED, 0,RECORD_TYPE, 19, Non-Negotiable Service ConfigurationRECORD_LEN1 FPC_INCL 0 GATING_RATE_INCL 0FOR_SCH_INCL, 0, No USE_FLEX_NUM_BITS, 0, NoUSE_VAR_RATE, 0, No LTU_INFO_INC, 0, No RESERVED, 0,CC_INFO_INCL, 0, No November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 61
  • 62. Course RF200 Lets Upload Data! Lets Upload Data!November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 62
  • 63. Reverse Supplemental Channel Assignment Mobile: Send Mobile: Send Walsh Code 1 Walsh Code 1 Starting in 320 ms Starting in 320 ms For 1000 ms. For 1000 ms. W23 F-FCH ESCAM ESCAM W1 PAGING KGKSAKKNKGGKSKPGSASPPCKGKSAKGKSAKKNKGGKSKPGSASPPCKGKSAKGKSAKKNKGGKSKPGSASPPCK N N NPN 168 W32 SYNC SSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSS BTS W0 PILOT TIME ACCESS CHANNEL R-FCH SCRM SCRM Supplemental Supplemental R-SCH Channel Burst Channel Burst System: I need to System: I need to Send you the Send you the Following blocks: Following blocks: November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 63
  • 64. Reverse Link Supplemental Channel Burst SUPPLEMENTAL CHANNEL EXTENDED SUPPLEMENTAL REQUEST MESSAGE CHANNEL ASSIGNMENT 000280, Time 17:26:11.063, Record 2783, QcpCdmaLogMsgRevTrafChan MESSAGE MSG_TYPE: Supplemental Channel Request Message000281, Time 17:26:11.217, Record 2810, QcpCdmaLogMsgForTrafChan ACK_SEQ: 1MSG_TYPE: Extended Supplemental Channel Assignment Message MSG_SEQ: 6ACK_SEQ: 5 MSG_SEQ: 2 ACK_REQ: No ACK_REQ: NoENCRYPTION: Encryption Disabled ENCRYPTION: Encryption DisabledSTART_TIME_UNIT: 0 ms SIZE_OF_REQ_BLOB: 3 bytesREV_SCH_DTX_DURATION: 200 ms REQ_BLOB: 228USE_T_ADD_ABORT: No USE_SCRM_SEQ_NUM: No REQ_BLOB: 39ADD_INFO_INCL: Yes REQ_BLOB: 255FPC_PRI_CHAN: Forward Fundamental Channel inner loop estimation USE_SCRM_SEQ_NUM: NoREV_CFG_INCLUDED: Yes REF_PN: 236NUM_REV_CFG_RECS: 0 REV_SCH_ID: 0 PILOT_STRENGTH: -4.50 dBREV_WALSH_ID: Forward Dedicated Control Channel inner loop NUM_ACT_PN: 0estimation NUM_NGHBR_PN: 0REV_SCH_NUM_BITS_IDX: RC 1,3,5=1512; RC 2,4,6=2280; 12 CRC bits RESERVED: 0NUM_REV_SCH: 1 REV_SCH_ID: 0REV_SCH_DURATION: ReservedREV_SCH_START_TIME_INCL: YesREV_SCH_START_TIME: 21REV_SCH_NUM_BITS_IDX: RC 1,3,5=1512; RC 2,4,6=2280; 12 CRC bits 000284, Time 17:26:11.691, Record 2893,FOR_CFG_INCLUDED: No QcpCdmaLogMsgRevTrafChanNUM_FOR_SCH: 0 MSG_TYPE: Supplemental Channel Request MessageFPC_INCL: No RPC_INCL: Yes RPC_NUM_SUP: 0 SCH_ID: 0 ACK_SEQ: 1RLGAIN_SCH_PILOT: 1.000000 dB MSG_SEQ: 63X_SCH_INFO_INCL: No ACK_REQ: YesCCSH_INCLUDED: No ENCRYPTION: Encryption DisabledFOR_SCH_CC_INCL: error: 1 bit field, 0 bits available SIZE_OF_REQ_BLOB: 0 bytesREV_SCH_CC_INCL: error: no bits available USE_SCRM_SEQ_NUM: No RESERVED: 0 November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 64
  • 65. Course RF200 Lets Download Data! Lets Download Data!November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 65
  • 66. Forward Supplemental Channel Assignment Mobile: Watch Mobile: Watch Walsh Code 2 Walsh Code 2 Starting in 320 ms Starting in 320 ms For 1000 ms. For 1000 ms. Supplemental Supplemental W2 F-SCH Channel Burst Channel Burst W23 F-FCH ESCAM ESCAM W1 PAGING KGKSAKKNKGGKSKPGSASPPCKGKSAKGKSAKKNKGGKSKPGSASPPCKGKSAKGKSAKKNKGGKSKPGSASPPCK N N NPN 168 W32 SYNC SSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSS BTS W0 PILOT TIME ACCESS CHANNEL R-FCH November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 66
  • 67. Assigning a Forward Supplemental Channel Burst EXTENDED SUPPLEMENTAL CHANNEL ASSIGNMENT MESSAGE 005878, Time 17:34:48.913, Record 105364, QcpCdmaLogMsgForTrafChan MSG_TYPE: Extended Supplemental Channel Assignment Message ACK_SEQ: 0 MSG_SEQ: 1 ACK_REQ: No ENCRYPTION: Encryption Disabled START_TIME_UNIT: 0 ms REV_SCH_DTX_DURATION: 200 ms USE_T_ADD_ABORT: No USE_SCRM_SEQ_NUM: No ADD_INFO_INCL: Yes FPC_PRI_CHAN: Forward Fundamental Channel inner loop estimation REV_CFG_INCLUDED: No NUM_REV_SCH: 0 FOR_CFG_INCLUDED: Yes FOR_SCH_FER_REP: Yes NUM_FOR_CFG_RECS: 0 FOR_SCH_ID: 0 SCCL_INDEX: 0 FOR_SCH_NUM_BITS_IDX: RC 1,3,4,6,7=3048; RC 2,5,8,9=4584; 12 CRC bits NUM_SUP_SHO: 0 PILOT_PN: 112 ADD_PILOT_REC_INCL: No CODE_CHAN_SCH: 3 QOF_MASK_ID_SCH: 0 (rot 256 bit) NUM_FOR_SCH: 1 FOR_SCH_ID: 0 FOR_SCH_DURATION: 320 ms FOR_SCH_START_TIME_INCL: Yes FOR_SCH_START_TIME: 17 SCCL_INDEX: 0 FPC_INCL: Yes FPC_MODE_SCH: 1 FPC_SCH_INIT_SETPT_OP: FPC_SCH_INIT_SETPT has Offset value of initial F- SCH Eb/Nt setpoint FPC_SEC_CHAN: 0 NUM_SUP: 1 SCH_ID: 0 FPC_SCH_FER: 0.5% - 10% (in units of 0.5%) FPC_SCH_INIT_SETPT: 5.000 dB FPC_SCH_MIN_SETPT: 2.000 dB FPC_SCH_MAX_SETPT: 8.000 dB FPC_THRESH_SCH_INCL: No RPC_INCL: No 3X_SCH_INFO_INCL: No CCSH_INCLUDED: No FOR_SCH_CC_INCL: No REV_SCH_CC_INCL: No RESERVED: 0 November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 67
  • 68. Course RF200 Let’s End A Call! Let’s End A Call!November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 68
  • 69. A Beautiful End to a Normal Call MOBILE RELEASE ORDER 008091, Time 17:39:26.108, Record 167760, QcpCdmaLogMsgRevTrafChan MSG_TYPE: Order Message ACK_SEQ: 4 BASE STATION ACKNOWLEDGMENT MSG_SEQ: 1008090, Time 17:39:26.020, Record 167747, ACK_REQ: NoQcpCdmaLogMsgForTrafChan ENCRYPTION: Encryption DisabledMSG_TYPE: Order Message ORDER: Release Order (Normal Release)ACK_SEQ: 5 ADD_RECORD_LEN: 0 octetsMSG_SEQ: 2 RESERVED: 0ACK_REQ: NoENCRYPTION: Encryption DisabledUSE_TIME: NoACTION_TIME: 0 msORDER: Release Order (No Reason Given)ADD_RECORD_LEN: 0 octetsRESERVED: 0 SYNC CHANNEL MESSAGE008092, Time 17:39:26.514, Record 167820, The mobile left the traffic channel,QcpCdmaLogMsgSyncChanMSG_TYPE: Sync Channel Message scanned to find the best pilot, and readP_REV: IS-95B the Sync Channel Message.MIN_P_REV: IS-95ASID: 179NID: 0PILOT_PN: 468LC_STATE: 0x02 87 7C F3 7F BASYS_TIME: 06/29/2002 06:15:19LP_SEC: 13LTM_OFF: -660 minutes November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 69
  • 70. An Aeronautical Analogy: Investigation Tools Control & Parameters Messaging 114.50 118.25 11500 11500 130.75 Aeronautical Investigations Flight Data Recorder Cockpit Voice Recorder CDMA Investigations BTS Temporal Analyzer Data Layer 3 Message FilesTo study the cause of an aeronautical accident, we try to recover the Flight Data Recorder and the Cockpit Voice Recorder.To study the cause of a CDMA call processing accident, we review data from the Temporal Analyzer and the Layer 3 Message Files -- for the same reasons. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 70
  • 71. Example 8 Let’s Do A Handoff! Let’s Do A Handoff!November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 71
  • 72. Basic Rules of Soft HandoffThe Handset considers pilots in sets PILOT SETS • Active: pilots of sectors actually in use Active 6 Req’d. By Std. Min. Members • Candidates: pilots mobile requested, but not yet set up & transmitting by system Candidate 5 • Neighbors: pilots told to mobile by system, as nearby sectors to check Neighbor 20 • Remaining: any pilots used by system but not already in the other sets (div. by PILOT_INC) RemainingHandset sends Pilot Strength MeasurementMessage to the system whenever: HANDOFF • It notices a pilot in neighbor or remaining set PARAMETERS exceeds T_ADD T_ADD T_DROP • An active set pilot drops below T_DROP for T_TDROP time T_TDROP T_COMP • A candidate pilot exceeds an active by T_COMP Exercise: How does a pilotThe System may set up all requested handoffs, in one set migrate intoor it may apply special manufacturer-specific another set, for all cases?screening criteria and only authorize some Identify the trigger, and the messages involved.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 72
  • 73. The Call is Already Established. What Next? All PN Offsets 0 Ec/Io -20 Chips 0 10752 14080 32002 32K PN 0 168 220 500 512Mobile Rake RX Active PilotF1 PN168 W61 Rake Fingers The call is already in progress.F2 PN168 W61 PN 168 is the only active signal,F3 PN168 W61 and also is our timing reference.Srch PN??? W0 Continue checking the neighbors. Reference PN Neighbor Set T_ADD ! ! If we ever notice a neighbor with Ec/Io above T_ADD, ask to use it! Send a Pilot Strength Measurement Message! November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 73
  • 74. The Handoff Process The handset pilot searcher notices energy from another sector or BTS, meeting any of these criteria: •Neighbor or Remaining Pilot Ec/Io stronger than T_Add •Candidate Pilot just got T_Comp better than an ac tive •An Active Pilot stayed below T_DROP for T_TDROP time BTS Pilot Strength Measurement Message Base Station Acknowledgment Order •Selector arranges channel elements/Walsh codes in requestedForward sectors and begins using them, too. Reverse Traffic Traffic Extended Handoff Direction MessageChannel Channel Mobile Station Acknowledgment Order •Handset verifies which assigned PNs it can now hear. Handoff Completion Message Base Station Acknowledgment Order Neighbor List Update Message Mobile Station Acknowledgment Order The new Handoff condition is now officially Established! November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 74
  • 75. Mobile Requests the Handoff! PILOT STRENGTH MEASUREMENT MESSAGE 98/05/24 23:14:02.205 [RTC] Pilot Strength Measurement Message MSG_LENGTH = 128 bitsJust prior to this message, this particular MSG_TYPE = Pilot Strength Measurement Message ACK_SEQ = 5 MSG_SEQ = 0 ACK_REQ = 1mobile already was in handoff with PN 168 ENCRYPTION = Encryption Mode Disabledand 220. REF_PN = 168 Offset Index (the Reference PN) PILOT_STRENGTH = -6.0 dBThis pilot strength measurement message KEEP = 1 PILOT_PN_PHASE = 14080 chips (PN220+0chips)reports PN 500 has increased above PILOT_STRENGTH = -12.5 dBT_Add, and the mobile wants to use it too. KEEP = 1 PILOT_PN_PHASE = 32002 chips (PN500 + 2 chips) PILOT_STRENGTH = -11.0 dB KEEP = 1 RESERVED = 0 BASE STATION ACKNOWLEDGMENT98/05/24 23:14:02.386 [FTC] Order MessageMSG_LENGTH = 64 bitsMSG_TYPE = Order Message The base station acknowledges receivingACK_SEQ = 0 MSG_SEQ = 0 ACK_REQ = 0 the Pilot Strength Measurement Message.ENCRYPTION = Encryption Mode DisabledUSE_TIME = 0 ACTION_TIME = 0ORDER = Base Station Acknowledgment OrderADD_RECORD_LEN = 0 bitsOrder-Specific Fields = Field OmittedRESERVED = 0 November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 75
  • 76. System Authorizes the Handoff! HANDOFF DIRECTION MESSAGE98/05/24 23:14:02.926 [FTC] Extended Handoff Direction Message The base station sends a HandofMSG_LENGTH = 136 bitsMSG_TYPE = Extended Handoff Direction Message Direction Message authorizing theACK_SEQ = 0 MSG_SEQ = 6 ACK_REQ = 1 mobile to begin soft handoff with allENCRYPTION = Encryption Mode DisabledUSE_TIME = 0 ACTION_TIME = 0 HDM_SEQ = 0 three requested PNs. The pre-existingSEARCH_INCLUDED = 1SRCH_WIN_A = 40 PN chips link on PN 168 will continue to useT_ADD = -13.0 dB T_DROP = -15.0 dB T_COMP = 2.5 dB Walsh code 61, the new link on PN220T_TDROP = 4 secHARD_INCLUDED = 0 FRAME_OFFSET = Field Omitted will use Walsh Code 20, and the newPRIVATE_LCM = Field Omitted RESET_L2 = Field Omitted link on PN500 will use Walsh code 50.RESET_FPC = Field Omitted RESERVED = Field OmittedENCRYPT_MODE = Field Omitted RESERVED = Field OmittedNOM_PWR = Field Omitted NUM_PREAMBLE = Field OmittedBAND_CLASS = Field Omitted CDMA_FREQ = Field OmittedADD_LENGTH = 0PILOT_PN = 168 PWR_COMB_IND = 0 CODE_CHAN = 61PILOT_PN = 220 PWR_COMB_IND = 1 CODE_CHAN = 20PILOT_PN = 500 PWR_COMB_IND = 0 CODE_CHAN = 50RESERVED = 0 MOBILE STATION ACKNOWLEDGMENT 98/05/24 23:14:02.945 [RTC] Order Message MSG_LENGTH = 56 bits MSG_TYPE = Order Message ACK_SEQ = 6 MSG_SEQ = 6 ACK_REQ = 0 The mobile acknowledges it has received ENCRYPTION = Encryption Mode Disabled ORDER = Mobile Station Acknowledgment Order the Handoff Direction Message. ADD_RECORD_LEN = 0 bits Order-Specific Fields = Field Omitted RESERVED = 0 November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 76
  • 77. Mobile Implements the Handoff! HANDOFF COMPLETION MESSAGE 98/05/24 23:14:02.985 [RTC] Handoff Completion Message The mobile searcher quickly re-checks MSG_LENGTH = 72 bits MSG_TYPE = Handoff Completion Message all three PNs. It still hears their pilots! ACK_SEQ = 6 MSG_SEQ = 1 ACK_REQ = 1 ENCRYPTION = Encryption Mode Disabled The mobile sends a Handoff Completion LAST_HDM_SEQ = 0 Message, confirming it still wants to go PILOT_PN = 168 Offset Index PILOT_PN = 220 Offset Index ahead with the handoff. PILOT_PN = 500 Offset Index RESERVED = 0 BASE STATION ACKNOWLEDGMENT The base station confirms it has98/05/24 23:14:03.085 [FTC] Forward Traffic Channel: OrderACK_SEQ: 1 MSG_SEQ: 1 ACK_REQ: 0 ENCRYPTION: 0 received the mobile’s HandoffUSE_TIME: 0 ACTION_TIME: 0 Completion message, and willBase Station Acknowledgement Order continue with all of the links active. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 77
  • 78. Neighbor List Updated, Handoff is Complete! NEIGHBOR LIST UPDATE MESSAGE98/05/24 23:14:03.166 [FTC] Neighbor List Update MessageMSG_LENGTH = 192 bitsMSG_TYPE = Neighbor List Update MessageACK_SEQ = 1 MSG_SEQ = 7 ACK_REQ = 1ENCRYPTION = Encryption Mode Disabled In response to the mobile’s HandoffPILOT_INC = 4 Offset Index Completion Message, the base stationNGHBR_PN = 164 Offset IndexNGHBR_PN = 68 Offset Index assembles a new composite neighborNGHBR_PN = 52 Offset Index list including all the neighbors of each ofNGHBR_PN = 176 Offset IndexNGHBR_PN = 304 Offset Index the three active pilots.NGHBR_PN = 136 Offset IndexNGHBR_PN = 112 Offset Index This is necessary since the mobileNGHBR_PN = 372 Offset Index could be traveling toward any one ofNGHBR_PN = 36 Offset IndexNGHBR_PN = 8 Offset Index these pilots and may need to requestNGHBR_PN = 384 Offset Index soft handoff with any of them soon.NGHBR_PN = 216 Offset IndexNGHBR_PN = 328 Offset IndexNGHBR_PN = 332 Offset IndexNGHBR_PN = 400 Offset IndexNGHBR_PN = 96 Offset IndexRESERVED = 0 MOBILE STATION ACKNOWLEDGMENT 98/05/24 23:14:03.245 [RTC] Order Message The mobile confirms receiving the MSG_LENGTH = 56 bits MSG_TYPE = Order Message Neighbor List Update Message. It is ACK_SEQ = 7 MSG_SEQ = 7 ACK_REQ = 0 ENCRYPTION = Encryption Mode Disabled already checking the neighbor list and ORDER = Mobile Station Acknowledgement Order will do so continuously from now on. ADD_RECORD_LEN = 0 bits Order-Specific Fields = Field Omitted The handoff is fully established. RESERVED = 0 November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 78
  • 79. Handoff Now In Effect, but still check Pilots! All PN Offsets 0 Ec/Io -20 Chips 0 10752 14080 32002 32K PN 0 168 220 500 512Mobile Rake RX Active SetF1 PN168 W61 Rake FingersF2 PN500 W50 T_DROPF3 PN220 W20Srch PN??? W0 Reference PN Neighbor Set T_ADD Continue checking each ACTIVE pilot. If any are less than T_DROP and remain so for T_TDROP time, send Pilot Strength Measurement Message, DROP IT!! Continue looking at each NEIGHBOR pilot. If any ever rises above T_ADD, send Pilot Strength Measurement Message, ADD IT! November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 79
  • 80. The Complete Picture of Handoff & Pilot Sets All PN Offsets 0 Ec/Io -20 Chips 0 SRCH_WIN_A 32K Rake Fingers PN 0 Active Set 512 Pilots of sectorsSRCH_WIN_A T_DROP now used for Mobile Rake RX communication F1 PN168 W61 Reference PN F2 PN500 W50T_DROP Candidate Set SRCH_WIN_N F3 PN220 W20 Pilots requested Srch PN??? W0 by mobile but not set up by system Neighbor Set Pilots suggested T_ADD by system for more checking All other pilots divisible by PILOT_INC but not Remaining Set presently in Active, Candidate, or Neighbor sets T_ADD SRCH_WIN_R November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 80
  • 81. Improved Handoff Control Improved Handoff Control in 1xRTT in 1xRTTNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 81
  • 82. The IS-95 Situation IS-95 handoff is driven by fixed thresholds of pilot strength (Ec/Io) If the mobile notices a new pilot stronger than T_Add, it asks for it immediately If an active pilot drops below T_Drop and stays below for T_Tdrop seconds, the 0 mobile asks for permission to stop using it The mobile has no discretion – the T_Add Ec/Io THRESHOLDS, db -5 and T_Drop values apply no matter what -10 T_ADD -15 T_DROP -20November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 82
  • 83. Disadvantages of Standard Handoff Triggers Active Mobile requests soft handoff with all -3 All Six pilots above T_Add sectors in Pilot Strength • This occasionally leads to some soft handoff! (Ec/Io, db) rigid, less-than-optimum decisions! Active Active Active Problem Situation 1 Active Active T_Add • One dominant, strong signal and a lot of weak ones: -20 – Mobile asks for them all, but only one is really needed! Problem Situation 2 -3 • Heavy pilot pollution, many signals Only One lurk barely below the threshold Sector in soft Pilot Strength handoff! (Ec/Io, db) – Mobile starts call on the best one, but never asks for Active handoffs with any others T_Add – mobile needs handoff to survive! Four -16 signals are as -20 good as a single -10 signal!!November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 83
  • 84. 1xRTT Allows Dynamic Handoff Thresholds A handoff process more intelligent than fixed thresholds • Handoff events driven by smarter, situation-influenced triggers Candidate Set Removal: if that sector isn’t worth adding anymore Neighbor-to-Active transition: only if it’s a worthwhile improvement Removal from Active Set: if that sector isn’t needed anymoreNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 84
  • 85. Standard Equation of a Line The equation of a straight line y is pretty simple. It includes • y, the vertical-axis value • m, the slope of the line b – the ratio of rise/run x • x, the horizontal-axis value • b, the y intercept – the value of y where the line crosses the y axis y = mx + b slope interceptNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 85
  • 86. The Dynamic Handoff Threshold Line +10 +5 Add COMBINED Ec/Io, db Intercept -20 -15 -10 -5 Ec/Io THRESHOLDS, db 0 -5 Combined Ec/Io -10 of Existing Active Pilots T_Add -15 -20November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 86
  • 87. The Dynamic Handoff Threshold Line +10 +5 COMBINED Ec/Io, db Drop -20 -15 -10 -5 Intercept Ec/Io THRESHOLDS, db 0 -5 Combined Ec/Io -10 of Existing Active Pilots -15 T_Drop -20November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 87
  • 88. The Dynamic Handoff Threshold Line +10 +5 Add COMBINED Ec/Io, db Intercept -20 -15 -10 -5 Ec/Io THRESHOLDS, db 0 Drop Intercept -5 Combined Ec/Io -10 of Existing Active Pilots T_Add -15 T_Drop -20November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 88
  • 89. Section G Deeper Handoff Details: Deeper Handoff Details: Search Windows & Timing Search Windows & TimingNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 89
  • 90. The Pilot Searcher’s Measurement ProcessCURRENT PILOT SET CONTENTS R 3 A A A The searcher checks pilots in nested 1 C loops, much like meshed gears. R 12 N N N N N N N N N N N N Actives and candidates N occupy the fastest- N NR112 R R R R R R R R R R R R R R R R R R R R R R R R spinning wheel. N R R R R R R R R R R R R Neighbors are A R R R R R R R R R R R R R next, advancing A AN R R R R R R R R R R R R one pilot for each R R R R R R R R R R R R R Act+Cand. revolution. N R R R R R R R R R R R R R Remaining is slowest, N R R R R R R R R R R R R N N advancing one pilot each R R R R R R R R R R R R R R R R R time the Neighbors revolve. PILOT SEARCHER VIEWED IN SEQUENCE: Typical Elapsed Time = 4 seconds A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N R A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N R A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N A A A C N R Only 3 of 112 remaining set pilots have been checked thus far! November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 90
  • 91. A Quick Primer on Pilot Search Windows The phone chooses one strong sector and PROPAGATION DELAY “locks” to it, assumes the PN offset is what its SKEWS APPARENT PN OFFSETS messages announce. All other offsets are 33 4 defined by comparison to this received signal. Chips Chips In messages, system gives to handset a A BTS B neighbor list of nearby sectors’ true PNs BTS Propagation delay “skews” the apparent PN offsets of all other sectors, making them If the phone is locked to BTS A, the seem earlier or later than expected signal from BTS B will seem 29 chips To overcome skew, when the phone searches for a particular pilot, it scans an earlier than expected. extra wide “delta” of chips (called a “search If the phone is locked to BTS B, the window”), centered on the expected offset signal from BTS A will seem 29 chips Search window values can be set up later than expected. individually for each Pilot set: There are pitfalls if the window sizes are improperly set • too large: phone searches slower • too small: overlook pilots from far away • too large: phone might mis-recognize identity of a distant BTS’ signal One chip is 801 feet or 244.14 m 1 mile=6.6 chips; 1 km.= 4.1 chipsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 91
  • 92. Setting Pilot Search Window Sizes When the handset first powers up, it does an exhaustive search for the best pilot. No windows are used in this process. On the paging channel, the handset learns the SEARCH WINDOW SETTINGS window sizes SRCH_WIN_A, N, R and uses AND PROPAGATION DISTANCES them when looking for neighbors both in idle mode and during calls. Window Datafill N,R Delta Distance Size (Chips) Value Miles KM. When a strong neighbor is requested in a PSMM, the former neighbor pilot is now a candidate. Its 14 (±7) 4 1.06 1.71 offset is precisely remembered and frequently 20 (±10) 5 1.52 2.44 rechecked and tracked by the phone. 28 (±14) 6 2.12 3.42 Window size for actives and candidates can be 40 (±20) 7 3.03 4.88 small, since the windows “float”, drifting with the 60 (±30) 8 4.55 7.32 observed pilot energy of the signal. Only search 80 (±40) 9 6.07 9.77 wide enough to include multipath energy! 100 (±50) 10 7.59 12.2 • This greatly speeds up overall searching! 130 (±65) 11 9.86 15.9 Most post-processing tools deliver statistics on 160 (±80) 12 12.1 19.5 the spread (in chips) between fingers locked to 226 (±113) 13 17.1 27.6 the same pilot. These statistics literally show us 320 (±160) 14 24.3 39.1 how wide the SRCH_WIN_A should be set. 452 (±226) 15 34.3 55.2 Neighbor and Remaining search windows should be set to accommodate the maximum intercell distances which a mobile might experienceNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 92
  • 93. Handoff Problems: “Window” Dropped Calls SITUATION 1 Locked to distant Calls often drop when strong neighbors suddenly appear A 12 mo site, can’t see outside the neighbor search 80 mile un one nearby BTS Ch s tai window and cannot be used to ips ns B establish soft handoff. SRCH_WIN_N = 130 BTS Neighbor Search Window BTS A is reference. 1 mi. SRCH_WIN_N should be set BTS B appears (7-80) chips 7 Chips early due to its closer distance. vel to a width at least twice the This is outside the 65-chip window.Tra propagation delay between Mobile can’t see BTS B’s pilot, but its any site and its most distant strong signal blinds us and the call drops. neighbor site SITUATION 2 Remaining Search Window Locked to nearby SRCH_WIN_R should be set A mo site, can’t see 12 un distant one to a width at least twice the BTS 80 mile tai ns propagation delay between Ch s B ips any site and another site SRCH_WIN_N = 130 BTS which might deliver occasional BTS B is reference. 1 mi. BTS A appears (80-7) chips 7 Chips RF into the service area late due to its farther distance. l This is outside the 65-chip window. Trave Mobile can’t see BTS A’s pilot.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 93
  • 94. Overall Handoff Perspective Soft & Softer Handoffs are preferred, but not always possible • a handset can receive BTS/sectors simultaneously only on one frequency • all involved BTS/sectors must connect to a networked BSCs. Some manufacturers do not presently support this, and so are unable to do soft-handoff at boundaries between BSCs. • frame timing must be same on all BTS/sectors If any of the above are not possible, handoff still can occur but can only be “hard” break-make protocol like AMPS/TDMA/GSM • intersystem handoff: hard • change-of-frequency handoff: hard • CDMA-to-AMPS handoff: hard, no handback – auxiliary trigger mechanisms available (RTD)November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 94
  • 95. Meet the CDMA Meet the CDMA Performance Indicators Performance IndicatorsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 95
  • 96. CDMA Performance Indicators A Flight Data Recorder logs aircraft operational settings. Its CDMA equivalent is a file of RF performance indicators captured by drive-test equipment. Key CDMA parameters and measurements show the condition of the RF environment. They are the primary gauges used to guide CDMA optimization and troubleshooting • some indicate uplink conditions, some downlink, and some, both. • these parameters are collected primarily at the subscriber end of the link, and thus are easy to capture using readily available commercial equipment without requiring assistance at the BSC Understanding these parameters and their important implications requires basic knowledge in several subject areas: • General: RF units, transmitter and receiver basics • CDMA and spread-spectrum signal characteristics – channel definitions – power control systems – basic CDMA call processing flow – signal behavior characteristics in noise and interferenceNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 96
  • 97. Indicator #1: FER FER Frame Erasure Rate Forw • on forward channel ar d 0 2 5 100 (realized at Handset) • on reverse channel R ev er se FER (realized at base station) % • FER is an excellent call quality “summary” statistic FER is the end-result of the whole transmission link • if FER is good, then any other problems aren’t having much effect • if FER is bad, that’s not the problem - it is just the end-result of the problem – we must investigate other indicators to get a clue what is going onNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 97
  • 98. Indicator #2: Received Power at the Handset Mobile Receive Power RX Level overload>> • usually expressed in dBm Handset Receiver -40 • measured derived from Rake LNA IF R handset IF AGC voltage ≈ x ≈ R • broadband, “unintelligent” BW BW R ~30 LO 1.25 measurement: includes all S <<too weak MHz. MHz. RX Level RF in the carrier bandwidth (from AGC) -90 regardless of source, not -105 just RF from serving BTS Receive power is important, but it’s exact value isn’t critical • too much received signal (-35 dbm or higher) could drive the phone’s sensitive first amplifier into overload, causing intermod and code distortion on received CDMA signals • too little received signal (-105 or weaker) would leave too much noise in the signal after de-spreading, resulting in symbol errors, bit errors, bad FER, and other problemsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 98
  • 99. Indicator #3, Ec/Io - What does it mean? Why can’t we just use the handset’s received power level to guide Handset Receiver Rake handoffs? LNA IF R • Because it is a simple total RF ≈ x ≈ R BW BW R power measurement, the total of ~30 MHz. LO 1.25 MHz. S all sectors reaching the mobile RX Level (from AGC) We need a way to measure the signal strength of each sector individually, and we must be able to measure it quickly and simply The solution is to use each sector’s pilot (Walsh 0) as a test signal to guide handoffs • At the mobile, if the pilot of a certain sector is very strong and clean, that means we also should be able to hear a traffic channel on that sector, so handoff would be a good idea • if the pilot of a certain sector is weak, then we probably won’t be able to get much benefit from using a traffic channel on that sectorNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 99
  • 100. How Ec/Io Varies with Traffic Loading Light Traffic Loading Each sector transmits a certain amount of power, the sum of: • pilot, sync, and paging Ec/Io = (2/4) = 50% • any traffic channels in use = -3 db. Paging 1.5w at that moment Sync 0.5w I0 Pilot 2w EC Ec/Io is the ratio of pilot power to total power • On a sector with nobody Heavily Loaded talking, Ec/Io is typically Traffic Channels about 50%, which is -3 db • On a sector with maximum Ec/Io = (2/10) 6w = 20% traffic, Ec/Io is typically I0 = -7 db. about 20%, which is -7 db. Paging Sync 1.5w 0.5w Pilot 2w ECNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 100
  • 101. How Ec/Io varies with RF Environment One Sector Dominant In a “clean situation”, one Channels Traffic sector is dominant and the Io = -90 dbm 4w mobile enjoys an Ec/Io just Ec = -96 dbm I0 Paging 1.5w as good as it was when Ec/Io = -6 db Sync 0.5w Pilot 2w EC transmitted In “pilot pollution”, too many Many Sectors, Nobody Dominant sectors overlap and the Traffic BTS10 Sync & Paging mobile hears a “soup” made Pilot Traffic BTS9 up of all their signals Sync & Paging Pilot Traffic BTS8 • Io is the power sum of all Io = 10 signals Sync & Paging Pilot Traffic BTS7 each -90 dbm Sync & Paging the signals reaching the Pilot = -80 dbm Traffic BTS6 mobile Ec of any one Sync & Paging Pilot Traffic BTS5 I0 Sync & Paging • Ec is the energy of a sector = -96 Pilot Traffic BTS4 single sector’s pilot Ec/Io = -16 db Sync & Paging Pilot Traffic BTS3 Sync & Paging • The large Io overrides the Pilot Traffic Sync & Paging BTS2 weak Ec; Ec/Io is low! Pilot Traffic BTS1 Sync & Paging Pilot ECNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 101
  • 102. Indicator #4: Handset Transmitter Power Subscriber HandsetTXPO Handset Transmit Power BTS Receiver>> Rake R • Actual RF power output of the LNA ≈ Viterbi DUP x IF R Σ handset transmitter, including Decoder combined effects of open TXPO PA ∼ LO R LO Open Loop S loop power control from x ~ Closed Loop Pwr Ctrl receiver AGC and closed IF I Long PN Vocoder loop power control by BTS x Orth IF Mod x Mod FEC • can’t exceed handset’s x Q <<Transmitter maximum (typ. +23 dBm) Typical TXPO: +23 dBm in a coverage hole TXPO = -(RXdbm) -C + TXGA 0 dBm near middle of cell C = +73 for 800 MHz. systems -50 dBm up close to BTS = +76 for 1900 MHz. systems What is the right power TX level? Whatever the BTS asks for! • As long as closed loop control is working, the phone’s opinion isn’t the last word. Just do what the BTS wants!! • However, if the BTS ever asks the phone to do the impossible, something is wrong (lower than -60 dbm, higher than +23 dbm)November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 102
  • 103. Indicator #5: Transmit Gain AdjustWhat is Closed Loop Transmit Gain Adjust (TXGA)? • The power correction the base station is asking the mobile to make right now, in real-time • At the beginning of a call, before the power control bits begin, it is zero. Then the power control bits begin, 800 per second. • During a call, TXGA is the running total of all the power control bits which have been received thus far. • Each power control bit asks for a 1 db correction, up or down • Each power control bit is based on the base station’s latest new decision: mobile is too strong, or mobile is too weak -- there is no cumulative error, since each decision is “fresh” Typical Transmit Gain Adjust 0 dB TXPO = -(RXdbm) -C + TXGA C = +73 for 800 MHz. systems -10 dB = +76 for 1900 MHz. systems -20 dB Time, SecondsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 103
  • 104. Closed Loop Power Control Dynamics The figures at right show the power control reactions to a sudden change in path loss The sudden change in path loss causes a sudden change in handset received signal Both open loop and closed loop control race to get the phone back to the right new power and succeed in about 10 milliseconds Open loop continues to approach the correct value better and better on its own 40 milliseconds later, no net closed loop correction is needed.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 104
  • 105. Section Identification Problem “Signatures” Problem “Signatures”November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 105
  • 106. “Signatures” of Common Conditions SIGNATURE:The key CDMA RF Performance GOOD CALLIndicators provide powerful cluesin cause-and-effect analysis for FFER RXL EC/IO TxGa TxPounderstanding problem conditions 100% -30 0 +25 +23 -40 +10There are many common 0conditions which are easy to -6 +10 -10recognize from their characteristic 50% -10 0“signatures” -- unique -20relationships among the key -10 -30 -15indicators which are observed 10% -90 -40 5% -20when these conditions exist 2% -100 -50 0% -110 -20 -25We will use the simplified format FFER RXL EC/IO TxGa TxPoshown at right to display the keyindicators for each of several Messaginginteresting cases. BTSNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 106
  • 107. Signature of a Successful Call SIGNATURE: If the mobile station originates GOOD CALL successfully, remains in service FFER RXL EC/IO TxGa TxPo area, and makes normal release, 100% +23 -30 0 +25 data will show: -40 +10 • Low forward FER 0 -6 +10 • Receive power > -100 dBm -10 • Good Ec/Io (> -12 dB) 50% -10 0 -20 • Normal Transmit Gain Adjust -10 -30 (actual value depends on site -15 configurations, loading & 10% -90 -40 NOM_PWR setting) 5% -100 -20 -50 2% • Transmit power < +20 dBm 0% -110 -20 -25 • Good Messaging FFER RXL EC/IO TxGa TxPo • Parsed message files will contain a full set of normal messages. BTS MessagingNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 107
  • 108. Signature of a Dropped Call in Poor Coverage SIGNATURE: If a mobile station is taken out DROPPED CALL, BAD COVERAGE of the service area or into a coverage hole, and only data FFER RXL EC/IO TxGa TxPo 100% +23 -30 0 +25 from the mobile station is available, the log files will show -40 +10 the following characteristics: -6 +10 0 • High forward FER -10 50% -10 0 -20 • Low receive power (<-100 dBm) -10 -30 -15 -40 • Low Ec/Io (< -10 dB) 10% -90 5% -20 -100 -50 • Higher-than-normal Transmit 2% 0% -110 -20 -25 Gain Adjust (actual value depends on site configurations, loading, FFER RXL EC/IO TxGa TxPo NOM_PWR setting) • Higher-than-normal transmit BTS Messaging power (> +20 dBm) • Poor messaging on both linksNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 108
  • 109. Signature of Forward Link Interference SIGNATURE:Characteristics of data for a phone FORWARD LINK INTERFERENCEexperiencing forward linkinterference from a source other FFER RXL EC/IO TxGa TxPothan the current BTS: 100% -30 0 +25 +23 • High forward FER -40 +10 • Good receive power (> -100 dBm) -6 +10 0 • Low Ec/Io (< -10 dB) -10 50% -10 0 • Higher-than-normal Transmit Gain -20 Adjust -10 -30 -15 • Normal transmit power (< +20 10% -90 -40 dBm) 5% -20 -100 -50 2% • Poor forward link messaging 0% -110 -20 -25 – unreliable at best and may be FFER RXL EC/IO TxGa TxPo the actual cause of the drop. BTS MessagingNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 109
  • 110. A CDMA Drop Example: Forward Link Case A mobile using Site A comes FORWARD LINK DIES down the highway and s A ct ion suddenly begins to see the s tru BTS signal of Site B B Ob BTS If the mobile begins soft el av handoff with site B, everything Tr continues to go well If the mobile cannot begin handoff with B for any reason, the call is doomed B grows stronger and stronger. Mobile’s open-loop instinct is to transmit • site B’s signal will override weaker; closed-loop correction from A site A’s signal, making it goes higher and higher, maintaining the mobile at the right power. unreadable Finally B obscures A, which disappears • as soon as the FER goes in an explosion of FER. The mobile mutes since it can’t hear power control too high, a fade timer will bits, and a fade timer or message timer start the the mobile will kills the call in a few seconds. eventually dieNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 110
  • 111. Signature of Reverse Link Interference SIGNATURE:Characteristics of data for a phone REVERSE LINK INTERFERENCEwhose BTS has a raised noisefloor due to reverse link FFER RXL EC/IO TxGa TxPointerference 100% -30 0 +25 +23 • Good forward FER -40 +10 • Good receive power (> -100 dBm) -6 +10 0 • Good Ec/Io (> -10 dB) -10 50% -10 0 • Higher-than-normal Transmit Gain -20 Adjust -10 -30 -15 • Higher-than-normal transmit power 10% -90 -40 (< +20 dBm) 5% -20 -100 -50 2% • Poor reverse link messaging 0% -110 -20 -25 – in the message files, you’ll FFER RXL EC/IO TxGa TxPo see repeats of messages on the forward link and reverse Messaging BTS linkNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 111
  • 112. A CDMA Drop Example: Reverse Link Case When a cell is penetrated by a REVERSE LINK DIES mobile not under its own s ct ion power control, bad things s tru happen! B Ob BTS • The foreign mobile is being el av power controlled by a Tr more distant cell, so it is transmitting louder than It was a beautiful day in the neighborhood appropriate for all the mobiles on site B until the grim • the local mobiles must reaper arrived, transmitting at high power to maintain its link with distant Cell A. power up in a deadly race Cell B tried to power up each of its to keep up with the individual mobiles so they would be interferor received as strong as the new interferor, but mobiles more distant than the • local mobiles can still hear interferor just couldn’t keep up, and died. Eventually the interferor died from the cell fine; the forward forward link interference, too. link is just great, to the If only the interferor had a soft handoff, all very end of this violence could have been avoided.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 112
  • 113. Solving the #1 Death Scenario: Failed Handoff What Went Wrong??! ns FORWARD LINK DIES A REVERSE LINK DIESns tio tio ruc BTS uc B b st B bstr lSteps in the Handoff Process BTS O el BTS O ve av Tr a Tr Mobile’s searcher notices see the needed new pilot Mobile sends PSMM ask requesting handoff Why didn’t the mobile ask for handoff? System sets up the handoff: •channel elements • New sector not on neighbor list do •forward power • Neighbor Search Window too Small? BTS •space in packet pipes Simulcasting begins! • BTS in “island mode”, wrong PN? Now the system can hear the mobile better! Why didn’t the BTS set up the handoff? tell System tells mobile how to • Old BTS didn’t hear mobile – rev link hear the new sectors: BTS Handoff Direction Message interf? Now the mobile can hear • No resources available on new BTS? the system better, too! • T-1 unstable, messages lost Mobile confirms completion: ok! Handoff Completion Message Why didn’t the mobile do the handoff? System makes new neighbor list, • Couldn’t hear BTS, Fwd link interf? tell sends to mobile: Neighbor List BTS Update MessageNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 113
  • 114. Pilot Pollution Ec/Io value at each Io BTS TX When a large number of -80.0 -3 Ec/Io of Multiple CDMA Signals CDMA signals are received Signal Strength Ec/Io at about the same strength, -90 -13.0 1 -90 -13.0 2 they cause severe -90 -13.0 3 interference to each other -90 -90 -13.0 4 -13.0 5 -90 -13.0 6 • this is called Pilot -90 -13.0 7 Pollution -90 -90 -13.0 8 -13.0 9 -90 -13.0 10 1 2 3 4 5 6 7 8 9 10 The cure for pilot pollution -80.0 Sum Power is to eliminate unneeded Ec/Io value signals which really weren’t at each intended to serve this Io -73.9 BTS TX -3 location anyway, and to Signal Ec/Io of Multiple CDMA Signals boost the one or a few Strength Ec/Io -90 -19.1 1 signals which were -90 -19.1 2 intended to serve this -90 -90 -19.1 3 -19.1 4 location -75 -90 -4.1 5 -19.1 6 See the first page of the -90 -90 -19.1 7 -19.1 8 workbook ECIOPLAY.XLS -90 -90 -19.1 9 -19.1 10 1 2 3 4 5 6 7 8 9 10 -73.9 Sum PowerNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 114
  • 115. Pilot Pollution/Handoff/Composite Ec/Io Demo See the second page of the workbook ECIOPLAY.XLSEc/Io, Handoff, and Rake Finger Pilot Status Relative Energies of Multiple CDMA Signals % Nominal Sum Max # Over- Max RF Comp Max # Pilots in Pilot Energy Sync, Paging, Traf fic%Pilot head Power Power osite Lockable Soft 1Power Power W Io Ec/Io Rake Fingers Handoff T_ADD 10% 20% 12 -86.2 -3.0 3 6 -12 0.9Traffic Trans Path SignalLoading mitted Loss, Streng 0.8% Ec/Io dB th Ec/Io 0% -3.0 120 -86.2 -3.0 Rake Locked Handoff 1 0.7 0% -3.0 200 -166.2 -83.0 Interferor 2 0% -3.0 200 -166.2 -83.0 Interferor 3 0.6 0% -3.0 200 -166.2 -83.0 Interferor 4 0% -3.0 200 -166.2 -83.0 Interferor 5 0.5 0% -3.0 200 -166.2 -83.0 Interferor 6 0.4 0% -3.0 200 -166.2 -83.0 Interferor 7 0% -3.0 200 -166.2 -83.0 Interferor 8 0.3 0% -3.0 200 -166.2 -83.0 Interferor 9 0% -3.0 200 -166.2 -83.0 Interferor 10 0.2 0% -3.0 200 -166.2 -83.0 Interferor 11 0% -3.0 200 -166.2 -83.0 Interferor 12 0.1 0% -3.0 200 -166.2 -83.0 Interferor 13 0% -3.0 200 -166.2 -83.0 Interferor 14 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14Only grey-shaded fields can be changed. Other fields calculate automatically. I nd i vi d ual Sig nal sTo unlock all cells, select TOOLS>PROTECTION>UNPROTECT SHEET. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 115
  • 116. System Performance Optimization Basic PN Planning and Basic PN Planning and Search Window Considerations Search Window ConsiderationsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 116
  • 117. Introduction to PN Planning and Search Windows In PN planning and setting Search Windows, several pitfalls must be avoided. These slides explain most of the basic facts, background, principles, and practical considerations involved.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 117
  • 118. Short PN Basics: PN Offsets Distinguish Sectors A Phone B Rake Receiver LNA IF PN A Walsh X x ≈ x ≈ PN B Walsh Y ∑ Decoding Vocoder C BPF BPF PN C Walsh Z LO Pilot Searcher D Each sector uses the short PN code, but at a different timing delay called its PN offset • PN delays are settable in 64-chip steps called "PN offsets" – For example, PN offset 100 means 6,400 chips of delay • PN short code is 32,768 chips long - room for 512 different PN offsets In the rake finger of a mobile in soft handoff, the short PN code is generated in step with just one sector the mobile is trying to hear • The rake finger hears the matching sectors signal, ignores all others • The rake finger next decodes the walsh code of the desired channel from that sector, ignoring all other users on that sectorNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 118
  • 119. A Practical "Rule of Thumb" to Remember Received: Transmitted: PN 101 PN 100 6,464 chips delay 6,400 chips offset 9.70 miles = 64 chips = 1 PN ABCDEFGHIJKLMNOPQRSTUVWXYZ1234567890abcdefghijkmnopqrstuvwxyz!@#$%^&*()_+ The PN chips SEEN by the mobile are what the base station transmitted 64 chips in the past! What the base station is really doing now, its true PN offset, is 64 chips later than what the mobile sees. So the base stations signal at the mobile seems to be one PN lower than it was actually transmitted. Mobile BTS The signal of a base station roughly 10 miles distant will SEEM to be one PN higher than it was transmittedNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 119
  • 120. Propagation Delay changes apparent PN Offset Base stations transmit signals on assigned, fixed short PN delays called PN Offsets PN360 Transmitted signals encounter additional delay traveling to the mobile • ~6.7 chips/mile = ~4.1 chips/kilometer 10 KM 41 chips These additional delays can become significant and cause errors at the mobile! • Failure to recognize certain signals • Misidentification of signals, recognizing 2 KM on BTS as another 8 chips • Improper combination of signals - listening to the wrong BTS and trying to PN200 decode and combine its signal in a handoffNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 120
  • 121. Mobile Timing: the Reference PN UNKNOWN EXTRA PROPAGATION DELAY All PN Offsets How many chips???? 0 Active Pilot Ec/Io Pilot Searcher Scans All PNs -20 Chips 0 Rake Fingers 32KMobile System Acquisition Process PN 0 512 SYNC CHANNEL MESSAGE Scan entire range of PNs 98/05/24 23:14:09.817 [SCH] MSG_LENGTH = 208 bits Lock to strongest Pilot found MSG_TYPE = Sync Channel Msg • Put rake fingers on multipaths Reference PN P_REV = 3, MIN_P_REV = 2 SID = 179 NID = 0 PILOT_PN = 168 Offset Index LC_STATE = 0x0348D60E013 • Earliest arriving multipath is "reference PN" SYS_TIME = 98/05/24 3:14:10.160 LP_SEC = 12 Read sync channel message LTM_OFF = -300 minutes DAYLT = 0, PRAT = 9600 bps • Learn what PN this is! But theres no way to know how many chips of propagation delay have happened before this signal was received • The mobile is "blind" to whatever this error may be; so the mobiles internal PN reference is late by an unknown amount • Every pilot the mobile looks for will appear to be early or late too! November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 121
  • 122. What are "Search Windows"? New pilots usually seem earlier or later than their official PNs from the neighor list • Some have come from nearer, some PN360 from farther, than the reference PN A mobile must look for pilot energy through a range of chips earlier and later than the 10 KM exact expected PN offset of the signal it is 41 chips trying to measure These "tolerance" ranges are called "Search Windows" • SRCH_WIN_A applies to active and candidate pilots +41 • SRCH_WIN_N applies to neighbors PN200 +8 360 • SRCH_WIN_R applies to remaining 2 KM 360+33c 8 chips Search windows are chosen by RF SRCH_WIN_N engineers and transmitted to the mobile in messages from the BTSNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 122
  • 123. What Search Window Values Can Be Set? SRCH_WIN_val Width, Chips Search windows cant be set to the exact number 0 4 (±2) of chips desired; each window can be set to a 1 6 (±3) value from the list at right 2 8 (±4) Remember the widths are total and apply with the 3 10 (±5) mobiles reference at the center. 4 14 (±7) • For example, SRCH_WIN_N = 10 means 5 20 (±10) when the mobile is checking for neighbor 6 28 (±14) pilots, it will search a range 100 chips wide, centered on what it thinks is the reference PN. 7 40 (±20) 8 60 (±30) – The mobile will search from 50 chips earlier to 50 chips later than the exact PN 9 80 (±40) it expects to find 10 100 (±50) 11 130 (±65) Search windows should be wide enough to include needed signals, but not unnecessarily wide 12 160 (±80) 13 226 (±113) • Grossly over-wide search windows will slow down the mobiles overall pilot searching 14 330 (±165) speed 15 452 (±226)November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 123
  • 124. Search Window Settings: Neighbor Set Neighbor Search Window Example The neighbor search window must be set wide enough to include the energy of any needed neighbor pilot PN360 The mobile at right is using PN200 as its Neighbor Sector reference (and only active) pilot 10 KM To the mobile, the pilot of neighbor sector 41 chips PN360 seems 33 chips late SRCH_WIN_N must be set to at least 2 x 33 = 66 chips wide so the PN360 pilot can be noticed by the mobile +41 The closest search window setting above PN200 +8 66 chips is SRCH_WIN_N = 9, which is 80 360 2 KM chips wide 8 chips 360+33c Active SRCH_WIN_N SectorNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 124
  • 125. Worst-Case Wide Neighbor Window Situation BTS A BTS B 1/2 mile 12 miles In some terrain, it is possible for a mobile to be very close to one BTS and far from another BTS, yet need them both in soft handoff This occurs when local terrain or buildings obstruct the signal of the near BTS, making it much weaker than normal • The far BTS may have much more favorable conditions, such as an over-water path • The signals of the two BTSs may seem equally strong! Almost the entire distance between the BTSs appears as timing skew • If near BTS is reference PN, distant BTS is late this number of chips • If far BTS is reference PN, near BTS is late by this number of chipsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 125
  • 126. Safe Initial Neighbor Search Window Value Examine a cell map for an area of your system Determining Safe Initial SRCH_WIN_N Identify the farthest-apart pair of cells likely to be used in soft handoff F D • Their distance separation determines maximum timing skew a mobile could ever E possibly encounter in this part of the system B 11.5 KM Calculate the timing skew in chips C • 6.7 chips times miles or 4.1 chips times A kilometers • Safe required window size = two times the Required Window skew = 4.1 x 11.5 x 2 = 94.3 chips Refer to table to convert required window size If locations exist near site 10 SRCH_WIN_N = A in chips to required value of SRCH_WIN_N where mobiles are in handoff with site F, mobiles could encounter After thorough drive-test data is available, it neighbor pilot timing skews as may be possible to reduce SRCH_WIN_N if large as the A-F distance. If observed delay spread is significantly locked to A, F looks late by this narrower than the window amount. If locked to F, A looks early by this amount. Window must be twice the skew value.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 126
  • 127. Search Window Settings: Remaining Set Remaining set search window size is determined by maximum possible timing F skew in the same way as for neighbor set D window E Recommended SRCH_WIN_R is one or two steps greater than SRCH_WIN_N B 11.5 KM Remaining set pilots can be requested by the C mobile in a PSMM but the system cannot A assign traffic channels since it uses the Neighbor Pilot Database as its cross- reference for identification of their base stations There is still value in allowing mobiles to find and request remaining pilots, since the requests help system RF engineers identify missing pilots that should be added to the neighbor lists of various sectorsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 127
  • 128. Search Window Initial Settings: Active Set Neighbor and Remaining search window centers are indexed against the mobile’s Active Search Window Reference PN 40 chips wide (typical) Each active search window is different – a “floating” window centered over the earliest -20 0 +20 observed multipath energy during the previous Ec/Io mobile searcher scan of that individual pilot Active search windows need not accommodate distance-based timing skews – they float centered on their respective pilots! The only timing variations they must Earliest Detected accommodate are multipath delay spreads Multipath The earliest arriving multipath Multipath delay spreads are determined by seen by the mobile during this terrain and clutter-driven scattering and searcher sweep will be used reflection of the signal as the center of this active window on the next searcher Measurements are better than predictions to sweep! This makes each set SRCH_WIN_A active search window "track" individually with its pilot.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 128
  • 129. SRCH_WIN_A Settings from MeasurementsTypical active set delay spread from actual drive-testsNotice the narrow distribution of energy!28-chip width, SRCH_WIN_A = 6, is enough for this caseDrive-test your own system to determine your own value of spread • It is determined by the signal-scattering characteristics of your terrainNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 129
  • 130. SRCH_WIN_A Special Consideration SRCH_WIN_A, Chips 10 14 20 28 40 60 20 No No No No No No SRCH_WIN_N, Chips 28 No No No No No No 40 No No No No Yes NoActive set delay spread is very narrow – 60 No No No Yes Yes Yescan the active search window be set 80 No No No Yes Yes Yesnarrow too? 100 No Yes Yes Yes Yes Yes 130 Yes Yes Yes Yes Yes YesMobile reference timing occasionally 160 Yes Yes Yes Yes Yes Yes“jumps” due to false early-window 226 Yes Yes Yes Yes Yes Yesdetection of the reference pilotThere is a dynamic relationship between mobile reference timingstability and the active and neighbor search window sizesThe chart above shows which combinations of SRCH_WIN_Aand SRCH_WIN_N are safe and stable for all mobilesNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 130
  • 131. The Potential for PN Problems and Conflicts After seeing the skewing effects of propagation, it is easy to anticipate problems of PN confusion and misidentification!There are many different kinds of possible PN problems: Two same-PN base stations with areas of coverage overlap • Mobiles cant distinguish them, experience horrible FER Combining unintended signals into the handoff mix being heard • The new signals cause interference instead of helping Mistaken identity of signals when requesting handoff • The wrong base station is added, the mobile cant hear it Running out of available PNs due to bad parameter choicesFortunately, these problems can be avoided by careful planning!November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 131
  • 132. PILOT_INC Helps Avoid PN Problems Imagine a network with base stations spaced approximately 10 miles apart - this is 1 PN offset! D Recall if we use adjacent PNs for adjacent base stations, there will be locations where their PNs are close together or even indistinguishable It would be smart to assign PNs farther apart! If properly set, PILOT_INC can prevent this problem • Only PNs divisible by PILOT_INC are allowed to be assigned to sectors PILOT_INC can be chosen from 1 to 16 • If too small, interfering PNs can be assigned • If too large, the pool of available PNs is small PILOT_INC is set based on the density of cells • 3 or 4 in typical cities with suburban density • 2 in dense urban environments • 6 or 8 in very rural areasNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 132
  • 133. Co-Active PN Demodulation Errors ACTIVE SEARCH WINDOWBTS A BTS BPN 142 PN 142 x miles x miles Mobile is midway between two BTSs with the same PN, in a call on BTS A PN energy of BTS A and B is indistinguishable in active search window Rake fingers may be assigned to both A and B energy • If the walsh code used on A also happens to be in use by someone on BTS B, demodulation of B will cause severe FER • The mobile audio will frequently clip and mute, and the call may drop • All the while, the phone will see very good Ec/Io since both A and B are recognized as good energy! Solution: Two different BTS covering the same area should never have the same PN offset. Change the PN offset for one of the sectors involved.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 133
  • 134. Adjacent-Active-PN Demodulation ErrorsBTS A BTS BPN 100 PN 99 ACTIVE SEARCH WINDOW 1 mile 11 miles Mobile is in a call on BTS A from 1 mile away; A is the reference PN The signal from BTS B on PN 99 travels 11 miles to the mobile and is approximately as strong as BTS A due to terrain effects Due to propagation delay, the signal of B is skewed and falls inside the active search window of the mobile for A • A and B energy are indistinguishable to the mobile • Rake fingers may be assigned to both A and B multipaths • If the walsh code used by the mobile on A also is in use by someone else on B, the mobile may demodulate their symbols and combine them with its own symbols from BTS A • This would cause severe FER and possibly a dropped call Solution: The PNs of the two BTSs are too close together. Use a different PN offset for BTS B.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 134
  • 135. Adjacent-Neighbor PN Recognition Errors BTS A 20 miles BTS G PN 100 PN 198 BTS BTS NEIGHBOR SEARCH WINDOWmo un tai ns X BTS FPN 200 BTS Mobile is in a call on BTS A, PN 100 Mobile checks neighbor PN 200 to see if handoff needed with BTS F Energy from distant BTS G on PN 198 is skewed so that it falls in the neighbor search window for PN 200; mobile asks for handoff with F The system sets up a traffic channel on BTS F - but mobile hears G! If the walsh code assigned on F happens also to be in use on G, the mobile may put a rake finger on it and include it in the mix • Severe FER and a possible dropped call will result! Solution: Careful RF design to avoid such "pockets" of distant coverage • If signal of G cant be reduced by RF methods, assign it to a different PN November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 135
  • 136. Sector PN Assignments: Consecutive Assignment Use only PNs divisible by PILOT_INC. 12 4 • PILOT_INC is chosen large enough to prevent aliasing of pilots in adjacent cells 96 88 8 24 16 Assign PNs in sequence to the sectors of all 92 the base stations 84 76 20 Common Usage: This is the typical default 108 100 80 36 28 method used in Nortel and Motorola CDMA networks 104 32 72 64 Advantage • Simple assignment 120 112 68 48 40 • When adjacent PNs are observed in the 116 44 field, they are known to be from sister 60 52 sectors of the same BTS or from nearby 56 BTSsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 136
  • 137. Sector PN Assignments: Segment AssignmentAssign only PNs divisible by PILOT_INC 340 4 • PILOT_INC is chosen to avoid aliasingDifferent ranges of PN values are reserved 368 32 172 344 8 • First 1/3 of PN offsets for alpha sectors 200 176 364 28 • Second 1/3 of PN offsets for beta sectors • Third 1/3 of PN offsets for gamma sectors 372 36 196 348 12Although 512/3 = 170.666, the value 168 is 204 180usually used for the inter-sector PN increment 360 24Common Usage: default in Lucent networks 376 40 192 352 16Advantage: In the field, interference is 184suddenly noticed from PN 468. Quickly, what 208 356 20is the source of it? 188 • Definitely some cells gamma sector!November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 137
  • 138. Course RF200 Section II. Introduction to CDMA Introduction to CDMA Performance Data Performance DataNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 138
  • 139. What Data is Available for Performance Study? CDMA NETWORK EQUIPMENT HANDSET Switch Access Mgr./BSC-BSM BTS SLM CM NOIS Messages GPSR IS-95/J Std 8 GPSR Messages BSM DMS-BUS TFU1 NMIS Messages CDSU CDSU CDSU DISCO TFU1 DISCO 1 Ch. Card ACC CDSU CDSU LPP ENET LPP CDSU DISCO 2 CDSU Σα Txcvr A RFFE A DTCs CDSU Σβ Txcvr B RFFE B Handset SBS CDSU Σχ Txcvr C RFFE C Vocoders Messages PC-based IOC Selectors QC-Specific Messages Mobile Data Selector IS-95/J Std 8 Messages Capture ToolsSwitch OMs, Logs pegs, logs Unix-based, Various PC-based PC-based External Data Analysis Mobile Data Analysis Post-Processing Post-Processing Tools Tools Tools CDMA data for analysis flows from three sources: • Switch, CDMA peripherals and base stations, and the Handset Various software and hardware tools are available for collection and analysis of each of these streams of data Data contains messages and various indicators of RF performance November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 139
  • 140. Resources on System and Switch Data CDMA networks are complex, including large conventional telephone switches, high-capacity CDMA system peripherals such as BSCs, CBSCs, and Access Managers, and many base stations (BTSs) which are usually multi-carrier • A network is literally a CITY of processors and software The specific performance statistics and event counters (peg counts) are best described in official documentation from the network manufacturers • However, current documentation always seems to lag behind cutting- edge hardware and software releases Each manufacturer publishes help on its own hardware & software: • Lucent: Wireless Networks Systems Documentation CDs – Application notes; many good training courses • Nortel: Helmsman CD, documents, training courses • Motorola: Planning Guides, documents, training courses This course focuses on the generic key indications to observe, and the analytical skills and perspective necessary for optimization • The manufacturers documentation will describe the actual counters and measurements available from your networkNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 140
  • 141. System Data and System Data and Statistical Analysis Statistical AnalysisNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 141
  • 142. Statistical CDMA Performance IndicatorsEach network platform (Lucent, Nortel, Motorola) has its own Dropped Call Statistics unique set of available statistics. Failed Access AttemptsThese indications are collected from Blocking Statistics the Switch, CDMA peripherals, and the base stations. They can be • BTS sector level analyzed, tracked and trended for • BSC resource level system performance benchmarking. • Switch resource levelThese indications should be examined • PSTN trunking level from many perspectives: overall for an entire system, by individual Counts of specific call sector and cell, and both in processing error events absolute numbers and by percentages of total traffic.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 142
  • 143. Typical Network Performance RF Acc-Fails Screen Calls %RF Acc-Fail % Screen Cal Total-Block %Tot-Block Calls-Drop MTA-Name Call-Succ. Call-Att. %-Succ. %-Drop Period CellsExample H Week ALL 1,147,447 1,123,308 97.9% 443 0.04% 12,429 1.1% 20,015 1.7% 11,229 1.0%Average of Others 96.1% 2.1% 2.8% 2.1% 0.6%Comparing A Sys ll tems Sorted B D y aily Traffic LevelExample System D D ay All 1,338,386 1,240,937 92.7% 44,593 3.33% 35,329 2.64% 30,576 2.28% n/a n/aExample System E D ay All 355,247 347,325 97.8% n/a n/a 7,922 2.23% n/a n/a n/a n/aExample System B D ay All 227,257 222,425 97.9% 388 0.17% 4,444 1.96% n/a n/a n/a n/aExample System C D ay All 220,707 205,766 93.2% 6,312 2.86% 6,090 2.76% 5,088 2.31% n/a n/aExample System A D ay All 209,621 205,461 98.0% n/a n/a n/a n/a 3,297 1.60% 1,327 0.6%Example System F D ay All 206,482 198,945 96.4% n/a n/a 7,537 3.65% 4,556 2.29% n/a n/aExample Sys tem H D ay A LL 163,921 160,473 97.9% 63 0.04% 1,776 1.1% 2,859 1.7% 1,604 1.0%Example System G D ay All 148,765 143,633 96.6% n/a n/a 5,132 3.45% 3,074 2.14% n/a n/a Here is a comparison of typical network performance in the industry against a new rural wireless system with light loading How does your system compare against the industry norms? Against the lightly loaded rural system? November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 143
  • 144. Another Network Performance ExampleNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 144
  • 145. Lucent System Data Lucent System Data Examples ExamplesNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 145
  • 146. Lucent System Data Examples 179 2 67 179 2 28 179 2 30 179 2 121S ys/ ECP / Ce ll/ N a m S MIT H SP R IN G 179 2 10 IN GLEW OO N OLEN SV IL CLAR KSV IL 179 2 1 179 2 45 e / La be l TOTALS S BOBCAT D LE LE/ BR ILE Y T E XT R ON FAR MER S % CD MA E st Ca lls 96.83 93.55 93.58 94.18 94.36 94.44 94.67 94.73 R e Acquir e d_Ca lls 2.84 3.22 2.61 3.89 2.38 5.26 2.65 2.06 CCE e rla ngs 6,580.44 62.60 128.68 71.45 63.54 36.16 76.37 115.21 CD MA_CE U sa ge 2,368,959 22,535 46,323 25,722 22,873 13,016 27,494 41,476 Prim_CS CE _U se 1,451,816 9,300 19,788 13,689 11,113 8,448 15,965 23,219 % P rim_CS CE _U se 61.28 41.27 42.72 53.22 48.59 64.90 58.07 55.98 S e c_CS CE _U se 917,143 13,235 26,535 12,033 11,760 4,568 11,529 18,257 % CD MA SoftH O U se 38.72 58.73 57.28 46.78 51.41 35.10 41.93 44.02 % CD MA SU Fa il 2.79 6.14 5.68 5.44 3.62 3.68 4.64 5.04 CD MA Lost_Ca ll 1,722 15 42 20 10 64 15 35 % CD MA Lost Ca lls 1.17 1.67 2.18 1.18 0.89 5.98 0.98 1.44 T otCD MA Fa ilure s 7,856 95 208 143 77 108 102 206 CD MAT otl Origins 5,069 65 143 89 47 73 67 141 CD MAT otl T e rmins 2,787 30 65 54 30 35 35 65 CD MA Ma int Busy 0 0 0 0 0 0 0 0 CD MAOnly P rfSz 80 0 0 0 1 0 0 0 CD MA_Org T rm_Ovf 713 0 0 0 0 0 0 0 CD MA Org_S z 109,076 680 1,500 1,236 771 828 1,229 1,862 CD MA Org_Asn 105,970 659 1,454 1,197 752 786 1,192 1,824 CD MA Pg_R sp_S z 46,720 313 611 640 445 369 430 755 CD MA T rm_Asn 44,951 301 590 603 412 329 414 736 CD MA R e q_Alg 4,426 32 55 73 29 63 44 54 November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 146
  • 147. Lucent Overload Data Examples from AutopaceS ys/ E CP / Ce ll/ N 179 2 1 179 2 1 179 2 1 179 2 2 179 2 2 179 2 2a me / Ante nna T E XT R ON 1 T E XT R ON T E XT R ON BE LMON T BE LMON T BE LMON TID / Ant_N a me TOTALS Ante nna :1 2 Ante nna :2 3 Ante nna :3 1 Ante nna :1 2 Ante nna :2 3 Ante nna :3 CD MA_Acs Chn_Oc 5,921 30 28 10 27 13.00 13.00 CD MA_Avg S q_D G 1,123,466 6,187 6,157 6,088 6,168 5,016.00 4,818.00 CD MA_Fwd P COLdur 581 12 4 2 0 0.00 0.00 CD MA_Fwd P COLcnt 339 4 4 1 0 0.00 0.00 CD MA Intcpt_Msg 0 0 0 0 0 0.00 0.00 CD MA_P g Ch_Ocpn 489,506 2,771 2,763 2,754 2,795 2,756.00 2,766.00 CD MA_P k Acs_ChOc 91,989 985 563 281 563 422.00 281.00 CD MA_P k P g_ChOc 555,984 3,264 3,140 3,197 3,125 3,120.00 3,155.00 CD MA_R e v P COLdur 305 0 0 0 0 0.00 0.00 CD MA_R e v P COLcnt 6 0 0 0 0 0.00 0.00 CD MA R e orde r_Msg 2 0 0 0 0 0.00 0.00 CD MA_T f CdCh_U sg 245,143 1,360 1,188 980 953 821.00 862.00 CD MA_Jmr D e t_D ur 0 0 0 0 0 0.00 0.00 November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 147
  • 148. Nortel System Data Nortel System Data Examples ExamplesNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 148
  • 149. Nortel BTSC MO Attributes Each attribute is a periodic counter maintained during the 15-minute automatic logging period. Data Seq. Access, Attribute Name Description Type Number Range 0x0002A P Number of originations blocked because BlockedOriginationsNoTCE word16 42 full no idle channel elements were available 0x0002B P Number of originations blocked due toBlockedOriginationsNoFwdCap word16 43 full lack of BTS forward link excess capacity 0x0002C P Number of originations blocked due toBlockedOriginationsNoRevCap word16 44 full lack of reverse link capacity 0x0002D P Number of handoffs blocked because no BlockedHandoffsNoTCE word16 45 full idle channel elements were available 0x0002E P Number of handoffs blocked due to lack BlockedHandoffsNoFwdCap word16 46 full of BTS forward link excess capacity 0x0002F P Number of handoffs blocked due to lack BlockedHandoffsNoRevCap word16 47 full of reverse link capaicty 0x00030 P SuccessfulOriginations word16 Number of successful originations 48 full 0x00031 P SuccessfulHandoffs word16 Number of successful handoffs 49 full November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 149
  • 150. Nortel FA MO Attributes Each attribute is a periodic counter maintained during the 15-minute automatic logging period. FA MO FA MO Sequence Sequence Number OM name Number OM name 16 TCEUtilMaximum 2D soft4softer1Alpha 17 NumOfTCsConfigured 2E soft4softer1Beta 18 soft1softer1Alpha 2F soft4softer1Gamma 19 soft1softer1Beta 30 soft4softer2AlphaBeta 1A soft1softer1Gamma 31 soft4softer2BetaGamma 1B soft1softer2AlphaBeta 32 soft4softer2GammaAlpha 1C soft1softer2BetaGamma 33 soft4softer3 1D soft1softer2GammaAlpha 34 soft5softer1Alpha 1E soft1softer3 35 soft5softer1Beta 1F soft2softer1Alpha 36 soft5softer1Gamma 20 soft2softer1Beta 37 soft5softer2AlphaBeta 21 soft2softer1Gamma 38 soft5softer2BetaGamma 22 soft2softer2AlphaBeta 39 soft5softer2GammaAlpha 23 soft2softer2BetaGamma 3A soft6softer1Alpha 24 soft2softer2GammaAlpha 3B soft6softer1Beta 25 soft2softer3 3C soft6softer1Gamma 26 soft3softer1Alpha 3D TimeNotInUse 27 soft3softer1Beta 28 soft3softer1Gamma 29 soft3softer2AlphaBeta 2A soft3softer2BetaGamma 2B soft3softer2GammaAlpha 2C soft3softer3November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 150
  • 151. Nortel BTSC MO Events Each event counter is maintained during the 15-minute automatic logging period. Type Seq. Event Report Name Description Event Report Number Includes as parameters all attributes with P 0x000?BTSCPerformanceData PerformanceData access documented in the attribute table for 0? this MO. FA MO Events Each event counter is maintained during the 15-minute automatic logging period. Type Seq. Event Report Name Description Event Report Number Includes as parameters all attributes with P 0x000? FAPerformanceData PerformanceData access documented in the attribute table for 0? this MO. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 151
  • 152. Nortel BTSC MO Report ExampleXYZ 19971120 BTSC MO Report+----+----------------------------+------+------+------+------+------+------+------+------+|BTS | Start Date/Time - |OBlock|OBlock|OBlock|HBlock|HBlock|HBlock| Succ | Succ || | End Date/Time |No TCE|No Fwd|No Rev|No TCE|No Fwd|No Rev| Origs|Handof|+----+----------------------------+------+------+------+------+------+------+------+------+| 1|1997/11/20 01:30:00-02:00:00| 0| 0| 0| 0| 0| 0| 3| 5|| 1|1997/11/20 12:00:00-12:30:00| 0| 0| 0| 0| 0| 0| 46| 314|| 1|1997/11/20 12:30:00-13:00:00| 0| 0| 0| 0| 0| 0| 76| 470|| 1|1997/11/20 13:00:00-13:30:00| 0| 0| 0| 0| 0| 0| 45| 414|| 1|1997/11/20 13:30:00-14:00:00| 0| 0| 0| 0| 0| 0| 55| 375|| 1|1997/11/20 14:00:00-14:30:00| 0| 0| 0| 0| 0| 0| 50| 525|| 1|1997/11/20 14:30:00-15:00:00| 0| 0| 0| 0| 0| 0| 72| 433|| 1|1997/11/20 15:00:00-15:30:00| 0| 0| 0| 0| 0| 0| 66| 412|| 1|1997/11/20 15:30:00-16:00:00| 0| 0| 0| 0| 0| 0| 53| 323|| 1|1997/11/20 16:00:00-16:30:00| 0| 0| 0| 0| 0| 0| 63| 342|| 1|1997/11/20 16:30:00-17:00:00| 0| 0| 0| 0| 0| 0| 51| 331|| 1|1997/11/20 17:00:00-17:30:00| 0| 0| 0| 0| 0| 0| 39| 323|| 1|1997/11/20 17:30:00-18:00:00| 0| 0| 0| 0| 0| 0| 51| 310|| 1|1997/11/20 18:00:00-18:30:00| 0| 0| 0| 0| 0| 0| 45| 237|| 1|1997/11/20 18:30:00-19:00:00| 0| 0| 0| 0| 0| 0| 31| 299|| 1|1997/11/20 19:00:00-19:30:00| 0| 0| 0| 0| 0| 0| 37| 282|| 1|1997/11/20 19:30:00-20:00:00| 0| 0| 0| 0| 0| 0| 19| 143|| 1|1997/11/20 20:00:00-20:30:00| 0| 0| 0| 0| 0| 0| 18| 96|| 1|1997/11/20 20:30:00-21:00:00| 0| 0| 0| 0| 0| 0| 33| 192|| 1|1997/11/20 21:00:00-21:30:00| 0| 0| 0| 0| 0| 0| 25| 226|| 1|1997/11/20 21:30:00-22:00:00| 0| 0| 0| 0| 0| 0| 15| 235|| 1|1997/11/20 22:00:00-22:30:00| 0| 0| 0| 0| 0| 0| 15| 216|| 1|1997/11/20 22:30:00-23:00:00| 0| 0| 0| 0| 0| 0| 9| 162|| 1|1997/11/20 23:00:00-23:30:00| 0| 0| 0| 0| 0| 0| 3| 40|| |Totals for BTS 1 | 0| 0| 0| 0| 0| 0| 1235| 8895|November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 152
  • 153. Nortel Selector Log File Example=====================================================Status : OLFLR_OKRecord Type : NEIGHBOR_LIST_TUNING_DATA_ARRAYFile Offset : 414 (octal)Time Stamp : 97/10/29-00:29:25.380Record Length : 72Header Length : 51Source Node Id : 297543 (0x00048a47)OID:AgentId : 297536 (0x00048a40)OID:MOClass : 81 (0x0051)OID:MOVersion : 1 (0x0001)OID:MOInstance : 1 (0x0001)Call Id : SID 0x4026 EntryPoint 0x134a Count 0x0 Time 0x2cfe821IMSI : NumDigits 15 Digits 134006043294814 (123-63-251-3692bf)ESN : 0x9f0d02acPFFlags : 0x1fSecondary Agent Id : 0x8a40FramingBytes : 0xfaaeSequence Number : 57AttributeType : 0x0256AttributeInstance : 0x0030Log Attr -> Type : LogSBSNeighborListTuningDataArray Seq Num : 0030LogData object contents:Data Type : NEIGHBOR_LIST_TUNING_DATA_ARRAYResource Type : OCC_SBS_RESOURCETimeStamp : 97/10/29-00:29:25.380Count : 2ExtdBaseId PowerCombineBit PilotStrength PNOffset+++++=========================++++=========================+++++0x018002a3 1 8 0x01040x018002a1 1 19 0x01a4=====================================================November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 153
  • 154. Nortel FAMO Report ExampleXYZ 19971120 FA MO Report+----+----------------------------+---------+---------+-----+-------+-------+-------+-----+---+|BTS | Start Date/Time - | MOU | MOU | CE/ | MOU | MOU | MOU |%Soft|Max|| | End Date/Time | CE | Traffic | User| Alpha | Beta | Gamma | HO |TCE|+----+----------------------------+---------+---------+-----+-------+-------+-------+-----+---+| 1|1997/11/20 07:00:00-07:30:00| 41.99| 33.35| 1.26| 11.77| 4.62| 16.96|20.58| 15|| 1|1997/11/20 07:00:00-07:30:00| 73.06| 46.22| 1.58| 17.72| 14.10| 14.39|36.75| 15|| 1|1997/11/20 08:00:00-08:30:00| 109.87| 66.05| 1.66| 24.78| 20.21| 21.06|39.88| 15|| 1|1997/11/20 10:00:00-10:30:00| 153.79| 89.81| 1.71| 41.85| 32.19| 15.77|41.60| 15|| 1|1997/11/20 10:30:00-11:00:00| 181.09| 102.19| 1.77| 43.60| 28.22| 30.38|43.57| 15|| 1|1997/11/20 11:00:00-11:30:00| 152.59| 84.73| 1.80| 37.61| 18.51| 28.61|44.47| 15|| 1|1997/11/20 11:30:00-12:00:00| 143.70| 89.16| 1.61| 39.66| 24.78| 24.72|37.95| 15|| 1|1997/11/20 12:00:00-12:30:00| 156.58| 89.52| 1.75| 25.51| 21.91| 42.10|42.83| 15|| 1|1997/11/20 12:30:00-13:00:00| 165.54| 89.97| 1.84| 44.41| 22.89| 22.67|45.65| 15|| 1|1997/11/20 13:00:00-13:30:00| 170.36| 99.19| 1.72| 52.81| 24.58| 21.79|41.78| 15|| 1|1997/11/20 13:30:00-14:00:00| 145.34| 93.71| 1.55| 41.88| 24.05| 27.77|35.53| 15|| 1|1997/11/20 14:00:00-14:30:00| 189.61| 121.49| 1.56| 52.43| 30.99| 38.06|35.93| 15|| 1|1997/11/20 14:30:00-15:00:00| 153.65| 108.08| 1.42| 47.58| 37.52| 22.99|29.65| 15|| 1|1997/11/20 15:00:00-15:30:00| 165.08| 106.66| 1.55| 49.00| 29.69| 27.97|35.39| 15|| 1|1997/11/20 15:30:00-16:00:00| 159.27| 94.72| 1.68| 42.04| 28.43| 24.25|40.53| 15|| 1|1997/11/20 16:00:00-16:30:00| 172.52| 114.62| 1.51| 56.57| 28.50| 29.55|33.56| 15|| 1|1997/11/20 16:30:00-17:00:00| 156.83| 105.46| 1.49| 53.29| 30.38| 21.80|32.76| 15|| 1|1997/11/20 17:00:00-17:30:00| 129.13| 82.52| 1.56| 31.50| 24.28| 26.73|36.10| 15|| 1|1997/11/20 17:30:00-18:00:00| 134.80| 81.76| 1.65| 35.80| 30.20| 15.77|39.35| 15|| 1|1997/11/20 18:00:00-18:30:00| 96.91| 60.49| 1.60| 27.80| 15.38| 17.31|37.58| 15|| 1|1997/11/20 18:30:00-19:00:00| 124.25| 73.62| 1.69| 22.37| 30.93| 20.33|40.75| 15|| 1|1997/11/20 19:00:00-19:30:00| 75.50| 41.14| 1.83| 18.03| 14.88| 8.24|45.50| 15|| 1|1997/11/20 19:30:00-20:00:00| 40.58| 23.56| 1.72| 12.50| 5.72| 5.33|41.95| 15|| 1|1997/11/20 20:00:00-20:30:00| 51.14| 29.81| 1.72| 13.26| 10.37| 6.19|41.71| 15|| 1|1997/11/20 20:30:00-21:00:00| 102.45| 55.26| 1.85| 16.36| 18.49| 20.41|46.07| 15|| 1|1997/11/20 21:00:00-21:30:00| 108.48| 74.86| 1.45| 28.32| 17.26| 29.27|30.99| 15|| 1|1997/11/20 21:30:00-22:00:00| 109.92| 68.50| 1.60| 26.53| 19.22| 22.75|37.68| 15|| 1|1997/11/20 22:00:00-22:30:00| 86.58| 59.36| 1.46| 26.09| 15.11| 18.15|31.45| 15|| 1|1997/11/20 22:30:00-23:00:00| 94.96| 63.48| 1.50| 27.73| 20.85| 14.90|33.15| 15|| 1|1997/11/20 23:00:00-23:30:00| 28.07| 20.76| 1.35| 9.06| 8.14| 3.55|26.04| 15|| |Totals for BTS 1 | 3690.90| 2280.64| 1.62| 980.80| 655.61| 644.22|38.21| 15| November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 154
  • 155. Motorola System Data Motorola System Data Examples ExamplesNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 155
  • 156. Motorola System Data Examples Usage OOS Orig Orig Orig Term Term Term RF RF Usa Cell MCC CE min min Atts Comps Fail% Atts Comps Fail% Loss Loss% Att ---- --- --- ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ 88 1 2 383.1 146.2 170 160 5.9 20 19 5.0 2 1.1 121.0 88 1 3 426.3 146.2 154 150 2.6 10 10 0.0 3 1.9 156.0 88 1 4 456.9 146.2 160 156 2.5 22 22 0.0 7 3.9 150.6 88 1 5 448.2 146.2 163 162 0.6 18 18 0.0 4 2.2 148.6 88 1 6 439.5 146.2 162 159 1.9 20 20 0.0 2 1.1 144.9 88 1 7 439.9 146.2 160 157 1.9 14 14 0.0 5 2.9 151.7 88 1 8 351.6 146.2 186 182 2.2 23 23 0.0 5 2.4 100.9 88 1 9 397.4 146.2 164 161 1.8 20 20 0.0 3 1.7 129.6 88 1 10 422.5 146.2 177 174 1.7 15 15 0.0 2 1.1 132.0 88 1 11 402.2 146.2 183 179 2.2 22 22 0.0 1 0.5 117.7 88 1 12 398.2 146.2 179 176 1.7 13 13 0.0 5 2.6 124.4 88 1 13 447.5 146.2 163 161 1.2 26 26 0.0 11 5.9 142.1 88 1 14 263.5 146.2 290 83 71.4 31 19 38.7 5 4.9 49.3 88 1 15 307.8 146.2 264 68 74.2 36 9 75.0 3 3.9 61.5 88 2 2 403.1 105.9 165 162 1.8 14 14 0.0 1 0.6 135.1 88 2 3 477.0 105.9 163 158 3.1 18 18 0.0 3 1.7 158.1 88 2 4 419.4 105.8 166 161 3.0 24 24 0.0 2 1.1 132.4 88 2 5 445.8 105.8 174 171 1.7 14 14 0.0 7 3.8 142.3 88 2 6 525.1 105.8 157 155 1.3 17 17 0.0 3 1.7 181.1 88 2 7 422.0 105.8 165 161 2.4 18 17 5.6 1 0.6 138.4 88 2 8 430.3 105.8 188 183 2.7 14 14 0.0 7 3.6 127.8 88 2 9 419.9 105.8 167 166 0.6 12 11 8.3 6 3.4 140.7 88 2 10 391.0 105.3 165 164 0.6 22 22 0.0 4 2.2 125.5 88 2 11 443.5 105.3 174 168 3.4 11 11 0.0 5 2.8 143.8 88 2 12 412.5 105.3 177 171 3.4 21 21 0.0 4 2.1 125.0 88 2 13 394.2 105.3 196 192 2.0 16 16 0.0 6 2.9 111.6 88 2 14 432.0 105.3 141 139 1.4 18 18 0.0 5 3.2 163.0 88 2 15 388.5 105.3 178 176 1.1 17 17 0.0 2 1.0 119.5November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 156
  • 157. Metrica Examples Metrica ExamplesNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 157
  • 158. Metrica: Forward Power Overload Reports DAILY SECTOR POWER OVERLOAD REPORT ON 10/26/98 FOR ALL SECTORS IN REGION (DAILY TOTALS) Sector RPwrOvldDur RPwrOvlds FPwrOvldDur FPwrOvlds Id. (min) Incidents (min) Incidents ------------------------------------------------------------------------------------------------- VX2-ECP:2-CELL:3-SECT:3 0.85 1 0.03 1 VX2-ECP:2-CELL:4-SECT:1 0.00 0 0.00 0 VX2-ECP:2-CELL:4-SECT:2 0.00 0 0.35 15 VX2-ECP:2-CELL:4-SECT:3 0.00 0 0.62 13 VX2-ECP:2-CELL:5-SECT:1 0.00 0 0.00 0 VX2-ECP:2-CELL:5-SECT:2 0.00 0 0.02 1 VX2-ECP:2-CELL:6-SECT:2 0.00 0 0.00 2 VX2-ECP:2-CELL:7-SECT:2 0.00 0 0.33 8 VX2-ECP:2-CELL:7-SECT:3 0.00 0 0.60 18 VX2-ECP:2-CELL:11-SECT:3 0.00 0 0.02 1 VX2-ECP:2-CELL:12-SECT:1 0.67 1 0.00 0 VX2-ECP:2-CELL:12-SECT:2 0.52 1 0.05 1 VX2-ECP:2-CELL:14-SECT:1 0.00 0 0.00 1 VX2-ECP:2-CELL:15-SECT:3 1.02 1 0.02 1 VX2-ECP:2-CELL:16-SECT:1 0.00 0 0.28 8 VX2-ECP:2-CELL:16-SECT:2 0.00 0 0.00 0 VX2-ECP:2-CELL:16-SECT:3 0.00 0 0.00 0November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 158
  • 159. Metrica Data Examples Although Metrica is the preferred tool of some PCS operators for performance analysis across all network platforms, it is more useful in systems of some manufacturers and less useful in others (see external examples)November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 159
  • 160. Metrica: Switch Traffic Statistics DAILY MSC SWITCHED TRAFFIC REPORT ON 10/26/98 (DAILY NORMALISED TOTALS/AVERAGES) -------- Switched ------ MSC Calls Failures Failure Data Id. (%) (%) ------------------------------------------ VX1-ECP:1 158556 10637 6.71 100 MSC Traffic Calls Failures Failure Data Id. Type (%) (%) -------------------------------------------------------------------------------------------- VX1-ECP:1 Mobile-to-Mobile Originating (Lucent) 111 2 1.80 100 VX1-ECP:1 Mobile-to-Mobile Terminating (Lucent) 2157 913 42.33 100 VX1-ECP:1 Mobile Originating 128034 5772 4.51 100 VX1-ECP:1 Roamer Mobile Originating (Lucent) 104015 155 0.15 100 VX1-ECP:1 Roamer Mobile Terminating (Lucent) 27197 8 0.03 100 VX1-ECP:1 Mobile Terminating 30522 4865 15.94 100November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 160
  • 161. Metrica: Traffic Engineering Counts Traffic Engineering Report Run Date: 10/27/98 Time(GMT-0): 21:02 DAILY CELL TRAFFIC REPORT ON 10/26/98 FOR ALL CELLS IN REGION (DAILY NORMALIZED TOTALS/AVERAGES)Sorted by Traffic (E) RF ------- Traffic ------- ResvCell Channels Blk RF Loss Channel Soft Code HO HO Pwr TCC Max Fwd Rev DataID Def Avail Atts Blks (%) Seizs Loss (%) (E) (E) (E) Blks Blks Dur Fail Busy Ovld Ovld (%)-------------------------------------------------------------------------------------------------------------------------------VX2020053 42 42.0 8605 0 0.0 8558 106 1.2 297.3 102.4 394.8 2 0 71 286 39 33 1 100VX2020050 40 40.0 6159 0 0.0 6120 72 1.2 228.4 82.0 289.4 0 0 202 171 31 80 0 100VX2020057 42 42.0 5082 0 0.0 5037 52 1.0 215.6 97.8 267.6 1 0 199 185 31 99 0 100VX2020062 42 42.0 4755 0 0.0 4716 54 1.1 207.2 91.9 257.2 1 0 72 150 27 29 0 100VX2020011 28 28.0 3685 0 0.0 3670 38 1.0 160.8 69.5 201.3 3 0 1 75 27 1 0 100VX2020112 28 28.0 2512 0 0.0 2496 20 0.8 158.4 92.8 181.6 0 0 5 81 23 2 0 100VX2020004 42 42.0 3845 0 0.0 3822 32 0.8 154.0 47.1 208.0 1 0 58 60 26 28 0 100VX2020073 30 30.0 4182 0 0.0 4137 26 0.6 152.3 55.6 192.4 0 0 4 91 24 2 0 100VX2020078 28 28.0 3429 0 0.0 3379 26 0.8 151.1 68.7 179.3 1 0 148 65 25 3 2 100VX2020060 28 28.0 2978 0 0.0 2965 28 0.9 150.4 69.5 190.4 2 0 0 93 24 0 0 100VX2020051 42 42.0 3193 0 0.0 3174 24 0.8 144.3 66.6 172.8 0 0 0 58 24 0 0 100VX2020023 54 54.0 3330 0 0.0 3307 27 0.8 139.6 43.8 197.8 0 0 4 67 23 4 0 100VX2020007 28 28.0 3520 0 0.0 3502 28 0.8 131.6 47.9 165.9 1 0 97 74 26 26 1 100VX2020017 30 30.0 2902 0 0.0 2889 24 0.8 130.0 45.9 181.7 1 0 51 70 23 0 1 100VX2020076 28 28.0 2888 0 0.0 2867 25 0.9 129.6 54.8 174.5 0 0 312 74 22 4 1 100VX2020012 28 28.0 3313 0 0.0 3297 21 0.6 122.1 37.8 166.2 1 0 74 61 23 1 2 100VX2020016 28 28.0 3049 0 0.0 3027 18 0.6 120.8 46.5 162.1 2 0 17 44 22 8 0 100VX2020067 28 28.0 2537 0 0.0 2519 30 1.2 118.3 51.2 145.9 0 0 0 46 19 0 0 100VX2020001 28 28.0 2746 0 0.0 2737 25 0.9 114.9 42.2 150.6 0 0 0 37 22 0 0 100VX2020003 28 28.0 2031 0 0.0 2019 19 0.9 113.4 54.6 153.7 0 0 53 41 20 1 1 100 November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 161
  • 162. Metrica: Daily Sector Traffic Reports Traffic Engineering Report Run Date: 10/27/98 Time(GMT-0): 21:02 DAILY SECTOR TRAFFIC REPORT ON 10/26/98 FOR ALL SECTORS IN REGION DropCity_2 (BUSY HOUR VALUES)Sorted by Code Traffic (E)Note: Busy Hour is in local timeSector Busy ----- Attempts ------ RFLoss HO Cd Traff OverloadName Hour Total Orig Term Seizs RFLoss (%) Seizs (E) Duration-----------------------------------------------------------------------------------------------------------------------VX2-ECP:2-CELL:53-SECT:3 17:00 372 312 60 369 6 1.63 3325 12.62 8VX2-ECP:2-CELL:57-SECT:1 17:00 183 158 25 180 2 1.11 2464 10.07 51VX2-ECP:2-CELL:53-SECT:1 17:00 251 205 46 251 6 2.39 2336 9.57 4VX2-ECP:2-CELL:62-SECT:1 18:00 193 144 49 193 5 2.59 2507 9.32 8VX2-ECP:2-CELL:50-SECT:3 17:00 289 243 46 284 4 1.41 2042 8.82 74VX2-ECP:2-CELL:4-SECT:2 16:00 165 131 34 164 4 2.44 2203 8.67 15VX2-ECP:2-CELL:4-SECT:3 16:00 120 86 34 118 1 0.85 2653 8.31 36VX2-ECP:2-CELL:50-SECT:1 17:00 184 148 36 183 0 0.00 2174 8.23 2VX2-ECP:2-CELL:23-SECT:3 11:00 194 140 54 193 1 0.52 2282 8.11 0VX2-ECP:2-CELL:53-SECT:2 18:00 143 111 32 143 5 3.50 2060 8.02 2VX2-ECP:2-CELL:78-SECT:3 15:00 133 99 34 133 3 2.26 2198 7.94 1VX2-ECP:2-CELL:17-SECT:3 15:00 201 143 58 200 1 0.50 1853 7.75 0VX2-ECP:2-CELL:12-SECT:2 16:00 144 91 53 144 2 1.39 1662 7.29 3VX2-ECP:2-CELL:7-SECT:3 17:00 191 133 58 188 0 0.00 2375 7.21 28VX2-ECP:2-CELL:3-SECT:3 15:00 144 99 45 143 1 0.70 2144 7.07 2VX2-ECP:2-CELL:57-SECT:2 20:00 155 116 39 151 1 0.66 1884 7.01 2VX2-ECP:2-CELL:11-SECT:1 17:00 132 98 34 132 2 1.52 1800 6.84 0 November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 162
  • 163. Metrica: Forward Power Overload Reports DAILY SECTOR POWER OVERLOAD REPORT ON 10/26/98 FOR ALL SECTORS IN REGION (DAILY TOTALS) Sector RPwrOvldDur RPwrOvlds FPwrOvldDur FPwrOvlds Id. (min) Incidents (min) Incidents ------------------------------------------------------------------------------------------------- VX2-ECP:2-CELL:3-SECT:3 0.85 1 0.03 1 VX2-ECP:2-CELL:4-SECT:1 0.00 0 0.00 0 VX2-ECP:2-CELL:4-SECT:2 0.00 0 0.35 15 VX2-ECP:2-CELL:4-SECT:3 0.00 0 0.62 13 VX2-ECP:2-CELL:5-SECT:1 0.00 0 0.00 0 VX2-ECP:2-CELL:5-SECT:2 0.00 0 0.02 1 VX2-ECP:2-CELL:6-SECT:2 0.00 0 0.00 2 VX2-ECP:2-CELL:7-SECT:2 0.00 0 0.33 8 VX2-ECP:2-CELL:7-SECT:3 0.00 0 0.60 18 VX2-ECP:2-CELL:11-SECT:3 0.00 0 0.02 1 VX2-ECP:2-CELL:12-SECT:1 0.67 1 0.00 0 VX2-ECP:2-CELL:12-SECT:2 0.52 1 0.05 1 VX2-ECP:2-CELL:14-SECT:1 0.00 0 0.00 1 VX2-ECP:2-CELL:15-SECT:3 1.02 1 0.02 1 VX2-ECP:2-CELL:16-SECT:1 0.00 0 0.28 8 VX2-ECP:2-CELL:16-SECT:2 0.00 0 0.00 0 VX2-ECP:2-CELL:16-SECT:3 0.00 0 0.00 0November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 163
  • 164. Metrica: RF Overload Blocking Estimation DAILY ESTIMATION OF BLOCKING DUE TO RF OVERLOAD ON 10/26/98 FOR ALL CELLS IN REGION (DAILY TOTALS) Definitions: ----------- - Percentage of total attempts (orig and term) that are terminations (L-M calls) %Term = CDMA_PAF3 / (CDMA_PAF2 + CDMA_PAF3) - Blocked terminations due to equipment blocks (no CEs avail or packet pipe blocked) TerCS7 = CDMA_CS7 * %Term - Blocked originations due to equipment blocks (no CEs avail or packet pipe blocked) OrgCS7 = CDMA_CS7 - TerCS7 - Blocked originations due to RF power overload OrgPwrBlk = SUM(CDMA_PAF26) - OrgCS7 - Total blocked originations/terminations due to RF power overload TotPwrBlk = (OrgPwrBlk * %Term) + OrgPwrBlk - Percentage of call attempts blocked at the cell due to RF power overload Pwr%Blk = TotPwrBlk / (SUM(CDMA_PAF2) + SUM(CDMA_PAF3) - CDMA_CP8 - SUM(CDMA_PAF25)) * 100 - Percentage of call attempts due to equipment blks/TCCF/RF power overload Tot%Blk = (TotPwrBlk + CS7 + TotTCC) / (SUM(CDMA_PAF2) + SUM(CDMA_PAF3) - CDMA_CP8 - SUM(CDMA_PAF25)) * 100November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 164
  • 165. Metrica: RF Overload Blocking Indications DAILY ESTIMATION OF BLOCKING DUE TO RF OVERLOAD ON 10/26/98 FOR ALL CELLS IN REGION (DAILY TOTALS) Org Tot Cell Rev Rev Fwd Fwd Org Ter Tot Re- Org Ter Tot Ter Org Pwr Pwr Pwr Tot Id. Pwr Dur Pwr Dur Att Att Att %Term Ord TCC TCC TCC cs7 cs7 cs7 Blk Blk %Blk %Blk paf25 cp8 ----------------------------------------------------------------------------------------------------------- VX2020001 0 0 0 0 1869 877 2746 31.9 1 26 11 37 0 0 0 1 1 0.0 1.4 0 0 VX2020002 0 0 0 0 809 373 1182 31.6 2 18 2 20 0 0 0 2 3 0.2 1.9 0 0 VX2020003 1 1 1 0 1428 603 2031 29.7 4 27 14 41 0 0 0 4 5 0.3 2.3 0 0 VX2020004 0 0 28 1 2780 1065 3845 27.7 7 40 20 60 0 0 0 7 9 0.2 1.8 0 2 VX2020005 0 0 1 0 544 182 726 25.1 0 6 0 6 0 0 0 0 0 0.0 0.8 0 0 VX2020006 0 0 2 0 731 339 1070 31.7 0 8 7 15 0 0 0 0 0 0.0 1.4 0 0 VX2020007 1 1 26 1 2493 1027 3520 29.2 5 59 15 74 0 0 0 5 6 0.2 2.3 0 1 VX2020008 0 0 0 0 2056 998 3054 32.7 0 23 19 42 0 0 0 0 0 0.0 1.4 0 0 VX2020009 0 0 0 0 626 212 838 25.3 3 10 2 12 0 0 0 3 4 0.4 1.9 0 0 VX2020010 0 0 0 0 877 349 1226 28.5 1 31 3 34 0 0 0 1 1 0.1 2.9 0 0 VX2020011 0 0 1 0 2577 1108 3685 30.1 1 60 15 75 0 0 0 1 1 0.0 2.1 0 0 VX2020012 2 2 1 0 2364 949 3313 28.6 4 45 16 61 0 0 0 4 5 0.2 2.0 0 0 VX2020014 0 0 1 0 670 245 915 26.8 3 11 4 15 0 0 0 3 4 0.4 2.1 0 0 VX2020015 1 1 1 0 1275 596 1871 31.9 4 24 9 33 0 0 0 4 5 0.3 2.0 0 0 VX2020016 0 0 8 0 2156 893 3049 29.3 4 36 8 44 0 0 0 4 5 0.2 1.6 0 0 VX2020017 1 1 0 0 2091 811 2902 27.9 1 54 16 70 0 0 0 1 1 0.0 2.4 0 0 VX2020018 0 0 0 0 950 393 1343 29.3 1 28 1 29 0 0 0 1 1 0.1 2.2 0 0 VX2020019 0 0 1 0 1254 513 1767 29.0 0 32 9 41 0 0 0 0 0 0.0 2.3 0 0November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 165
  • 166. Analyzing System Data Analyzing System DataNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 166
  • 167. Total Blocked Call Percentage Example Percent Total Block Call Percentage 8.0% 7.5% Blkd 7.0% 6.5% 6.0% 5.5% 5.0% 4.5% 4.0% 3.5% 3.0% 2.5% 2.0% 1.5% 1.0% Date This is an example of a cumulative system-wide total blocked call percentage chart maintained in one marketNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 167
  • 168. Dropped Call Percentage Tracking Example Percent Total Drop Call Percentage 5.0% 4.5% %Drops 4.0% 3.5% 3.0% 2.5% 2.0% 1.5% 1.0% 0.5% 0.0% Date Dropped call percentage tracking by one marketNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 168
  • 169. Total System Daily MOU Example Daily Total System MOU MOU 300000 Daily Total System MOU 250000 200000 150000 100000 50000 0 Date Total system daily MOU plotted by one marketNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 169
  • 170. “Top Ten” Performance Tracking ExampleCall AttemptsEng MSC Call %Call Block %Blck Acc %Acc Drop %DropSite Site Call Att Succ Succ Calls Calls Fail Fail Calls Calls Call Attempts6.1 13X 2561 2234 87.2 130 5.1 130 5.1 145 5.7 30002.1 2X 2244 2017 89.9 101 4.5 101 4.5 93 4.1 25001.2 1Y 1922 1743 90.7 83 4.3 83 4.3 66 3.4 200064.3 93Z 1833 1549 84.5 137 7.5 136 7.4 110 6.0108.2 30Y 1740 1589 91.3 46 2.6 45 2.6 83 4.8 15001.3 1Z 1630 1495 91.7 31 1.9 31 1.9 81 5.0 100063.2 57Y 1623 1486 91.6 49 3.0 49 3.0 66 4.1 500102.2 4Y 1615 1495 92.6 18 1.1 18 1.1 70 4.3 Calls 0 108.2 102.2 108.1 64.3 63.2 43.3 6.1 2.1 1.2 1.3108.1 30X 1490 1387 93.1 27 1.8 27 1.8 54 3.6 Sector43.3 42Z 1488 1410 94.8 4 0.3 4 0.3 53 3.6% Blocked Calls September 5, 1997Eng MSC Call %Call Block %Blck Acc %Acc Drop %DropSite Site Call Att Succ Succ Calls Calls Fail Fail Calls Calls % Blocked Calls64.3 93Z 1833 1549 84.5 137 7.5 136 7.4 110 6.0 8.06.1 13X 2561 2234 87.2 130 5.1 130 5.1 145 5.7 7.063.3 57Z 1282 1098 85.7 65 5.1 65 5.1 90 7.0 6.02.1 2X 2244 2017 89.9 101 4.5 101 4.5 93 4.1 5.0 4.01.2 1Y 1922 1743 90.7 83 4.3 83 4.3 66 3.4 3.063.2 57Y 1623 1486 91.6 49 3.0 49 3.0 66 4.1 2.064.1 93X 1027 926 90.2 30 2.9 30 2.9 58 5.7 1.026.3 35Z 855 698 81.6 24 2.8 24 2.8 112 13.1 0.0 % 108.2 64.3 63.3 63.2 64.1 26.3 6.1 2.1 1.2 1.3108.2 30Y 1740 1589 91.3 46 2.6 45 2.6 83 4.8 Sector1.3 1Z 1630 1495 91.7 31 1.9 31 1.9 81 5.0 Many markets use scripts or spreadsheet macros to produce ranked lists of sites with heavy traffic, performance problems, etc.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 170
  • 171. “Bracketing”: Fault Notification and Alarming Historic Performance Data and Automatic Alarming 0 Some operators develop their 1 2 3 own software for monitoring 4 5 6 and tracking performance data 7 8 9 10 11 Each new 30-minute period is 12 13 14 compared against a six-week 15 16 17 average for that day and time 18 19 20 21 22 If the new value is outside 23 SM TWT F S SM TWT F S SM TWT F S SM TWT F S SM TWT F S SM TWT F S SM TWT F S user-selectable tolerances (typically +/- 30%), an alarm is TOO LOW NORMAL TOO HIGH sent to operations personnel • By SMS or pager +30% +30% +30% The tolerance values can be 6-week average adjusted to produce -30% -30% -30% reasonable numbers of alarms • Typically 20-40 alarms per If an important performance statistic varies day outside a user-specified range, an alarm message is sent automatically to the performance specialist responsible for that base station.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 171
  • 172. CDMA System Parameters CDMA System ParametersNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 172
  • 173. Lucent System Parameters Lucent System ParametersNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 173
  • 174. Lucent BTS Parameters ExampleSysID 179 179 179 179 179 179 179 179 179 179 179 179 179 179 179 179 179 179 179 179 179ECPID 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2CellID 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2Antenna 1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 3 3 1 1 1CDMAPilotPN 4 4 4 4 4 4 172 172 172 172 172 172 340 340 340 340 340 340 8 8 8CDMAPilotDrpThrsh -15 -15 -15 -15 -15 -15 -15 -15 -15 -15 -15 -15 -15 -15 -15 -15 -15 -15 -15 -15 -15CDMAPilotDetThrsh -13 -13 -13 -13 -13 -13 -13 -13 -13 -13 -13 -13 -13 -13 -13 -13 -13 -13 -13 -13 -13CDMACompThrsh 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2 2 2CDMADropTimer 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3CDMASrchWinActCand 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7CDMASrchWinNbr 9 9 9 9 9 9 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7CDMASrchWinRemain 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0CDMAPilotGain 108 108 108 108 108 108 108 108 108 108 108 108 108 108 108 108 108 108 108 108 108CDMAPageGain 64 64 64 64 64 64 64 64 64 64 64 64 64 64 64 64 64 64 64 64 64CDMASyncGain 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34CDMABCRAtt 6 6 6 6 6 6 6 6 6 6 6 6 8 8 8 8 8 8 8 8 8SectorSize_ceqfaceBBAMaxPower 33.5 33.5 21 21 33.5 33.5 33.5 33.5 21 21 33.5 33.5 33.5 33.5 21 21 33.5 33.5 25 25 25CDMAMinTrfChnlGain_R2CDMAMaxTrfChnlGain_R2CDMATrafGain_R2CDMAFwdFrmErrRate_R2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1CDMARevFrmErrRate_R2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1CDMANomEbNoSetPt_R2 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8CDMAMinEbNoSetPt_R2 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8CDMAMaxEbNoSetPt_R2 8.8 8.8 8.8 8.8 8.8 8.8 8.8 8.8 8.8 8.8 8.8 8.8 8.8 8.8 8.8 8.8 8.8 8.8 8.8 8.8 8.8Srchwincell 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 174
  • 175. Nortel System Parameters Nortel System ParametersNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 175
  • 176. Nortel System Parameters ExampleProto type datafill for 1900 CDMA System ParametersParameter Name Range Recommended Value Remarks1. CDMA Channel ParametersSystem Determination and Acquisition CDMA_ AVAIL 0-1 1 CDMA_FREQ (CDMA_CHAN) 0 - 2047 See Remarks As determined by the local MTA BAND_CLASS 0 - 31 1 1900 MHzSystem Acquisition (Sync channel Information) P_REV 0-255 1 MIN_P_REV 0-255 1 SID 0 - 32,767 See Remarks As determined by the local MTA NID 0 - 65,535 See Remarks As determined by the local MTA PILOT_PN 0 - 511 See Remarks As determined by the local MTA LC_STATE See Remarks Determined by the system. TBA SYS_TIME TFU_1 or TFU_2 See Remarks As detetermined by the System time PRAT 0-3 1 LP_SEC 0-255 13 TBA LTM_OFF 0-63 16 TBA DAYLT 0-1 0 or 1 Depending on whether Daylight saving is On/Off Nortel parameters are built in files on the BSM, then downloaded to BTS and SBS locations November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 176
  • 177. Nortel System Parameters Example2. Access ParametersRequest Response Parameters PSIST(0-9) 0 - 63 0 ACCOLC(0 -9) are all permitted to transmit PSIST(10-15) 0-7 0 ACCOLC(10 -15) are all permitted to transmit MAX_CAP_SZ 0-7 3 3 Frames message PAM_SZ 0 - 15 4 4 Frames preamble REG_PSIST 0-7 0 MSG_PSIST 0-7 0 PROBE_PN_RAN 0 - 15 0 ACC_CHAN 0 - 31 0 1 Access channel ACC_TMO 0 - 15 (x80 ms) 3 (2+1), 240 ms PROBE_BKOFF 0 - 15 0 (0 + 1) slot delay BKOFF 0 - 15 1 (0 + 1) slot delay MAX_REQ_SEQ 0 - 15 2 MAX_RSP_SEQ 0 - 15 2 AUTH 0-3 0 No standard Authentication RAND 0-(232-1) 0 Not applicable without Authentication Parameters here determine contents of Access Parameters Message on the Paging Channel November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 177
  • 178. Nortel System Parameters ExampleRegistaration Parameters SID 0 - 32,767 See Remarks As determined by the local MTA NID 0 - 65,535 See Remarks As determined by the local MTA REG_ZONE 0 - 4095 As determined by the network Zone Registration not currently supported TOTAL_ZONES 0-7 0 Zone Registration not currently supported ZONE_TIMER 0-7 0 Zone Registration not currently supported MULTI_SIDS 0-1 0 If roaming is permitted, this should be set to 1 MULTI_NIDS 0-1 0 If roaming or more than one NID in the MTA, set to 1 BASE_ID 0 - 65,535 See Remarks As determined by the local MTA BASE_CLASS 0 - 15 0 Public macro cellular system PAGE_CHAN 0-7 1 One paging channel MAX_SLOT_CYCLE_INDEX 0-7 5 HOME_REG 0-1 1 FOR_SID_REG 0-1 1 FOR_NID_NEG 0-1 1 POWER_UP_REG 0-1 1 POWER_DOWN_REG 0-1 1 PARAMETER_REG 0-1 1 REG_PRD 0 - 127 0 Periodic registration every 2621 sec (43 min) BASE_LAT -1296000, +1296000 See Remarks As determined by the local MTA BASE_LONG -2592000, +2592000 See Remarks As determined by the local MTA REG_DIST 0 No distance based registration RESCAN 0-2047 0 Parameters here determine the contents of the registration fields of the System Parameters Message on the paging channel November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 178
  • 179. Nortel System Parameters Example3. Power Control ParametersOpen Loop NOM_PWR 0 -15 8 8 = 0 dB INIT_PWR 0 - 31 16 16 = 0 dB PWR_STEP 0-7 3 3 dB NUM_STEP 0 - 15 6 (6 +1) access probes per sequenceForward Power Control PWR_REP_THRESH 0 - 31 2 Only applicable to RateSet1 (8 kbps) data PWR_REP_FRAMES 0 - 15 7 Only applicable to RateSet1 (8 kbps) data PWR_THRESH_ENABLE 0-1 1 Only applicable to RateSet1 (8 kbps) data POWER_PERIOD_ENABLE 0 -1 0 Only applicable to RateSet1 (8 kbps) data POWER_REP_DELAY 0 - 31 1 4 frames, only applicable to RateSet1 (8 kbps) data4. Handoff ParametersPilot Search Parameters PILOT_PN 0-1 1 As determined by the local MTA SEARCH_WIN_A 0 - 15(4 - 452 PN Chps) 8 60 PN chips SEARCH_WIN_N 0 - 15(4 - 452 PN Chps) 10 100 PN chips SEARCH_WIN_R 0 - 15(4 - 452 PN Chps) 10 100 PN chips NGHBR_MAX_AGE 0 - 15 2 PILOT_INC 0 - 15 4 NGHBR_CONFIG 0-7 0Pilot Strength Parameters T_ADD 0 - 63(-0.5x dB) 28 -14 dB T_DROP 0 - 63(-0.5x dB) 32 -16 dB T_TDROP 0 - 15 (=< 0.1 - 319 sec) 3 4 sec T_COMP 0 - 15 (x0.5 dB) 5 2.5 dB These parameters are communicated to the mobile in the overhead messages on the Paging Channel. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 179
  • 180. Nortel System Parameters Example NMIS Parameter Range Recommended Value RemarksAcquisition AccessChannelAcquisitionSearchWidth 25 - 4095 TBA Used by the BTS for the revese link AccessChannelDemodulationSearchWidth 25 - 4095 TBA Used by the BTS for the revese link TrafficChannelAcquisitionSearchWidth 25 - 4095 TBA Used by the BTS for the revese link TrafficChannelDemodulationSearchWidth 25 - 4095 TBA Used by the BTS for the revese linkPowerControlRateSet1Data, RateSet2Data PrRXerror (FER %) Full 1/16 - 256/16 16/16 1% Half 1/16 - 256/16 80/16 5% Quarter 1/16 - 256/16 80/16 5% Eighth 1/16 - 256/16 80/16 5% Unknown 1/16 - 256/16 16/16 1% RRXincrease Full 1/256 - 4095/256 42/256 Half 1/256 - 4095/256 7/256 Quarter 1/256 - 4095/256 7/256 Eighth 1/256 - 4095/256 7/256 Unknown 1/256 - 4095/256 14/256RateSet1Data PRXlower (Ew/Nt) 1/256 - 4095/256 2048/256 (8 - 10log2) = 5 dB Eb/Nt PRXupper (Ew/Nt) 1/256 - 4095/256 3328/256 (11 - 10log2) = 8 dB Eb/Nt PRXstart (Ew/Nt) 1/256 - 4095/256 2688/256 (10.5 - 10log2) = 7.5 dB Eb/NtRateSet2Data PRXlower (Ew/Nt) 1/256 - 4095/256 2509/256 (10 - 10log3) = 5.2 dB Eb/Nt PRXupper (Ew/Nt) 1/256 - 4095/256 3789/256 (13 - 10log3) = 8.2 dB Eb/Nt PRXstart (Ew/Nt) 1/256 - 4095/256 3149/256 (12.5 - 10log3) = 7.7 dB Eb/NtRateSet1Data PrTXerror 1/16 - 256/16 16 1% RTXincrease 1/256 - 4095/256 20/256 PTXlower -4095/256 - 0/256 -2304/256 -9 dB PTXupper -4095/256 - 0/256 -768/256 -3 dB PTXstart -4095/256 - 0/256 -1536/256 -6 dBRateSet2Data PrTXerror 1/16 - 256/16 16 1% RTXincrease 1/256 - 4095/256 133/256 PTXlower -4095/256 - 0/256 -3072/256 -12 dB PTXupper -4095/256 - 0/256 -256/256 -1 dB PTXstart -4095/256 - 0/256 -1536/256 -6 dB PowerControlGainOffset -127 to 128 0 November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 180
  • 181. Nortel System Parameters ExampleWilting, Blossoming and Breathing Parameters WiltBlossStepSize 0/16 - 255/16 dB (0-255) 4/16 dB (4) Rate should be 1 dB/sec WiltBlossStepTime 1 - 20 (units of 100 ms) 1 100 msec WiltBlossEnabled 0-1 1 BreathingStepSize 1/16 - 255/16 dB/ms 4/16 dB/ms Rate should be 1 dB/sec BreathingStepTime 1 - 20 (units of 100 ms) 1 100 msec BreathingDelta 0/16 - 255/16 dB (0-255) 192/16 (4 dB [64] or more) 12 dB rise over the noise floor BreathingEnabled 0-1 0 Disabled RecPowerEstimationFilterRate 2 - 40 (units of 5 ms) 4 Steps of 5 ms. 4 =20 ms RecPowerDecayExponential 0 - 16 6 TXAttenNormal 0-70 (0/16 - 1120/16 dB) See Remarks As given by the installtion & calibration TXPowerMax 384/16 - 736/16 dBm 672/16 42 dBm(16 W) TXAttenAntenna 0-6 (0/16 - 96/16 dB) 0 As determined by Pilot Channel calibration process. As measure TPEFilterDecayExponential 0 - 16 5 ReverseLinkCapacityEstimationRate 20*5 to 255*5 ms 100 (20*5) Not supported HandoffBlockingThreshold 0-100 % 5? Not supported CallBlockingThreshold 0-100 % 10? Not supported RXFEGain RcvrA 0/16 - 480/16 dB See Remarks As given by the installtion & calibration RCVRB 0/16 - 480/16 dB See Remarks As given by the installtion & calibration RXFENoiseFigure RCVRA 0/16 - 160/16 dB See Remarks As given by the installtion & calibration RCVRB 0/16 - 160/16 dB See Remarks As given by the installtion & calibration RXCableAtten According to above cable loss of antenna path for the specific ap RcvrA 0/16 - 480/16 dB See Remarks As given by the installtion & calibration RCVRB 0/16 - 480/16 dB See Remarks As given by the installtion & calibration RXCableNoiseFigure Close to RxCableAtten According to noise figure introd due to cable loss of antenna pat RCVRA 0/16 - 160/16 dB See Remarks As given by the installtion & calibration RCVRB 0/16 - 160/16 dB See Remarks As given by the installtion & calibration RXCardNoiseFigureMin 0/16 - 1120/16 5 dB for 800, 4 dB for 1900 As given by the Rx card calibration RXCardNoiseFigureMax 0/16 - 1120/16 960/16 60 dB Wilting and blossoming are techniques for gracefully taking a sector from service or returning it to service without dropping traffic. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 181
  • 182. Nortel System Parameters ExamplePilot Data BasePilotChannel CDMACenterFrequency ?? See Remarks As determined by the Preferred Channel Set ExtendedBaseId word32 See Remarks BandClass, CDMAFreq,BASE_ID,Sector Available 0 -1 1 QuickRepeat 0 -1 0 disabled BlankAndBurst 0 -1 0 Not used ForwardGain 0 - 255 TBA PilotGain 0 - 255 216 216 for 800 MHz MinPilotToTotalPwrRatio -255/16 to 0/16 dB -7 20% of HPA power NeighborList word32Array, up to 20 nieghb See Remarks As determined by the RF design CellType CELL_STANDARD, CELL_P CELL_STANDARD If no HHO in the cell CELL_BORDERSyncChannel SyncGain 0 - 255 68 10 dB down from pilot for 800 MHzPagingChannel PagingGain 0 - 255 130 4.4 dB down from pilot for 800 MHz These parameters set the power levels of the overhead channels. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 182
  • 183. Motorola System Motorola System Parameters ParametersNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 183
  • 184. Motorola System ParametersList all BTSs for MM1 MM2.System: BRDVILBOCM0OMC1 Prev BTS <-Id: OMC1 MM2 BTS88 Next BTS ->Name: 265A_S2_Location: 41.58.39, -87.54.14Band: 1900 MHz PilotInc: 3Carriers: 1-375SiteConf: 120°-SC600Service Option: 13KVoiceRelease: Apr 28 13:18 2.8.1.21.22Exported on: Mon Jun 14 00:00:17 CDT 1999November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 184
  • 185. Motorola System Parameters Motorola customers should obtain the proprietary Motorola document, “CDMA RF Application Note: Parameters and Optimization”, draft version 8.1 or later, available from your Motorola representative. This document gives descriptions of most system parameters and many operational peg counts and valuable guidance for setting parameters.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 185
  • 186. Motorola System Parameters Forward Pwr Ctrl Sector 1 Sector 2 Sector 3 Default C1 C1 C1 PilotPn 66 237 408 0 PilotGain 127 127 127 127 SchGain 40 40 40 40 PchGain 110 110 110 110 SifPilotPwr 31 31 31 31 dBm MinPcbGain 20 20 20 20 PcbGainFact 0.75 0.75 0.75 0.75 FwdPwrThresh 2 2 2 2 Frm PwrThreshEna 1 1 1 1 PwrPeriodEna 0 0 0 0 PwrRepThresh 3 3 3 3 Frm PwrRepFrames 7 7 7 7 Frm PwrRepDelay 12 12 12 12 FrmNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 186
  • 187. Motorola System Parameters Reverse Pwr Ctrl Sector 1 Sector 2 Sector 3 Default C1 C1 C1 NomPwr 3 3 3 3 dB InitPwr -3 -3 -3 -3 dB PwrStep 5 5 5 5 dB NumStep 4 4 4 4 RPCMaxEbNo 12.5 12.5 12.5 12.5 dB RPCNomEbNo 9 9 9 9 dB RPCMinEbNo 6 6 6 6 dB RPCThrshNom 1930 1930 1930 0 Cell Size Sector 1 Sector 2 Sector 3 Default C1 C1 C1 TchacqWinSz 125 125 125 125 chp TchmpthWinSz 25 25 25 25 chp TchPamWinSz 25 25 25 25 chp CellRadius 6 6 6 10 kmNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 187
  • 188. Motorola System Parameters Handoff Sector 1 Sector 2 Sector 3 Default C1 C1 C1 SrchWinA 6 6 6 6 chp SrchWinN 8 8 8 8 chp SrchWinR 9 9 9 9 chp TAdd -14 -14 -14 -14 dB TDrop -16 -16 -16 -16 dB TComp 4 4 4 4 dB TTDrop 3 3 3 3November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 188
  • 189. Motorola System Parameters TCH Gain Sector 1 Sector 2 Sector 3 Default C1 C1 C1 MaxGain1Way 127 127 127 127 NomGain1Way 80 80 80 80 MinGain1Way 20 20 20 20 MaxGain2Way 127 127 127 127 NomGain2Way 80 80 80 75 MinGain2Way 20 20 20 20 MaxGain3Way 127 127 127 127 NomGain3Way 80 80 80 75 MinGain3Way 20 20 20 20 StepUp 10 10 10 5 StepDown 1 1 1 1 DeltaTime 7 7 7 7 Frm StepDownDel 21 21 21 21 Frm OrigDelay 100 100 100 100 Frm TchWCCnt 42 42 42 42 TCHNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 189
  • 190. Motorola System Parameters N-Way Sector 1 Sector 2 Sector 3 Default C1 C1 C1 HoConstr 1 1 1 1 MaxActSetSz 6 6 6 6 MaxCEPerCall 3 3 3 3 TcompEnaThr -14.5 -14.5 -14.5 -14.5 MaxBTSLegs1 3 3 3 3 MaxBTSLegs2 2 2 2 3 MaxBTSLegs3 2 2 2 2 AggActLimit1 35 35 35 35 AggActLimit2 43 43 43 43 AggActLimit3 51 51 51 51 EnableSofter Enable Enable Enable 0 EnableBTS Enable Enable Enable 0 EnableSoft Enable Enable Enable 0 AggrStr1 0 0 0 -6 dB AggrStr2 -7 -7 -7 -8 dB AggrStr3 -9 -9 -9 -10 dB NumCandidate 10 10 10 10 XCTComp 3 3 3 4 dB SofterShuff 3 3 3 3 dB BTSShuffleC 3 3 3 0 dB SoftShuffle 3 3 3 3 dBNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 190
  • 191. Course RF200 Section III Introduction to Introduction to Optimization Tools Optimization ToolsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 191
  • 192. Introduction To CDMA Field Tools: Topics Two Important Concepts • The Department Store Analogy - Tops-Down vs. Bottoms-Up • The Aeronautical Analogy - Accident Investigation Resources Survey of CDMA Field Tools • Mobile Tools • Handsets - Maintenance DisplaysNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 192
  • 193. Department Store Analogy: Tops-Down, Bottoms-Up Dis e nc Ta osses ce Management trib utio su ra S ervi on Test Shopper ecti s n In Sel xe Profits Capital L Lea sts Complex!!! s es Simpler Co ng Con rtisi Stocking Su pp ven ve Labor Relations lie Price ienc Ad rs e ce System are Administration rferen s Phone S oftw Inte all Provisioning Trans- pp ed C mission Dro Switch CBSC Complex!!! Simpler Data C Cov PSTN TrunkingData apture erag Analys Acces is s Failur e Neighbor Lists Configuration BTS es Field Tools Some things are easier to measure from the customer side!November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 193
  • 194. Aeronautical Analogy: Tools for Problem Investigation Control & Parameters Messaging 114.50 118.25 11500 11500 130.75 Aeronautical Case Flight Data Recorder Cockpit Voice Recorder CDMA Case BTS Temporal Analyzer Data Layer 3 Message FilesTo study the cause of an aeronautical accident, we try to recover the Flight Data Recorder and the Cockpit Voice Recorder.To study the cause of a CDMA call processing accident, we review data from the Temporal Analyzer and the Layer 3 Message Files -- for the same reasons. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 194
  • 195. Sources of CDMA Data and Tools for Processing CDMA NETWORK EQUIPMENT HANDSET Switch CBSC BTS SLM CM GPSR IS-95/J-STD-8 GPSR Messages BSM CDSU CDSU DISCO TFU1 TFU1Switch Data DMS-BUS DISCO 1 CDSU Ch. Card ACC CDSU CDSU pegs, logsLPP ENET LPP CDSU System DISCO 2 Σα Txcvr A Internal Messages B CDSU RFFE A DTCs CDSU Σβ Txcvr RFFE B Handset SBS CDSU Σχ Txcvr C RFFE C Vocoders Messages PC-based IOC Selectors Mobile Data Capture Tools IS-95/J-STD-008 Messages Unix-based, Various PC-based PC-based External Data Analysis Mobile Data Analysis Post-Processing Post-Processing Tools Tools Tools CDMA optimization data flows from three places: • Switch • CDMA peripherals (CBSC & BTS) • Handset Each stream of data has a family of software and hardware tools for collection and analysis November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 195
  • 196. Autonomous Data Collection Autonomous Data Collection By Subscriber Handsets By Subscriber HandsetsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 196
  • 197. Autonomous Collection: A New Way to See Network Performance Collection Server •software download •collected data upload BTS •data management, analysis PDSN/Foreign Agent Backbone BTS Internet Network VPNs T SECURE TUNNELS T PDSN Authentication Home Agent Authorization AAA R-P Interface Accounting BTS PSTN v SEL t1 t1 t1 Switch (C)BSC/Access Manager BTS An exciting new trend in network RF performance is to embed data collection software on mobile platforms Offers big advantages for RF optimization cost/effectivenessNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 197
  • 198. Using Autonomous Collection Collection Server •software download •collected data upload BTS •data management, analysis PDSN/Foreign Agent Backbone BTS Internet Network VPNs T SECURE TUNNELS T PDSN Authentication Authorization Home Agent Accounting AAA R-P Interface BTS t1 v PSTN SEL t1 t1 Switch (C)BSC/Access Manager BTS A Server downloads software to a large population of subscriber mobiles Mobiles collect on custom profiles • all or groups of mobiles can be enabled/disabled • new triggers can be rapidly developed and downloaded when desired Mobiles upload compacted packets to server driven by custom triggers • may be immediately if needed, or at low-traffic pre-programmed times • collected data can include location/GPS/call event/L3 messaging/timestamps/etc. Server manages data, provides filtering and reporting Performance optimizers use terminals and post-processing softwareNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 198
  • 199. Advantages of Autonomous Collection Mobile-reported data can be location-binned • post-processing provides visual identification of problem areas Collection can be rapidly enabled per cell or area for immediate investigation of problem reports Requires less employee drive time for collection Customer mobiles cover area more densely than drivetesters Customer mobiles include in- building populations Individual mobile identification can be included with customer permission for direct customer service interactionNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 199
  • 200. Current Issues in Autonomous Collection Collection Server •software download •collected data upload BTS •data management, analysis PDSN/Foreign Agent Backbone BTS Internet Network VPNs T SECURE TUNNELS T PDSN Authentication Authorization Home Agent Accounting AAA R-P Interface BTS t1 v PSTN SEL t1 t1 Switch (C)BSC/Access Manager BTS Requires extensive software capability to develop/manage • current progress is from specialty application consulting houses Requires cooperation of handset vendor to effectively integrate software onto handset platform • caution required to avoid negative call processing side-effects Privacy issues involved if any user-specific data tracking Additional network capacity required for large-scale reportingNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 200
  • 201. Conventional Field Tools Conventional Field ToolsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 201
  • 202. CDMA Field Test Tools Field Collection Tools using Handset Data PN Scanners Motorola Qualcomm MDM, CAIT Agilent Berkeley (HP + SAFCO) Varitronics Grayson Agilent Willtech (HP + SAFCO) Grayson Qualcomm Comarco Ericsson TEMS DTI Willtech There are many commercial CDMA field test tools Characteristics of many test tools: • capture data from data ports on commercial handsets • log data onto PCs using proprietary software • can display call parameters, messaging, graphs, and maps • store data in formats readable for post-processing analysis • small and portable, easy to use in vehicles or even on foot A few considerations when selecting test tools: • does it allow integration of network and mobile data? • Cost, features, convenience, availability, and support • new tools are introduced every few months - investigate!November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 202
  • 203. Qualcomm’s MDM: Mobile Diagnostic MonitorThe Qualcomm Mobile DiagnosticMonitor was the industry’s first fielddiagnostic tool • used industry-wide in the early deployment of CDMA • pictures at right from Sprint’s first 1996-7 CDMA trials in Kansas CityQualcomm’s Mobile Diagnostic Monitor • CDMA handset (customer provided) • Proprietary connecting cable • PC software for collection and field pre- analysis – Temporal analyzer display mode – MessagingNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 203
  • 204. Grayson’s Invex3G Tool 100 MB ethernet connection to PC the eight card slots can hold receivers or dual-phone cards there’s also room for two internal PN scanners Multiple Invex units can be cascaded for multi-phone load- test applications Cards are field-swappable - Users can reconfigure the unit in the field for different tasks without factory assistanceNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 204
  • 205. Grayson Invex Playback Example 76.8 kb/sThis mobile is in a 4-way soft handoff(four green FCH walsh codesassigned) in the middle of a downlinkSCH burst. Notice walsh code #2, 8chips long, is assigned as an SCHbut only on one sector, and thedownlink data speed is 76.8kb/s. November, 2004 (c) 2004 Scott BaxterTechnical Introduction to Wireless -- ©1997 Scott Baxter - V0.0 RF200 v4.0 205
  • 206. Grayson Invex Playback Example 153.6 kb/sThis mobile is in a 2-way soft handoff(two green FCH walsh codesassigned) in the middle of a downlinkSCH burst. Notice walsh code #3, 4chips long, is assigned as an SCHbut only on one sector, and thedownlink data speed is 153.6kb/s. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 206
  • 207. Grayson Invex Playback Example F-SCH rates 153.6 kbps; R-SCH 76.8kbps CDMA Status PN Scanner DataCurrent Data Task Status Layer-3 Messages November, 2004 (c) 2004 Scott BaxterTechnical Introduction to Wireless -- ©1997 Scott Baxter - V0.0 RF200 v4.0 207
  • 208. WillTech Tools Blue Rose platform can manage multiple phones and collect data • Internal processor manages test operations independently for stand- alone operation • Internal PCMCIA flash card provides storage • An external PC can display collected data during or after data collectionNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 208
  • 209. Agilent Drive-Test Tools Agilent offers Drive-Test tools • Serial interfaces for up to four CDMA phones • A very flexible digital receiver with several modes PN Scanner • Fast, GPS-locked, can scan two carrier frequencies Spectrum Analyzer • Can scan entire 800 or 1900 mHz. Bands Base-Station Over-Air Tester (BOAT) • Can display all walsh channel activity on a specific sector • Useful for identifying hardware problems, monitoring instantaneous traffic levels, etc. Post-Processing tool: OPAS32November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 209
  • 210. IS-95 Busy Sector Snapshot of Walsh UsageNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 210
  • 211. 1xRTT Busy Sector Walsh Code UsageNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 211
  • 212. Comarco Mobile Tools X-Series Units for more data- intensive collection activities • Multiple handsets can be collected • Data is displayed and collected on PC LT-Series provides integrated display and logging "Workbench" Post-Processing tool analyzes drive-test filesNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 212
  • 213. PN Scanners Why PN scanners? Because phones can’t scan remaining set fast enough, miss transient interfering signals Berkeley Varitronics • high-resolution, GPS-locked – full-PN scan speed 26-2/3 ms. • 2048 parallel processors for very fast detection of transient interferors Agilent (formerly Hewlett-Packard) • high resolution, GPS-locked – full-PN scan speed 1.2 sec. • Integrated with spectrum analyzer and phone call-processing tool Grayson Wireless • New digital receiver provides CDMA PN searcher and and sector walsh domain displaysNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 213
  • 214. Post-Processing ToolsPost-Processing tools display drive-test files for detailed analysis - Faster, more effective than studying data playback with collection tools alone Actix Analyzer • Imports/analyzes data from almost every brand of drive-test collection tool Grayson Interpreter • Imports/analyzes data from Grayson Wireless Inspector, Illuminator, and Invex3G Agilent OPAS32 • Imports/analyzes a variety of data OPAS32 Nortel RF Optimizer • Can merge/analyze drive-test and Nortel CDMA system data Wavelink Comarco "Workbench" Tool Verizon/Airtouch internal tool “DataPro” COMARCONovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 214
  • 215. Drive-Testing Some General Guidelines Some General GuidelinesNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 215
  • 216. Safety Considerations Don’t worry for the company’s loss due to your accidental death • Qualified and eager replacements have resumes on file • We’re constantly buying more drive-test vehicles • We were going to replace that old drive-test equipment soon • We’re not really sure we needed your last drive test, anyway • Your death will serve as a warning to others, so it’s not in vain It’s OK to be careful and continue living for your own sake if you wish! Always start and stop drive test file collection in a safe place off the road and out of traffic patterns • Set up a graph window, message window, etc., whose motion can provide a quick-glance visual reassurance that collection is running OK While on the road, do not attempt to start or stop files, open or close windows, or review results - just glance occasionally for signs of activity If the PC freezes, the power cord pops out, or any other problem occurs while collecting, don’t try to deal with it or correct it while driving • Just pull over to the next really safe place to assess and correctNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 216
  • 217. Physical Considerations Be sure the connections (power, phone, PC and GPS cables) are secure so they won’t dislodge during collection and distract you Be sure the equipment is physically restrained so it won’t go flying around and hit you in case of a panic stop or sudden swerve Some GPS antennas are not weatherproof. Try to avoid getting them drenched in heavy rain The GPS antennas should be mounted where they have a view of the sky as unobstructed as possible External PCS or Cellular antennas should not be mounted closer than about 1 foot to each other or to GPS antennas to ensure there is no significant electromagnetic interference or pattern distortionNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 217
  • 218. Operational Concerns The ideal length for drive-test files is 30 minutes to an hour • You’d hate to lose bigger files in case the PC locks up! • Larger files are a hassle to move around, load, and analyze • When interesting call processing events occur, it’s nice if they are in small files that can be easily processed and stored Always make sure you have at least 2 or 3 GB of free hard drive space before you start a new drive-test collection • Don’t open other programs while collecting data - they can tie up all your free space in swap files and cause a crash • Check your hard drive for errors and defragment it every week or so if you’re collecting and transferring big files Don’t retrace large parts of your travel path during a drive-test run • It’s harder to distinguish what happened on each run when analyzing drives that cruise the same road multiple times Always stop the test call before you stop recording -- otherwise, post- processing software may misinterpret calling eventsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 218
  • 219. Some Manufacturer-Specific Concerns When using Agilent drive-test equipment, data accumulates very quickly. A day’s driving can easily produce a 500 MB data file (data.mdb), or even larger. • Don’t allow file data.mdb to grow too large to manage or copy to other media and computers. Rename or copy and delete the file when it reaches a few hundred MB, and do your next collection in a new file. • Without caution, this file can grow so large that it can’t be practically copied to other computers, and so large that it takes many hours just to open it using post-processing software. On both Grayson and Agilent drive-test equipment, be sure you aren’t configuring the phone to try to dump more data than the data rate of its serial port (38.4 kbps). • On the Grayson, click “Default” settings then click Markov boxes if you wish to see FER data on regular or Markov calls. Also click the SearcherInfo2 box but only if you are investigating search windows. • On the Agilent, don’t create unnecessary measurements that you won’t ordinarily use. But be careful since unlike the Grayson, you can’t open a display window during playback unless that display window existed during collection.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 219
  • 220. Getting Location Data into Drive-Test Files In order to be able to build maps from drive-test data, location information must be imbedded in the data files while they are collected in the field. Several methods for obtaining location data have been popular: GPS Global Positioning System • This is the least expensive and most popular source of location information for drive-testing since 1992 Stored Vector Maps and position-recognition software • Commercial products take raw vehicle distance and direction data and match it to a stored road database to deduce location • Bosch TravelPilot and other tools used this method • More expensive and troublesome than GPS, not popular today LORAN • MF Loran transmissions are only reliable in some coastal areas and are being phased outNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 220
  • 221. GPS Basics GPS (Global Positioning System) was funded and implemented by the US military and serves both civilian and military users • approved military users use a high precision signal (“C/A”) • civilian users use a lower-precision component of the signal • GPS signals are spread-spectrum at 1.545 and 1.2 GHz. Other Global Navigation Systems: • Europe: Galileo (not yet launched) • Russia: GLONASS (in poor repair) GPS uses 21 active satellites and 3 parked spares, all in mid-level orbits at about 10,000 KM • Hour-by-hour, 5 to 7 satellites are usually in view anywhere • Reception of four satellites is enough to fix determine location • Three satellites are enough if user’s elevation already known • GPS reception is often blocked in cities, under bridges, dense forests, or wherever obstacles interrupt the signal path Dead Reckoning is a method of supplementing GPS with independent location information when GPS can’t be received Differential GPS is a technique adding independent corrections to received GPS data for better accuracy. GPS civilian accuracy was improved in May, 2000. DGPS hasn’t been widely used since thenNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 221
  • 222. Dead-Reckoning Systems Dead-reckoning systems normally use a combination of magnetic compass and wheel rotation sensors to augment GPS The manufacturer’s instructions should be followed for installation. Major factors requiring attention are: • If used, Wheel sensors must be securely mounted to prevent accidental breakaway while driving (major injury hazard) • Magnetic compasses should be located as far as possible from magnetic field sources in or on the vehicle – example: mag-mount antennas – (experimentation is often required) • Calibration by actual test is required to achieve workable accuracy for dead-reckoning systemsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 222
  • 223. Drive-Tests: Phones Maintenance Features of Maintenance Features of CDMA Handsets CDMA HandsetsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 223
  • 224. Handsets as Tools: Simple but always Available! Most CDMA handsets provide some form of maintenance display (“Debug Mode”) as well as instrumentation access • all CDMA drive-test tools use handsets as their “front-ends” Using the handset as a manual tool without Commercial Test Tools: Enter the maintenance mode by special sequence of keystrokes Displayed Parameters • PN Offset, Handset Mode, Received RF Level , Transmit Gain Adjust Maintenance Display Applications • best serving cell/sector • simple call debugging (symptoms of weak RF, forward link interference, etc.) Handset Limitations during manual observation • no memory: real-time observations only; no access to messages or call details; serving PN offset not updated during voice calls November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 224
  • 225. Older Qualcomm/Sony Maintenance Displays Press This: See This: continue: See This: Menu D D MAIN MENU DEBUG 0 1:Volume 1:Screen 2:Call Info 2:Test Calls 3:Security 3:CDMA Only 4 * D D DEBUG 0 FEATURES 4 4:Errors 1:AutoAnswer 5:Clr Errors 2:AutoRetry 6:13K Voice 3:Scratch 1 0 D D 318 2 9D X A 7F ENTER FIELD SERVICE CODE ****** See following 0 0 0 0 0 0 * legend for maintenance(* or correct code, if different) display values November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 225
  • 226. Qualcomm & Sony Phones with Jog Dials Enter 111111 Press dial in for OPTIONS Dial to FIELD DEBUG, press enter Field Debug Security Code press ScreenNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 226
  • 227. Interpreting the QCP Maintenance Display 0 - Pilot Channel Acquisition Substate 1 - Sync Channel Acquisition Substate 2 - MS Idle State QCP- QCP- 3 - System Access State 1900 800 4 - Traffic Channel State FF -67 -64 Receive State F5 -70 -67 E6 -75 -72 D D7 -80 -77 C8 -85 -82 B9 -90 -87PN Offset 318 2 94 Receive Power AA -95 -92 9B -100 -97 X A 7F 8C -105 -102 80 -109 -106 Unsupported Transmit Adjust Receive Power Conversion: RXdbm=XXDEC / 3 - 63.25 (800 MHz) A = active pilots 80 -109 RXdbm=XXDEC / 3 - 66.25 (1900 MHz) X = exit reason 80 -109 (if XX>7F, use XX = XXDEC-256) 00 0 Transmit Gain Adjust Conversion: 0A -5 TXADJdb=XXDEC / 2 14 -10 Transmit Power Output Conversion: 1E -15 TXdbm= -73 -RXDBM - TXADJdb (800 MHz) 28 -20 TXdbm= -76 -RXDBM - TXADJdb (1900 MHz) November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 227
  • 228. Kyocera 2035 Maintenance ModeSteps to enter maintenance mode: 111111 Enter Options: Debug Enter Enter Field Debug Code • 000000 Field Debug Debug Screen Enter Basic EnterNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 228
  • 229. Kyocera 6035 Maintenance Mode 111111 Jog > Options Jog > Debug Open flip to continue Enter Code • 000000 OK SCREENNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 229
  • 230. Early Samsung Maintenance Display Press This: See This: continue: See This: SVC SVC Menu Main Menu ↑ Debug Menu ↑ 1:Call Logs 1:Screen 2:Phone Book 2:Test Calls 8 * SVC SVC Debug Menu ↑ Setup ↑ 3:Errors 1:Auto Retry 4:Erase Error 2:Anykey Ans 1 0 SVC SVC S04379 SI0 1 Service Code T-63 D105-06 ?????? P016 CH0600 See following 0 0 0 0 0 0 * legend for maintenance(* or correct code, if different) display values November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 230
  • 231. Samsung SCH-3500 Maintenance DisplayHere are the steps to enter maintenance mode: MENU SETUP 0 (undocumented “trap door”) 000000 (operator’s code) ScreenSee the Samsung idle and in-call maintenance screens at the end of the Samsung phones.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 231
  • 232. Samsung SCH-8500 Maintenance DisplayHere are the steps to enter maintenance mode: [Menu] [down][down][down][down] [down][down][down] Setup/Tool [OK] [0] Service Code ?????? [0] [4] [0] [7] [9] [3] Screen [OK]See the Samsung idle and in- call maintenance screens at the end of the Samsung phones.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 232
  • 233. Samsung SCH-A500 Maintenance Display Here are the steps to enter maintenance mode: Select settings select display select enter 0 enter 040793 See the Samsung idle and in- call maintenance screens at the end of the Samsung phones.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 233
  • 234. Interpreting Samsung Maintenance Display: Acquisition, Idle, and Access States 0 - Pilot Channel Acquisition Substate 1 - Sync Channel Acquisition SubstateDisplay toggles between: 2 - MS Idle State Slot Cycle Index 3 - System Access State System Identifier (SID) 4 - Traffic Channel State Network Identifier (NID) 5,6,7 - various call service options Processing State Receive svc Power, S04379 SI0 1 dbm Transmit Gain Adjust, T-63 D085-06 Ec/Io, db (primary PN only) db P016 CH0600 Frequency PN Offset (channel #) Transmit Power Output Calculation: TXdbm= -73 -RXDBM - TXADJdb (800 MHz) TXdbm= -76 -RXDBM - TXADJdb (1900 MHz) November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 234
  • 235. Interpreting Samsung Maintenance Display: Traffic Channel State 0 - Pilot Channel Acquisition Substate 1 - Sync Channel Acquisition Substate Transmit Receive Walsh 2 - MS Idle State Vocoder Rate Vocoder code 3 - System Access State 1 = 1/8 Rate assigned 4 - Traffic Channel State 5,6,7 - various call service options 2 = 1/4 Processing State 4 = 1/2 svc Receive 8 = Full Power, Transmit TV1 RV8 08 7 dbmGain Adjust, T-63 D085-06 Ec/Io, db db P016 CH0600 (primary PN only) Frequency PN Offset (channel #) Transmit Power Output Calculation: TXdbm= -73 -RXDBM - TXADJdb (800 MHz) TXdbm= -76 -RXDBM - TXADJdb (1900 MHz)November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 235
  • 236. Entering Denso Debug Mode D Enter ##DEBUG (##33284) CBV: 3957 ABU: 3954 ABT: 031 Scroll down to SAVE ARF: 0000 CCL: 01 SID: 04157 Press OK NID: 00001 CH: 0100 RSSI: 093 Highlight SERVICE SCREEN DPN: 084 TX:-46 Press OK BFRM:0000000968 TFRM:0000135712 FER:% 000.71 LT: 036:06:36 If you want to make a test call, LG: -086:45:36 EC: -16 -63 -63 dial the digits and press OK PN: 084 084 084 while in idle mode FNGLK: Y Y N WLSH: 01 01 01 ACT: 084 484 096 -01 -01 200 CND: 220 332 200 200 332 NGH: 076 080 340 068 196 O56 320 220 316 344 488 196 200 392 124 128 084 224 008 084November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 236
  • 237. Denso Maintenance Display Charging Battery Voltage D Average Battery Voltage CBV: 3957 Average Battery Temperature ABV: 3954 ABT: 031 System ID ARF: 0000 CCL: 01 SID: 04157 Network ID NID: 00001 Received Signal Strength RF Channel Frequency CH: 0100 RSSI: 093 DPN: 084 TX:-46 Estimated Transmitter Digital PN Offset Power Output BFRM:0000000968 Number of Bad Frames TFRM:0000135712 Number of Good Frames FER:% 000.71 Frame Erasure Rate, Percent Base Station coordinates LT: 036:06:36 LG: -086:45:36 EC: -16 -63 -63Current status of Rake Fingers PN: 084 084 084 FNGLK: Y Y N WLSH: 01 01 01 Active Pilot Set ACT: 084 484 096 -01 -01 200 Candidate Pilot Set CND: 220 332 200 200 332 NGH: 076 Neighbor Pilot Set 080 340 068 196 O56 320 220 316 344 488 196 200 392 124 128 084 224 008 084 November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 237
  • 238. Early Sanyo Dual-Band Phones Press This: press menu 7, 0 Menu enter in DEBUGM (332846) screens are similar to QCP phones 7 D 0 318 2 94 X A 7F3 3 2 8 4 6November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 238
  • 239. Sanyo SPC-4500 Maintenance Display Choose the following: DISPLAY OK 0 OK Enter Code: 0 0 0 0 0 0 Debug Menu SCREEN OKNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 239
  • 240. Sanyo SPC-4900 Maintenance Display Call Proc. State Receive ## Power PN offset Io 040793 select MENU/OK button scroll to save Phone # Channel select FrequencyNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 240
  • 241. Entering Maintenance Mode: Motorola StarTacContact your service provider to obtain your phone’s Master Subscriber entity Lock (MSL). Then enter the following: FCN 000000 000000 0 RCL Youll be prompted for your MSL, enter it and press STO. • New prompts will appear, Press STO in response to each prompt until no more appear. Don’t delay - continue quickly and enter: FCN 0 0 * * T E S T M O D E STO • The display will briefly show US then just . Press 55#. • Step 1 will appear with its current setting displayed. Press * to accept and move on to the next step. Repeat for steps 2-8. Step 9 (Option byte 2) is the only step requiring manual changes. Enter 1 0 0 0 0 0 0 0 (The leftmost bit now set to 1 is what enables test mode.) Now press STO to accept the entry and exit back to the prompt. Power off and back on. You should now be in test mode!November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 241
  • 242. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 242
  • 243. Last Call Indicator N5 N5M failure NI No Indication yet BS BS Ack failure MR Mobile Release WO L3 WFO State Timeout BR Base Sta. Release MP Max Probe Failure TC Traffic Channel Lost PC Paging Channel loss L2 Layer 2 Ack Fail RR Reorder or Release on PCH NC No Channel Assn Msg ?? Unknown Condition Battery RX Power Local Time Condition Strongest Active # # Channel PN Ec/Io Actives Neighbors Number Strongest Neighbor # Cand- Call Proc Last Call PN Ec/Io idates State Exit Reason Rx Power Tx Power Last Call FER% # Drops dbm (Io) dbm Current # CallsService Option SID NID Call Processing States ORG Call Origination CP CP Exit SMS Short Message SvcCurrent Service Option RST CP Restart ORD Order Response8V 8K voice original 13S 13K SMS RTC Restricted REG Registration PLT Pilot Acquisition TCI Tfc Ch InitializationIL 8K loopback 8MO 8K Markov Old SYN Sync Acquisition WFO Waiting for Order8EV 8K EVRC DAT Data TIM Timing Change WFA Waiting for Answer8S 8K SMS 8M 8K Markov New BKS Background Sch Conversation state 13M 13K Markov New IDL Idle CON13L 13K loopback OVD Overhead REL Release November, 2004 13V 13KRF200 v4.0 (c) 2004 Scott Baxter Voice PAG Paging NON No State RF200 - 243
  • 244. Motorola V120C Series MENU 073887* Enter 000000 for security code. Scroll down to Test Mode. Enter subscriber entity lock code if required by your phoneSame maintenance display as shown for StartacNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 244
  • 245. Motorola V60C MENU 073887* Enter 000000 for security code. Scroll down to Test Mode. Enter subscriber entity lock code if required by your phoneSame maintenance display as shown for StartacNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 245
  • 246. Audiovox 8100, 9155 Press ##27732726 [End] Select the Debug screen. PN, channel#, SID, NID, mode (13K, EVRC, etc) Ec/Io, RX Level, TX Level. You cannot make a call while in any of the maintenance screens.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 246
  • 247. NeoPoint PhonesAlthough NeoPoint went out of business in June, 2001, there are still some NeoPoint handsets in general use Press the M (menu) key Select Preferences (using the up-arrow key) Enter 040793 Choose Debug Screen [Select] Now you’re in maintenance mode!November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 247
  • 248. GoldStar TouchPoint To enter maintenance mode, just key in:# # D E B U G SAVENovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 248
  • 249. Nokia 6185 Maintenance Display Enter *3001#12345# MENU Scroll down to Field test Press Select Scroll up to Enabled Press OK Power the phone off and on You should now be in Field test modeNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 249
  • 250. Older Nokia Models Maintenance Display Enter *3001#12345# MENU Scroll down to Field test Press Select Scroll up to Enabled Press OK Power the phone off and on You should now be in Field test mode and the following screens will be available:November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 250
  • 251. Maintenance Display Screens of Nokia Handsets The following screens appear in field test mode on Nokia HD881 series of Handsets: Screen 1: General Screen 5: NAM Info CSST CS State PPCA Primary Channel A Idle: PN Offset SPCA Secondary Channel A XXXXX TFC: #Actv, FER PPCB Primary Channel B RSSI RSSI dBm SPCB Secondary Channel B CCCC Paging Channel # L Local Use RX RX power, dbm A Access Overload Class TX TX power, dbm Screen 6: BS & Access. Info. Screen 2: Paging CH Info SID Current SID CSST CS State NID Current NID PGCH Paging Channel # DBUS DBUS (Handsfree?) CURSO Current Service Option FER Frame Error Rate Screen 7: BS Protocol Rev. Level BASE# BASE_ID (sys par msg) Screen 4: NAM Info P_REV P_REV (sync msg) OwnNumber Mobile MIN MIN_P_REV MIN_P_REV (sync msg. ESN Mobile Station ESN Screen 8: Time Information Preferred Sys CSST CS State P 1=AMPS, 2=CDMA MMDDYY Date from System Time Operator Selected HHMMSS System Time A (1=A, 2=B, 3=bothNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 251
  • 252. Nokia Maintenance Display Screens (continued)Screen 9: Acquisition Information Screen 11: Active Set (#4-6) TA TADD PPN Pilot PN Offset TD TDROP EC Ec/Io in 1/2 db units TC TCOMP K Keep? 1 TT TTDROP PPN Pilot PN Offset WW1 Active Window EC Ec/Io in 1/2 db units WW2 Neighbor Window K Keep? 1 WW3 Remaining Window PPN Pilot PN Offset EC Ec/Io in 1/2 db units Screen 10: Active Set (#1-3) K Keep? 1 PPN Pilot PN Offset EC Ec/Io in 1/2 db units K Keep? 1 PPN Pilot PN Offset EC Ec/Io in 1/2 db units K Keep? 1 PPN Pilot PN Offset EC Ec/Io in 1/2 db units K Keep? 1November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 252
  • 253. Nokia Maintenance Display Screens (continued) Screen 12: Neighbor Set (#1-5) Screen 14: Neighbor Set (#11-15) PPN NBR 1 PN Offset PPN NBR 11 PN Offset EC Ec/Io in 1/2 db units EC Ec/Io in 1/2 db units PPN NBR 2 PN Offset PPN NBR 12 PN Offset EC Ec/Io in 1/2 db units EC Ec/Io in 1/2 db units PPN NBR 3 PN Offset PPN NBR 13 PN Offset EC Ec/Io in 1/2 db units EC Ec/Io in 1/2 db units PPN NBR 4 PN Offset PPN NBR 14 PN Offset EC Ec/Io in 1/2 db units EC Ec/Io in 1/2 db units PPN NBR 5 PN Offset PPN NBR 15 PN Offset EC Ec/Io in 1/2 db units EC Ec/Io in 1/2 db units Screen 13: Neighbor Set (#6-10) Screen 15: Neighbor Set (#16-20) PPN NBR 6 PN Offset PPN NBR 16 PN Offset EC Ec/Io in 1/2 db units EC Ec/Io in 1/2 db units PPN NBR 7 PN Offset PPN NBR 17 PN Offset EC Ec/Io in 1/2 db units EC Ec/Io in 1/2 db units PPN NBR 8 PN Offset PPN NBR 18 PN Offset EC Ec/Io in 1/2 db units EC Ec/Io in 1/2 db units PPN NBR 9 PN Offset PPN NBR 19 PN Offset EC Ec/Io in 1/2 db units EC Ec/Io in 1/2 db units PPN NBR 10 PN Offset PPN NBR 20 PN Offset EC Ec/Io in 1/2 db units EC Ec/Io in 1/2 db unitsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 253
  • 254. Nokia Maintenance Display Screens (continued) Screen 16: Candidate Set (#1-5) PPN CAND 1 PN Offset EC Ec/Io in 1/2 db units PPN CAND 2 PN Offset EC Ec/Io in 1/2 db units PPN CAND 3 PN Offset EC Ec/Io in 1/2 db units PPN CAND 4 PN Offset EC Ec/Io in 1/2 db units PPN CAND 5 PN Offset EC Ec/Io in 1/2 db units Screen 17-22: Task Stack Ck Info TASKN Task Name FREE Worst-Cs Stack Free Sp Screen 23: Stack Status Info.Task Stack Overflow ind. by shiftSys Stack 2=sys stack overflow Screen 24: Codec RegistersNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 254
  • 255. New CDMA PhonesNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 255
  • 256. New CDMA PhonesNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 256
  • 257. New CDMA PhonesNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 257
  • 258. New CDMA PhonesNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 258
  • 259. New CDMA PhonesNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 259
  • 260. Novatel Merlin C201 Card Enter # # D E B U G to enter maintenance mode. To exit, just click “OK” box in the Debug window.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 260
  • 261. Audiovox Thera Maintenance Mode ScreensHow to enterDebug Mode:[ctrl] [D] [enter]Advanced Usr Pwd:##DEBUG [enter]Protocol Statistics November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 261
  • 262. RF200 Section IV Multi-Carrier Operation Multi-Carrier Operation and Its Complications and Its ComplicationsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 262
  • 263. A CDMA network with 5 carriersNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 263
  • 264. It’s AMulti-Carrier/Multi-System/Multi-Manufacturer World! Systems are forced to use multiple carriers to achieve needed traffic capacity • It’s important that the traffic load be divided between carriers Physically adjacent friendly systems often desire to allow seamless mobile operation across their borders, although they use different carrier frequencies Even within one large network, seamless mobile operation is desired across serving switch boundaries These situations are not completely solved in the original IS-95 CDMA vision, so additional standards documents and additional proprietary processes provide the needed functionality • IS-95: hashing or GSRMs can distribute idle mobiles among carriers • IS-41 - provides intersystem handoffs and call delivery • Proprietary algorithms can distribute in-call traffic among carriers • RF tricks and network proprietary algorithms can support inter-carrier handoff Multi-Carrier Operation is a complex sportNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 264
  • 265. Transitions at System Boundaries IDLE IDLE IN-CALL IN-CALL Boundary types • between different operators – same frequency, different frequency, even different band • between different BSCs or Switches of Same Operator – same frequency, different frequency, even different band • between different carriers where number of carriers changes – same frequency, different frequency, even different band! A reliable transition method must be planned for users in all circumstances • all directions of approach • all modes of operation (idle, active voice call, dormant data session, active data session)November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 265
  • 266. Foundation for Transition Troubleshooting Multi-carrier and intersystem boundary transitions are complex relationships between mobile, air interface, and system • to solve problems, it’s necessary to understand the basic actions of mobile and the system – this information comes from the standard, summarized in the next few slides The mobile’s actions are generic, defined by the standards, and simpler/more specific than the steps taken by the system • A thorough knowledge of the mobile side is the easiest-to-get resource for general troubleshooting of problems For in-call transition troubleshooting, the system’s generic and proprietary algorithms must also be understood • artificial proprietary trigger mechanisms and internal system order communications and IS-41 implementation – this information comes from manufacturer documentation • trunking and networking between adjoining systems – this information comes from operator’s own network designNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 266
  • 267. Course RF200 MultiCarrier Operation: MultiCarrier Operation: Big-Picture View of Frequency Changes Big-Picture View of Frequency ChangesNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 267
  • 268. Multi-Carrier Operation: Mobiles Change Frequencies. When/Why/How? System Idle Mode Call Start: In Call: Acquisition Reselection Ch. Assignment Hard Handofff5f4 Hashing: Proprietary IS-95 or Network 1xRTT Algorithmsf3 Nortel: MCTA Auxiliary Lucent: Handoff Triggers GSRM SDA Motorola: •Beacons Multi-f2 MRU PRL-AI Freq •Ec/Io, RTD Proprietary Nbrs Processesf1 Remember: Different Mechanisms Apply at Different Stages November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 268
  • 269. f1 f2 f3 f4 W0 Pilot W0 Pilot W0 Pilot W0 Pilot w1 Paging w1 Paging w1 Paging w1 Paging wa Traffic wa Traffic wa Traffic wa Traffic wb Traffic wb Traffic wb Traffic wb Traffic w32 Sync w32 Sync w32 Sync w32 Sync wx Traffic wx Traffic wx Traffic wx Traffic wy Traffic wy Traffic wy Traffic wy TrafficNovember, 2004 Operation wz Traffic wz Traffic wz Traffic wz Traffic Basic Multi-Carrier IS-95 IS-95 IS-95 IS-95 f1 f2 f3 f4 W0 Pilot W0 Pilot W0 Pilot W0 Pilot w1 Paging wa Traffic wa Traffic wa Traffic wa Traffic wb Traffic wb Traffic wb Traffic wb Traffic wc Traffic wc Traffic wc Traffic w32 Sync wd Traffic wd Traffic wd Traffic wx Traffic wx Traffic wx Traffic wx Traffic wy Traffic wy Traffic wy Traffic wy Traffic wz Traffic wz Traffic wz Traffic wz Traffic can carry more traffic! IS-95 IS-95 IS-95 IS-95 Non-originating carriersRF200 v4.0 (c) 2004 Scott Baxter f1 f2 f3 f4 W0 Pilot W0 Pilot W0 Pilot W0 Pilot w1 Paging wa Traffic wa Traffic w1 Paging wa Traffic wb Traffic wb Traffic wa wb Traffic wc Traffic wc Traffic Data w32 Sync wd Traffic wd Traffic w32 Sync wx Traffic wx Traffic wx Traffic wx Traffic wy Traffic wy Traffic wy Traffic wy Traffic wz Traffic wz Traffic wz Traffic wz Traffic support 1xRTT Many Network/Carrier Configurations are Possible!RF200 - 269 Some Carriers may IS-95 IS-95 IS-95 1xRTT
  • 270. Within My Systems The Adjoining Network View: The Adjoining Network View: Different Common Configurations Different Common ConfigurationsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 270
  • 271. The Big Picture:CDMA Multicarrier System Overlaying Analog System CDMA F3 CDMA F2 CDMA F1 Analog SystemImportant Questions: How do idle dual-mode mobiles choose a system? • When do they select analog operation? How do idle CDMA mobiles change carrier frequencies? How do CDMA mobiles in a call handoff to other carrier frequencies? Can CDMA mobiles in a call hand down to analog operation? When can a dual mode mobile return from analog to CDMA?November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 271
  • 272. Adjoining CDMA Networks of Different Manufacturers F3 F2 F2 F1 Brand X System Brand Y SystemPSTN Ordinary Interswitch Trunks (can’t transmit packets, so soft handoff impossible)At present, only Hard Handoffs work between different manufacturersImportant Questions: What happens if bordering cells are on the same frequency? • Advantages and drawbacks What happens if bordering cells are on different frequencies? • Advantages and drawbacksNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 272
  • 273. Adjoining CDMA Networks of the Same Manufacturer F3 F4 F2 F1 F1 Brand X System Brand X System ATM linksPSTN between CDMA packet networks (soft handoffs are desired)At present, most manufacturers support intersystem soft handoffImportant Questions: What happens if bordering cells are on the same frequency? • Advantages and drawbacks What happens if bordering cells are on different frequencies? • Advantages and drawbacksNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 273
  • 274. My Mobile The Mobile View: The Mobile View: When Do II Change Frequencies? When Do Change Frequencies?November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 274
  • 275. Multi-Carrier Operation: Mobiles Change Frequencies. When/Why/How? System Idle Mode Call Start: In Call: Acquisition Reselection Ch. Assignment Hard Handofff5f4 Proprietary Hashing Network Algorithmsf3 Nortel: MCTA Auxiliary Lucent methods Handoff Triggers GSRM SDA Motorola methods •Beacons Multi-f2 MRU PRL-AI Freq •Ec/Io, RTD Proprietary Nbrs Processesf1 Remember: Different Mechanisms Apply at Different Stages November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 275
  • 276. System Acquisition/Idle Mode ReSelection System Idle Mode Idle mobiles are like Acquisition Reselection automobile drivers! • There are rules whichf5 they obey, but • They decide where they want to go withoutf4 personal intervention Hashing by the systemf3 The SDA guides mobiles to choose the appropriate GSRM system SDA Multi-f2 MRU PRL-AI Freq Paging channel messages Nbrs can cause idle mode carrier or systemf1 reselection Mobiles find the proper system through both standard-defined acquisition steps and a proprietary System Determination Algorithm November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 276
  • 277. Basic Principles: System Determination in Idle Mode CDMA Idle Mode Mobile has control, follows the System Determination Algorithm Look at most recently used frequency. Find Strongest Pilot > Read Sync > If system denied or not preferred, check other frequencies in PRL. Read Paging/Config Messages If Multiple Frequencies appear in CDMA Channel List Message, Hash and go to proper frequency If GSRM transmitted, go wherever directed Monitor Paging Channel Analog Idle Mode Mobile has control, follows procedures of the Standard Find Strongest CCH Monitor Paging Channel Every 3 minutes, rescan for CDMA signalNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 277
  • 278. Summary: How Idle Mobiles Choose CDMA CarriersAt turnon, Idle mobiles use proprietary System Determination Algorithms(SDA) to find the initial CDMA carrier intended for them to useOn the paging channel of the idle mobile’s newly-found home signal, themobile might be sent to a different frequency if it hears • CDMA Channel List Message • Global Service Redirection Message (GSRM) Start System Determination Algorithm Preferred MRU Only Bit 0 PRL Acq Idx Go to last Strongest Is better Yes Idle Mode Carrier Selection Is SID frequency PN, read SID permitted? from MRU Sync available? F3 No Signal Denied SID No CDMA Ch HASH using F2 Config List Message IMSI F1 Messages: Read remain Last Resort: Paging GEO escape Channel Global Svc my ACCOLC? Or Analog Redir Msg redirect to another CDMA frequency or system Legend to Analog Steps from Steps from Proprietary the CDMA proprietary SDA standards SDAs databases November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 278
  • 279. Avoiding Unwanted Acquisition of Supplemental CDMA Carriers CDMA Carrier Frequency 2 GSRM GSRM CDMA Carrier Frequency 1 System acquisition is primarily controlled by the mobile • dual-mode mobiles look for last-used frequency first Distant mobiles may notice weak Carrier 2 signals beyond the edge of Carrier 2 coverage, and originate calls likely to drop • system can transmit Global Service Redirection Messages on all out-looking Carrier 2 sectors to immediately force any distant mobiles to reacquire Carrier 1 – there will be no F2 originations on outermost F2 sectors! – However, still possible to soft-handoff into F2 outer sectorsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 279
  • 280. Frequency Change at Initial Channel Assignment Call Start: When a new call is about to be assigned to Ch. Assignment its first traffic channel - it’s an ideal time to change carrier frequencies for intercarrierf5 traffic distribution purposes • No call yet exists, so no muting occursf4 Proprietary Each network manufacturer is free to Network design its own internal algorithms to make Algorithms the decision of which carrier should bef3 Nortel: MCTA used Lucent methods Motorola methods • These algorithms consider the presentf2 loading condition of each carrier Mobiles simply follow the instructions sent to them in the Channel Assignmentf1 Message on the paging channel Call Start is a painless time to change to a different frequency.Different networks have different proprietary triggers to distribute traffic. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 280
  • 281. System Determination During Call Setup and Call Continuation CDMA Conversation State System has control, follows Standard or proprietary procedures •Initial channel assignment: system can select which frequency (most common trigger would be congestion on present frequency) •Normal handoffs are soft, on same frequency, to mobile-selected pilots •Artificial trigger mechanisms can force mobile handoff to different: 1) CDMA frequency, 2) CDMA system, or 3) analog system Analog Conversation State System has control, follows procedures of the Standard •Mobile can be handed off to different analog cell or even different analog system based on locate receiver measurements •No handoff possible to CDMA from ongoing analog callNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 281
  • 282. Interfrequency Hard Handoff In Call: Mobiles in a call are receiving only their Hard Handoff current operating frequency • They’re unaware of the presence orf5 absence of signals on different carrier frequencies, so they don’t realize when they need to do intercarrier handoffsf4 Networks use a variety of methods to trick mobiles into appropriate handoffsf3 Auxiliary • Pilot beacons - “decoy” signals on the current frequency that lure the mobile Handoff Triggers •Beacons into disclosing needed informationf2 •Ec/Io, RTD Proprietary • Tier-based triggers Processes – Round trip delay thresholdsf1 – Ec/Io and other parameter thresholdsMobiles in conversation can’t see pilots on different carrier frequencies. We must “trick” these mobiles into handoff by artificial means. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 282
  • 283. Intersystem Hard HandoffSame Frequency causes Interference Problems! BSC1 SW1 City 2 Frequency 1 Interference SW2 BSC2 City 1 Consider two adjacent CDMA systems: • Same frequency • If not equipped for intersystem soft handoff, only hard handoff is possible between them; “dragged” handoffs become a big problem Handoff Performance Results: • Mobiles CAN see pilots from adjoining system, so mobile-directed handoff is possible • However, due to hard handoff mobiles can use only one system or the other, not both, and simultaneous shared power control is not possible • “dragging” mobiles cause severe interference in border cells • border area has poor capacity, high access failures and dropped calls November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 283
  • 284. Intersystem Soft Handoff: Avoids Border Area Interference Problems BSC1 SW1 City 2 Frequency 1 no problems! SW2 BSC2 City 1 Intersystem Soft Handoff ATM link Consider two adjacent CDMA systems: • Same frequency • ATM connection between BSCs allows soft handoff Handoff Performance Results: • Mobiles CAN see pilots from adjoining system, so mobile-directed handoff is possible • Intersystem soft handoff is possible, so simultaneous power control is possible for mobiles in border area • Border RF environment is the same as internal RF environment, no special problemsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 284
  • 285. Avoid Interference, Use Different Frequencies? Hard Handoff Logistical Problems F2 Mobiles can’t see F1 pilots! BSC1 SW1 Frequency 2 City 2 Frequency 1 SW2 BSC2 F1 Mobiles can’t see F2 pilots! City 1 Consider two adjacent CDMA systems: • Suppose intersystem soft handoff is not available • Systems are deliberately on different frequencies. This definitely avoids interference in the border area, but causes other complications Conversation-State Handoff Logistical Problems: • Mobiles on one system can’t see the pilots of adjoining cells on the other system! So, the mobiles will never request trans-border handoff • Some method must be employed to force unsuspecting mobiles into transborder handoffs • Common solutions: 1) implement intersystem soft handoff, 2) Pilot beacon cells, 3) auxiliary trigger mechanisms (Ec/Io, RTD, etc.) November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 285
  • 286. One Solution to the Multi-Frequency Problem 2-Frequency Trigger Method: Beacon Cells F2 Mobiles can see F2 beacon BSC1 SW1Frequency 2 City 2Frequency 1 SW2 BSC2 F1 Mobiles can see F1 beacon City 1 The Beacon Solution • A pilot beacon cell is a “mannequin” -- a signal which can be seen by arriving mobiles from the other system on their own frequency, inducing them to request handoff as soon as it is appropriate • When mobiles request soft handoff with the beacon, the old system steps in and instructs the mobiles to do intersystem hard handoff to the real cell which the mobiles are approaching on the other system Special Logistical Concerns with Beacons • Of course, it’s possible for mobiles of one system to “wake up” looking at the pilot of a beacon cell in the border area, rather than a real cell. • Therefore, a beacon cell must transmit not only its pilot, but also a sync channel and a paging channel with global service redirectionNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 286
  • 287. Another Solution for Multi-Frequency Handoffs Bridge Cells, RTD Trigger in Boundary Sectors BSC1 SW1 Frequency 2 Boundary Sector City 2 Frequency 1 Boundary Sector SW2 BSC2 City 1 All along the intersystem border, a one-cell-thick “transition zone” is created. The “bridge” cells in this zone are equipped with dual equipment, one set operating on each system. • The outlooking sector of each bridge cell is tagged in the site database as a “boundary sector”. Whenever a mobile is served exclusively by a boundary sector, the system continuously monitors that mobile’s round trip delay (RTD). • When the mobile’s RTD passes upward through a datafilled threshold, the system steps in and orders a hard handoff to the matching sector of the bridge cell on the other system – this ensures the handoffs happen in clean environments with high probability of success – disadvantage: more BTS hardware needed than otherwise November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 287
  • 288. Another Solution for Multi-Frequency Handoffs Arbitrary Ec/Io Trigger Mechanisms BSC1 SW1 Frequency 2 Boundary Sector City 2 Frequency 1 Boundary Sector SW2 BSC2 City 1 Outlooking sectors of border cells are tagged as “boundary sectors” in the system database • Whenever a mobile is served exclusively by a boundary sector, the system frequently interrogates the mobile with pilot measurement request messages • When the mobile’s reports the boundary sector’s Ec/Io is below a preset threshold, the system immediately commands a hard handoff to a previously defined sector on the other system. Everyone hopes (prays?) that sector is able to hear the mobile for a successful handoff. – The Ec/Io trigger threshold is sometimes a fixed value (usually 11 db above the T_Drop in the serving sector, although some networks’ later software allows an arbitrary trigger level to be set November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 288
  • 289. CDMA/Analog Overlay Considerations CDMA/Analog Overlay ConsiderationsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 289
  • 290. CDMA/AMPS Overlays: Idle CDMA Acquisition CDMA Overlay GSRM GSRM AMPS Existing System System acquisition is primarily controlled by the mobile • dual-mode mobiles look for CDMA first, then AMPS if needed Distant mobiles may notice weak CDMA signals beyond the edge of CDMA coverage, and originate calls likely to drop • most systems transmit Global Service Redirection Messages on all out-looking sectors to immediately force any distant mobiles to reacquire AMPS system – hence no CDMA originations on outermost CDMA sectors! – However, still possible to soft-handoff into outer sectors Most operators request handset manufacturers to add feature of periodic rechecking by idle handsets seeking to acquire CDMANovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 290
  • 291. CDMA/AMPS Overlays: Analog Handdown CDMA Overlay AMPS Existing System CDMA mobiles approaching the edge of CDMA coverage must hand down to AMPS • however, CDMA mobiles cannot see AMPS signals during CDMA calls, and therefore will not request handoff Methods for triggering CDMA-to-AMPS Handdown: the same ones we considered for CDMA-CDMA intersystem handoff • beacon cells • bridge cells with RTD trigger • arbitrary Ec/Io thresholds on boundary sectors Once a CDMA phone hands down to analog, it cannot be handed back up during the same call (due to long CDMA acquisition time)November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 291
  • 292. Course RF200 Section V. Applied Optimization Applied OptimizationNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 292
  • 293. Good Performance is really Simple!! Although there are many phases of optimizationBTS BTS activities, good performance is really just compliance with a very simple idea One, Two, or Three good signals in handoff • Composite Ec/Io > -10 dbBTS Enough capacity for the offered traffic • No resource problemsEc/Io BTS A BTS B BTS C -10 In principle, A COW next door can solve almost any CDMA problem! FORWARD Reality Check: available LINK 1. But who has enough regular cells OR cows or money to power fix every problem location?!! Traffic 2. Problems occur in the areas between cells’ dominant Channels coverage. Adding a cow only pushes the problems In use out to its own boundary with other cells. Paging Conclusion: We need to design better, and to use our Sync existing cells more effectively. We need to provide Pilot one, two, or three dominant signals everywhere. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 293
  • 294. Bad Performance Has Many Causes +41 Weak Signal / Coverage Hole 360 +8 Pilot Pollution A 360+33c • Excessive Soft Handoff BTS B Handoff Failures, “Rogue” mobiles BTS • Missing Neighbors • Search Windows Too Small • BTS Resource Overload / No Resources BTS Rx Pwr No Overload – No Forward Power, Channel Available Elements Power! – No available Walsh Codes BTS Sector Transmitter Traffic – No space in Packet Pipes Channels In Use CEs Pilot “Surprise” ambush; Slow Handoffs Paging x PN Plan errors Slow Data Problems: RF or IP congestion Sync Pilot Vocoders Improper cell or reradiator configurationBTS A Selectors BTS B Hardware and software failuresPN 100 PN 99 ACTIVE SEARCH WINDOW But on analysis, all of these problems’ bad effects happen because the simple few-signal 1 mile 11 miles ideal CDMA environment isn’t possible. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 294
  • 295. Starting Optimization on a New System RF Coverage Control • try to contain each sector’s coverage, avoiding gross spillover into other sectors • tools: PN Plots, Handoff State Plots, Mobile TX plots Neighbor List Tuning • try to groom each sector’s neighbors to only those necessary but be alert to special needs due to topography and traffic • tools: PSMM data from mobiles; propagation prediction Search Window Settings • find best settings for SRCH_WIN_A, _N, _R • especially optimize SRCH_WIN_A per sector using collected finger separation data; has major impact on pilot search speed Access Failures, Dropped Call Analysis • finally, iterative corrections until within numerical goalsGetting these items into shape provides a solid baseline and foundation from which future performance issues can be addressed. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 295
  • 296. Performance Monitoring/Growth Management Benchmark Existing Performance • Dropped Call %, Access Failure %, traffic levels Identify Problem Cells and Clusters • weigh cells and clusters against one another Look for signs of Overload • TCE or Walsh minutes -- excessive ? Soft handoff excessive? • Required number of channel elements -- excessive? • Forward Power Overloads: Originations, Handoffs blocked Traffic Trending and Projection • track busy-hour traffic on each sector; predict exhaustion • develop plan for expansion and capacity relief – split cells, multi-sector expansions, multiple carriersThese steps must be continuously applied to guide needed growth. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 296
  • 297. CDMA Problems, Causes, and CuresPROBLEMS Excessive Access Failures Excessive Dropped Calls Forward Link Interference Slow Handoff Handoff Pilot Search Window Issues PN Planning Considerations Excessive Soft Handoff Grooming Neighbor Lists Software Bugs, Protocol ViolationsEXAMPLES Normal Call Dropped Call - Coverage Dropped Call - Neighbor List Dropped Call - Search WindowNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 297
  • 298. Solving CDMA Performance Problems CDMA optimization is very different from optimization in analog technologies such as AMPS AMPS: a skilled engineer with a handset or simple equipment can hear, diagnose, and correct many common problems • co-channel, adjacent channel, external interferences • dragged handoffs, frequency plan problems CDMA impairments have one audible symptom: Dropped Call • voice quality remains excellent with perhaps just a hint of garbling as the call approaches death in a hostile RF environment Successful CDMA Optimization requires: • recognition and understanding of common reasons for call failure • capture of RF and digital parameters of the call prior to drop • analysis of call flow, checking messages on both forward and reverse links to establish “what happened”, where, and why.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 298
  • 299. Normal Call Processing Normal Call Processing Event Templates Event TemplatesNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 299
  • 300. Registration Registration Message (by PROBING) BTS Base Station Acknowledgment OrderPaging AccessChannel Channel November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 300
  • 301. Voice Mail Notification Feature Notification Message BTS Mobile Station Ack’mt. (by PROBING)Paging Base Station Acknowledgment Order AccessChannel Channel November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 301
  • 302. Incoming Call Delivery Scenario General Page Message Page Response Message (by PROBING) BTS Base Station Acknowledgment OrderPaging Channel Assignment Message AccessChannel Channel Continuous frames of all 000’s Traffic Channel Preamble: Frames of 000’s Base Station Acknowledgment OrderForward Reverse Traffic TrafficChannel Mobile Station Acknowledgment Order Channel Service Connect Message Service Connect Complete Message The Call is now officially Established! November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 302
  • 303. Mobile-Originated Call Scenario Origination Message (by PROBING) BTS Base Station Acknowledgment OrderPaging Channel Assignment Message AccessChannel Channel Continuous frames of all 000’s Traffic Channel Preamble: Frames of 000’s Base Station Acknowledgment OrderForward Reverse Traffic TrafficChannel Mobile Station Acknowledgment Order Channel Service Connect Message Service Connect Complete Message The Call is now officially Established! November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 303
  • 304. The Handoff Process The handset pilot searcher notices energy from another sector or BTS, meeting any of these criteria: •Neighbor or Remaining Pilot Ec/Io stronger than T_Add •Candidate Pilot just got T_Comp better than an ac tive •An Active Pilot stayed below T_DROP for T_TDROP time BTS Pilot Strength Measurement Message Base Station Acknowledgment Order •Selector arranges channel elements/Walsh codes in requestedForward sectors and begins using them, too. Reverse Traffic Traffic Extended Handoff Direction MessageChannel Channel Mobile Station Acknowledgment Order •Handset verifies which assigned PNs it can now hear. Handoff Completion Message Base Station Acknowledgment Order Neighbor List Update Message Mobile Station Acknowledgment Order The new Handoff condition is now officially Established! November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 304
  • 305. Troubleshooting Access Failures Troubleshooting Access FailuresNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 305
  • 306. Investigating Access FailuresAn access attempt failure can occur at any point in the process: Successful Access Attempt Access probes exhausted (not received Origination Msg ACCESS by system) Access probes exhausted (seen by BTS MS Probing system but ACK not reaching mobile station) PAGING Base Sta. Acknlgmt. Order Ack received by mobile station but FW TFC TFC frames of 000s Channel Assignment Message not seen PAGING Channel Assnmt. Msg. Channel Assignment Message seen at mobile but mobile station does not TFC preamble of 000s RV TFC acquire Forward Traffic Channel FW FC Base Sta. Acknlgmt. Order Mobile station acquires Forward Traffic Mobile Sta. Ackngmt. Order RV TFC Channel but system does not acquire Reverse Traffic Channel FW TFC Service Connect Msg. System acquires Reverse Traffic Svc. Connect Complete Msg RV TFC Channel but Service Connect Message FW TFC Base Sta. Acknlgmt. Order is not seen at mobile station. Call is Established! November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 306
  • 307. Troubleshooting Access Failures & TCCFs Troubleshooting access failures (Traffic Channel Confirmation Failures) can be difficult There are many steps in the access process • Finding which step failed is not easy Rarely, circumstantial evidence points clearly to the problem Usually, it is necessary to debug the process leading up to the access failure • Consider each step in the access process • Get evidence to determine whether this step occurred successfully • Move on to the next step and keep checking steps until the unsuccessful step is found • Determine why this step failed The following slides describe the steps in the access process, where they take place, and some of the factors which may cause them to fail This narrative might be useful as a “template” for organizing your own thinking as you investigate access failures you are tracking! • Go out and capture actual drive tests of failed origination attempts • If possible, also collect system logs (RF call trace, etc.) for the same eventNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 307
  • 308. Troubleshooting Access Failures (1)BTS Steps in the Access Process Troubleshooting Comments Paging Channel Access Channel If the mobile does not hear acknowledgment from the BTS within ACC_TMO, this could mean either: Origination Msg. •The BTS did not hear the mobile Probe #1 •Maybe the mobile collided with another mobile transmitting at the same time Mobile waits to see if the BTS hears and •Maybe mobile was too weak to overcome acknowledges its probe within the time ACC_TMO. the existing reverse noise level at the BTS If not, the mobile must transmit the message again •In either case another probe should solve in another probe, this time PI db. louder. the problem, provided PI is set reasonably Origination Msg. and additional probes are allowed (check the Probe #2 Access Parameters Message to see if Num_Step and the power parameters make Mobile waits again to see if the BTS hears and sense; be sure also the cell size or Access acknowledges its probe within the time ACC_TMO. Channel acquisition search width is set large If not, the mobile must transmit the message again enough and the number of access preamble in another probe, this time PI db. louder. frames is large enough for the cell size) Origination Msg. •The BTS is acknowledging but the mobile cannot Probe #3 hear the acknowledgment •If the mobile can’t hear the BTS The mobile keeps probing until NUM_STEP probes acknowledging, Ec/Io is likely quite poor. If have been sent, then repeats the probe sequence so, check whether this is due to weak signal again until Max_Probe_Sequences have been (poor coverage) or pilot pollution (lots of sent. pilots all weak but no dominant server) Collect system logs if necessary to determine definitely whether the system heard the mobile’s origination or not November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 308
  • 309. Troubleshooting Access Failures (2)BTS The Access Process Troubleshooting Comments Paging Channel Access Channel If this problem happens frequently, the BTS traffic overload must be relieved. Here are some steps to try: One Dreaded Possibility: •Investigate BTS TX hardware to ensure everything is working correctly and properly calibrated, Reorder particularly gain settings in the TX chain •To free up more forward power for traffic channels, try: Mobile beeps and displays “Call Failed - System •Reduce PTXstart (initial traffic channel Busy” DGU) watching for less forward power control overloads. If you go too far, you will notice access failures increase. •Reduce PTXmax (maximum traffic channel DGU) watching for less forward power control overloads. If you go too far, dropped calls will increase. •Reduce sector traffic by reorienting the sectors to more closely balance the load carried by each •Or, add another carrier •Or split cells November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 309
  • 310. Troubleshooting Access Failures (3)BTS The Access Process Troubleshooting Comments Paging Channel Access Channel After hearing the BTS acknowledgment, the mobile will stop probing and wait for further instructions on Base Station the paging channel. Acknowledgment If the mobile does not hear the Channel Assignment Message within 12 seconds, the mobile will beep and display “Call Failed”. Possible causes: •The BTS did not transmit the Channel Assignment Message •Check system logs to see if this was not transmitted. If not transmitted, get troubleshooting help from the system manufacturer -- this should never occur •The BTS did transmit the Channel Assignment Message, but the mobile did not hear it •Was this because the paging channel faded? (Did the Ec/Io drop momentarily)? If Channel Assignment so, see If this is a recurring problem such as Message a coverage hole or severe pilot pollution Finally! The mobile hears the Channel Assignment STOP! Leave the Paging Channel, and don’t Message! transmit again on the access channel. Now it will immediately leave the paging channel The mobile now goes to try to hear the Forward and start trying to hear the new Forward Traffic Traffic Channel. Channel. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 310
  • 311. Troubleshooting Access Failures (4)BTS The Access Process Troubleshooting Comments FWD Traffic Channel REV Traffic Channel The mobile listens to the Walsh Code # given in the 00000000000000000000 Channel Assignment Message. It should hear N5M 00000000000000000000 good frames full of all zeroes within T2M seconds 00000000000000000000 (usually 2 frames in 10 frames). If the mobile does not hear the required number of Mobile beeps and displays “Call Failed” good empty frames, it will beep and give an error message, then reacquire the system. 00000000000000000000 If the mobile hears the required number of good 00000000000000000000 empty frames, it starts transmitting its own 00000000000000000000 “Reverse Traffic Channel Preamble” of empty all- zero frames. If the BTS does NOT hear the mobile’s access preamble within a prescribed delay, it will abort the process and release all the resources, and the mobile will reacquire the system. . This is what Lucent terms a “Traffic Channel Confirmation Failure (TCCF).” Base Station Acknowledgment If the BTS DOES hear the mobile’s access preamble, it will send an acknowledgment. Mobile Station The mobile responds with an acknowledgment, or Acknowledgment maybe even a pilot strength measurement message if it already needs a handoff. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 311
  • 312. Troubleshooting Access Failures (5)BTS The Access Process Troubleshooting Comments FWD Traffic Channel REV Traffic Channel Now that the BTS and mobile see each other on the traffic channels, the next step is service Service Connect Message negotiation. The BTS sends a Service Connect message listing the type and rate set of the vocoder or other primary traffic source. The mobile either accepts the proposal with a Service Connect Service Connect Complete message, or Complete Message counterproposes a different mode. This is still just an ongoing access attempt The BTS acknowledges the Service Connect Base Station Complete message. Acknowledgment Now this is officially a call in progress The call is now officially in progress. If anything happens to interrupt it after this point, that is considered a dropped call. If any of these steps is unsuccessful, the call attempt will probably fail. Suspect RF conditions on the link which was supposed to carry the unsuccessful command. Look at system logs and message logs from mobile drive testing to pin down just what happened. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 312
  • 313. Access Failure/TCCF Troubleshooting Access Attempt Failed Were any probes acknowledged? Forward Power Optmz Fpwr DGUs Yes, No, Yes, Blocking Channel Elements Add chan cards BS Ack Nothing Reorder Rev. Link Noise Identify, fix source Weak Signal/Coverage Hole? Add coverage Paging Channel Strong Fwd interf / pollution? Identify, eliminate faded, lost Is T-1unstable/blocking? Report/repair no Check System Logs. Rev Link Overload? Identify, fix source Was mobile heard? Num_Step, Pwr_Step Ensure reasonable yes appropriate? values yes no Was Channel Assignment Check System Logs. Sector Size, Acq Width Ensure reasonable Message heard? Was CH ASN sent? appropriate? values for cell size yes no System Problem. Software problem Did mobile see N5M good no frames on F-TCH? no Investigate why Resource blocking yes Check System Logs. CH EL initialized OK? Check System Logs. Did yes Rev. Link Noise Identify, fix source BTS see mobile preamble? no yes Init TCH DGU large enough? Raise DGU no F-TFC Channel Weak Signal/Coverage Hole? Improve coverage Did mobile see BS Ack? faded, lost Strong Fwd interf / pollution? Identify, eliminate yes Is T-1unstable/blocking? Report/repair Check System Logs. no Weak Signal/Coverage Hole? Improve coverage Did BTS see mobile Ack? R-TFC Channel Strong Rev Noise? Identify, eliminate faded, lost OK Is T-1unstable/blocking? Report/repairNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 313
  • 314. Reducing Access FailuresIf the base station never sees the mobile’s probes, Access Attemptthe cause is probably coverage-related. If it happensin strong signal areas, suspect BTS hardware. Also Origination Msg ACCESScheck datafill for proper NOM_PWR and PWR_INC.Be sure the BTS datafill access channel acquisition BTS MSand demodulation search windows are adequate. Probing1. If the failures occur in areas where one BTS PAGING Base Sta. Acknlgmt. Orderis dominant, suspect BTS hardware problems. FW TFC TFC frames of 000s2. Plot the access failures to see if they correlatewith areas of BTS overlap. If so, suspect PAGING Channel Assnmt. Msg.forward link problems. This is probable TFC preamble of 000s RV TFCbecause the mobile does not have the normaladvantage it would get from soft handoff on a FW FC Base Sta. Acknlgmt. Ordertraffic channel. During access, it must Mobile Sta. Ackngmt. Order RV TFCsuccessfully demodulate all five BTS messageswithout the benefit of soft handoff. If the FW TFC Service Connect Msg.handset is in an area of multiple BTS overlaps Svc. Connect Complete Msg RV TFCor weak signal, this can be risky. In such cases,try to make the serving BTS more dominant. FW TFC Base Sta. Acknlgmt. OrderAlso check the access/probing parameters. Call is Established! November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 314
  • 315. Troubleshooting Dropped Calls Troubleshooting Dropped CallsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 315
  • 316. Dropped Call Troubleshooting - Mobile Side Just arrived on sync channel! Is this a drop? yes Were there release OK, normal messages? end of call no This is a drop! yes Was the Sync Channel PN Weak Signal/Coverage Hole? Improve coverage Active before the drop? Check Strong Fwd/Rev interference? Identify, eliminate for: no Is T-1unstable/blocking? Report/repair yes Did mobile request Sync CH Why didn’t handoff happen? PN in PSMM before drop? no PN not in neighbor list Add PN to Nbr List! noAdd PN to Neighbor List! Is PN in neighbor list? Weak Signal/Coverage Hole? Add coverage yes FER already too bad? Push earlier no Widen SRCH_WIN_N! Is SRCH_WIN_N adequate? Border configuration problems Debug, reconfigure yes Incr Sector Overlap yes Fast-rising pilot, slow reaction Speed up searcherRepair/Re-initialize Cell! Is cell in “island Mode”? Forward Power Optmz Fpwr DGUs no Blocking Channel Elements Add chan cards Is T-1unstable/blocking? Rev. Link Noise Identify, fix source Is T-1unstable/blocking? Report/repair More information needed. Collect system logs and merge with mobile data, analyze November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 316
  • 317. Investigating Dropped Calls BAD COVERAGE If the radio link fails after the mobile sends the FFER 100% RXL -30 -40 EC/IO 0 TxGa +25 TxPo +23 +10 Service Connect Complete Message then it is 50% -6 -10 +10 0 0 -10 -20 considered a dropped call. Using the signatures 10% 5% 2% -90 -100 -15 -10 -20 -30 -40 -50 0% -110 -20 -25 described earlier, it is possible to recognize and FFER RXL EC/IO TxGa TxPo separate the dropped calls into the categories at BTS Messaging right. FWD. INTERFERENCE FFER RXL EC/IO TxGa TxPo 100% +23 Each category has its own causes and solutions -30 -40 0 -6 +25 +10 +10 0 -10 50% -10 0 -20 Dropped call analysis can consume a 10% -90 -15 -10 -30 -40 5% -100 -20 -50 considerable amount of time. Using good post- 2% 0% FFER -110 RXL -20 EC/IO -25 TxGa TxPo processing analysis tools, the root cause of BTS Messaging some of the drops can be determined from REV. INTERFERENCE mobile data alone. However, there will be FFER 100% RXL -30 EC/IO 0 TxGa +25 TxPo +23 -40 +10 cases where the cause cannot be reliably 50% -6 -10 +10 0 0 -10 -20 confirmed unless system data is also used 10% 5% -90 -15 -10 -20 -30 -40 2% -100 -50 0% -110 -20 -25 FFER RXL EC/IO TxGa TxPo BTS MessagingNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 317
  • 318. Handoff Problems: “Window” Dropped Calls SITUATION 1 Locked to distant Calls often drop when strong neighbors suddenly appear A 12 mo site, can’t see outside the neighbor search 80 mile un one nearby BTS Ch s tai window and cannot be used to ips ns B establish soft handoff. SRCH_WIN_N = 130 BTS Neighbor Search Window BTS A is reference. 1 mi. SRCH_WIN_N should be set BTS B appears (7-80) chips 7 Chips early due to its closer distance. vel to a width at least twice the This is outside the 65-chip window.Tra propagation delay between Mobile can’t see BTS B’s pilot, but its any site and its most distant strong signal blinds us and the call drops. neighbor site SITUATION 2 Remaining Search Window Locked to nearby SRCH_WIN_R should be set A mo site, can’t see 12 un distant one to a width at least twice the BTS 80 mile tai ns propagation delay between Ch s B ips any site and another site SRCH_WIN_N = 130 BTS which might deliver occasional BTS B is reference. 1 mi. BTS A appears (80-7) chips 7 Chips RF into the service area late due to its farther distance. l This is outside the 65-chip window. Trave Mobile can’t see BTS A’s pilot.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 318
  • 319. Optional: Quick Primer on Pilot Search Windows The phone chooses one strong sector and PROPAGATION DELAY “locks” to it, accepting its offset at “face value” SKEWS APPARENT PN OFFSETS and interpreting all other offsets by 33 4 comparison to it Chips Chips In messages, system gives to handset a A BTS B neighbor list of nearby sectors’ PNs BTS Propagation delay “skews” the apparent PN offsets of all other sectors, making them If the phone is locked to BTS A, the seem earlier or later than expected signal from BTS B will seem 29 chips To overcome skew, when the phone searches for a particular pilot, it scans an earlier than expected. extra wide “delta” of chips centered on the If the phone is locked to BTS B, the expected offset (called a “search window”) signal from BTS A will seem 29 chips Search window values can be datafilled later than expected. individually for each Pilot set: There are pitfalls if the window sizes are improperly set • too large: search time increases • too small: overlook pilots from far away • too large: might misinterpret identity of a distant BTS’ signal One chip is 801 feet or 244.14 m 1 mile=6.6 chips; 1 km.= 4.1 chips November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 319
  • 320. Pilot Search Order, Speed, and Implications WINDOW SIZE PILOT SEARCHING IN NESTED LOOPS: IN CHIPS AND DATA UNITS THE CAR ODOMETER ANALOGY Datafill Window The searcher checks pilots in the Value Size (Chips) order they would appear if pasted Active+CandRemaining on the wheels of a car odometer. 4 14 (±7) Neighbor Actives and candidates occupy the 5 20 (±10) fastest-spinning wheel. 6 28 (±14) Neighbors are next, advance one pilot each time Act+cand revolves. 7 40 (±20) Remaining is slowest, advance one 8 60 (±30) pilot each time Neighbors revolve. 9 80 (±40) Actives & candidates have the biggest influence. 10 100 (±50) • Keep window size as small as possible 11 130 (±65) • During soft handoff, this set dominates searcher 12 160 (±80) – Minimize excessive Soft HO! 13 226 (±113) Neighbor set is second-most-important • Keep window size as small as possible 14 320 (±160) • Keep neighbor list as small as possible 15 452 (±226) • But don’t miss any important neighbors! Remaining Set: pay your dues, but get no reward • You must spend time checking them, but the system can’t assign one to youNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 320
  • 321. Treating Drops with Poor-Coverage Symptoms Using a post-processing tool, Drops with weak-signal symptoms display a map of the locations of happened in predicted strong-signal area. Suspect bad BTS hardware. dropped calls that exhibit symptoms of poor coverage • It is particularly useful to be able to overlay the drop locations on a map of predicted or measured signal levels Verify this type of drop is not occurring in good-coverage areas • If so, suspect and investigate hardware at the serving site Coverage related drops occurring in poor-coverage areas are to be expected; additional RF (usually from new BTSs) is the only These drops are probably normal solution except in rare cases due to their locations in a predicted weak-signal area.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 321
  • 322. Treating Drops with Forward-Link Problems The call on sector A dropped here,Plot the data containing the apparently due to interferenceforward-link interference drops on from sector B. Find out why softmaps from your propagation handoff with B did not occur.prediction tool • Use the prediction tool to help identify other strong signals reaching the drop areas A • If the signals are from other CDMA carriers, add their Pilot PNs to the neighbor list B • Resolve any PN conflictsAnother technique is to examine Sync Channel Messagethe dropped call message files p_rev 1, bit_len: 170and identify the BTS from which min_p_rev 1the sync channel message is sid 4139 nid 41received immediately after each pilot_pn 0x164 = 356 ( RMCZ ) lc_state 1ED595B9632drop (this will be the cleanest pilot sys_time 189406BE8the handset sees at that time) lp_sec 13 ltm_off 0x10 (8.0 hours) daylt 0 prat 1 cdma_freq 50November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 322
  • 323. “Optimizable” Dropped Calls: Slow Handoff BTSWhen the mobile is suddenlyconfronted with a strong new signal,or when the signal it is using takes asudden deep fade, it will have poorEc/Io and high forward FER. The callwill drop unless it gets help quickly.Several steps which must occurwithout delay: x • The mobile search correlator must first notice the new pilot and send a PSMM to the system. BTS • The system must set up the soft handoff and notify the mobile. • The mobile must acquire the new signal by locking a fingerNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 323
  • 324. Sources of Delay Causing Slow HandoffEvery step in the handoff process can suffer delay if we’re not careful to control conditions: Mobile search correlator notices new pilot • Window sizes too large, searching is slow • Multi-sector soft handoff already underway, many active pilots, searching is slow… • Interferor not a neighbor, must find in remaining set: slow, DIE! – System cannot currently set up true remaining-set handoffs Mobile reports PSMM to system. • Reverse link noisy, PSMM must be re-requested & repeated System sets up handoff, sends EHDM to mobile • Resource congestion: no TCEs, or other problems • Forward link is noisy, mobile doesn’t hear EHDM, must repeatFortunately, these problems do not have to happen.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 324
  • 325. Auditing System Handoff Setup Time Typical Handoff After the mobile searcher recognizes the Setup Time pilot it needs, the job is only begun 100% • The mobile must send a PSMM to 90% system; it must be received • System must recognize reported PNCumulative Distribution Function 80% phase, set up resources in the 70% appropriate sector • An EHDM must be sent to the 60% mobile, received, acknowledged 50% • Mobile must acknowledge again when handoff implemented 40% Time required for this process can be 30% measured by watching messages 20% • most post-processing tools can show histogram or graph of this delay 10% • if system is healthy, almost all handoffs will happen in <200 msec. 0% and there will be no “stragglers” 0 100 200 300 400 500 Time (milliseconds) November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 325
  • 326. Setting Pilot Search Window Sizes When the handset first powers up, it does an exhaustive search for the best pilot and no windows apply to this process. When finally on the paging channel, the handset learns the window sizes SRCH_WIN_A, N, R and uses them when looking for neighbors both in idle mode and during calls. During a call, when a strong neighbor is recognized, a PSMM is sent requesting soft handoff. The former neighbor pilot is now a candidate set pilot and its offset is precisely remembered and frequently rechecked and tracked by the phone. The window size for active and candidate pilots doesn’t need to be very large, since the searcher has already found them and is tracking them very frequently. We need only enough width to accommodate all multipath components of these pilots. • This greatly speeds up the overall pilot search management! Most post-processing tools deliver statistics on the spread (in chips) between fingers locked to the same pilot. These statistics literally show us how wide the SRCH_WIN_A should be set. Neighbor and Remaining search windows should be set based on intercell distances as described in a preceding slide.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 326
  • 327. Maximum Timing Budget for CDMA CellsThe range of a CDMA cell is normally limited by the attenuation thatoccurs along ordinary propagation paths. Occasionally, a site islocated atop a high mountain or other location from which it can see avery large distance, so large that timing considerations must berecognized. Search windows are the main concern.The BTS uses acquisition and demodulation search windows muchlike the pilot search windows used by the mobile. The maximumsetting is 4095/8 chips (512 chips -1/8 chip). A mobile 38.8 miles fromthe site would be at the edge of this maximum window setting, andcould not originate or be acquired during handoff beyond this distance.The mobile is not restricted on acquiring the system forward channelsbut its pilot search windows are limited to 452 chips. Neighbor pilotscouldn’t be recognized if coming from a cell more than 34.3 milescloser or farther than the cell to which the mobile is locked.The IS-95 and J-Std008 specify a maximum of 350 µsec maximumround trip delay, BTS-Handset. This is a distance of 32.6 miles.General Observation: If your cell radius exceeds 30 miles, be careful.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 327
  • 328. A Word About Soft Handoff for Former AMPS/TDMA Personnel Former AMPS/TDMA optimizers may feel an instinctive obligation to minimize handoff activity, with good reason. In AMPS/TDMA, handoffs involved muting and real risk of a drop. Since the mobile could be served by just one sector at a time, there was pressure to be sure it was the best available sector, but also pressure not to do many handoffs. Ping-pong is unpopular in AMPS/TDMA. In CDMA, there is no muting or audible effect during soft/softer handoff, and there is no pressure to use just the right sector -- if several are roughly as good, use them all, up to 6 at a time. • The noise level on the reverse link actually decreases during soft handoff - by roughly 4 db. - allowing the system to handle from 1.5 to 2 times as many subscribers as otherwise. • The forward link noise does rise, but not to troublesome levels • There is an additional cost for doing soft handoff: each involved BTS must dedicate a TCE channel element to the handoff. However, even if every user is constantly involved in soft handoff, this increases the cost of a BTS a small percentage. So, to former AMPS/TDMA folks, don’t fear. “Use the force, Luke!” And to our GSM friends, “Resistance is futile…...”November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 328
  • 329. How Much Soft Handoff is Normal?How much soft handoff is normal? • Expectations in early CDMA development were for roughly 35% • The level of soft handoff which should be used depends on how much diversity gain can be achieved, and terrain roughness – If the reverse link budget assumed 4 dB soft handoff gain, and propagation decays 35 dB/decade, 42% of the sector’s area is within the last 4 dB. of coverage where soft handoff occurs. – In typical markets, terrain irregularities scatter RF beyond cleanly designed cell edges; soft handoff is typically 50-60% – In rough terrain, proper soft handoff may rise to 70% or moreIn a system not yet well-tuned, soft handoff may be clearly excessive • The main cause is usually excessive RF overlap between cells • RF coverage control is the most effective means of reducing and managing soft handoff (BTS attenuation, antenna downtilting) • Thresholds T_ADD and T_DROP can be adjusted to reduce soft handoff, but this penalizes mobiles that need soft handoff to escape interference from the excessively overlapping sites Controlling soft handoff percentage with T_ADD and T_DROP is like limiting allowed hospital days for various illnesses. Works, but some patients may drop.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 329
  • 330. Dangerous Environments The CDMA handset is designed with a digital “rake receiver” including three correlators (“fingers”) which can demodulate signals from up to three sectors simultaneously, combining and using the energy from all three to improve reception. Implications: • If One dominant signal: this is a good situation; the three fingers will be looking for resolvable multipath components; good diversity • If Two usable signals: good situation; soft handoff & diversity • If Three usable signals: good situation; soft handoff & diversity • If Four roughly equal signals: workable but not ideal. Three best signals are demodulated; other remains an interferor. 3 vs 1 • If Five roughly equal signals: probably workable but not good. Three best are demodulated; remaining two are interferors. 3 vs 2 • If Six roughly equal signals: very frightening. Three best signals are demodulated; three remaining signals are interferors. 3 vs 3 The system can provide up to 6-way soft handoff, but anything above three-way is an indication that there is too much RF coverage overlap. More than three-way soft handoff should be the notable exception rather than the rule.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 330
  • 331. Identifying Causes of Excessive Soft Handoff RF Drive Test data (preferred) or Propagation Prediction runs (second choice) can be used to identify the excessive coverage overlaps which cause soft handoff. Suggested Procedure: • Use a post processing tool to display all locations where a sector has strongest rake finger status, or • Use a propagation prediction tool to show all locations where a sector is “best server” • Draw a curve through all the adjacent surrounding sites • If more than 15% of the best-finger or best-server points lie outside this line, this sector’s coverage is excessive. • Reduce signal levels by at least 8 dB. through attenuation or downtilt and re-examine either using prediction or re-driving • Be aware that as strong unwanted signals are reduced or removed by this process, other signals formerly degraded may become apparent and also require similar treatment. This is therefore a somewhat iterative process.November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 331
  • 332. Grooming Neighbor Lists Earlier we described a general technique for creating initial neighbor lists. During initial optimization, and especially after your system generates data from commercial traffic, you’ll want to revisit and groom the neighbor lists. Use your post-processing tool to show you all handoff transitions requested by mobiles on a per-sector basis. If you don’t have a fancy software tool, you can still do it with fairly simple scripts parsing captured pilot strength measurement messages. For each sector, examine the statistics in conjunction with the Planet equal power boundaries plot. Consider removing any pilots that are currently in the neighbor list but have less than 1% of the handoff transitions. However, make sure that is not a consequence of no test drives being made across a particular sector boundary (for example, do not remove adjacent sectors of a sectored site). Consider adding pilots that are not currently in the neighbor list but have greater than 5% of the handoff transitions. Remember, though, that the goal is to keep neighbor lists to a minimum (see below) so avoid adding sites that are obviously not immediate neighbors of the serving cell (i.e. try to make use of the composite neighbor list as much as possible).November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 332
  • 333. TX Gain Adjust as a Per-Site Debugging Tool Collect Transmit Gain Adjust Statistics For an unloaded system, the average should be -7 to -12 db. and should be fairly constant throughout the coverage area Look for big “jumps” in TX GA from sector to sector. Look for hardware problems (antennas OK, RX noise figure OK?, etc.) If you see values generally outside the range above uniformly across the coverage area, look at the BS Eb/Nt. It should be 5-9 dB for mobile systems, or 3-4 dB. for fixed wireless access. Other parameters can have similar uses; compare and study. Typical Mobile Station Transmit Gain Adjust 0 dB -10 dB -20 dB Time, SecondsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 333
  • 334. Course RF200 CDMA2000 1xRTT CDMA2000 1xRTT System Performance Optimization System Performance OptimizationNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 334
  • 335. The Big Picture: IP Data Environment PDSN/Foreign Agent BackboneInternet Network VPNs T SECURE TUNNELS T CDMA RF Environment IP Data Environment PDSN Authentication Home Agent Authorization AAA R-P Interface Accounting •Coverage Holes BTS •Pilot Pollution •Missing Neighbors PSTN v SEL CE •Fwd Pwr Ovld t1 t1 t1 •Rev Pwr Ovld Switch •Search Windows (C)BSC/Access Manager Wireless •Island Cells Mobile Device Traditional Telephony CDMA IOS PPP •Slow Handoff 1xRTT services may include both traditional circuit-switched voice and new fast IP data connections • A Users link is in multiple jeopardy, both radio and packet worlds Radio environment portion • Problems: FER, drops, access failures, capacity woes • Causes: mainly in the RF world, because of mainly RF problems Packet environment • Problems: Setup failures, dropped connections, low throughput • Causes: could be IP-related, or could be RF related November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 335
  • 336. Optimization Issues Network Design and Configuration • Coverage holes, excessive coverage overlap Call Processing Problems due to Misconfiguration • Neighbor Lists • Search Windows • Power control parameters Physical Problems/Hardware Problems • Mismatched multicarrier sector coverage Capacity Issues • Forward and Reverse Power Control Overload • Physical resource congestion – Channel elements, packet pipes – IP network congestion Managing A New Dimension: circuit-switched and IP traffic blend • QoS-related competitive issuesNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 336
  • 337. Optimizing in Two Worlds Circuit-Switched Voice Traffic • Some operators are implementing 1xRTT mainly to gain capacity for additional voice traffic • Their optimization techniques remain about the same as for 2G voice networks today – Keep network adequately dimensioned – Control RF environment – Monitor and manage capacity utilization IP Data Traffic • Operators adding IP traffic to upgraded voice networks • Conventional optimization techniques are still appropriate for general RF environment and circuit-switched network performance • New IP and QoS issues require a new optimization focus for the blended total network – IP performance depends on both IP and RF factors – IP and Voice performance involve competitive tradeoffsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 337
  • 338. Managing Forward Link Sector Loading vs. Time Sector Maximum TX Power, Maximum Throughput Sector Total TX Power or Throughput Packet Data Traffic Voice Traffic Time, Seconds Both voice and data traffic loads a sector, driving up transmit power • Voice calls are typically given higher priority than data • MAC-layer throttling holds lower-priority data sessions off until there is enough free power available November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 338
  • 339. Starting Optimization on a New System RF Coverage Control • try to contain each sector’s coverage, avoiding gross spillover into other sectors • tools: PN Plots, Handoff State Plots, Mobile TX plots Neighbor List Tuning • try to groom each sector’s neighbors to only those necessary but be alert to special needs due to topography and traffic • tools: PSMM data from mobiles; propagation prediction Search Window Settings • find best settings for SRCH_WIN_A, _N, _R • especially optimize SRCH_WIN_A per sector using collected finger separation data; has major impact on pilot search speed Access Failures, Dropped Call Analysis • finally, iterative corrections until within numerical goals IP Data Performance Assessment • identify latency and throughput issuesGetting these items into shape provides a solid baseline and foundation from which future performance issues can be addressed. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 339
  • 340. Performance Monitoring/Growth Management Benchmark Existing Performance • Dropped Call %, Access Failure %, traffic levels Identify Problem Cells and Clusters • weigh cells and clusters against one another Look for signs of Overload • TCE or Walsh minutes -- excessive ? Soft handoff excessive? • Required number of channel elements -- excessive? • Forward Power Overloads: Originations, Handoffs blocked Traffic Trending and Projection • track busy-hour traffic on each sector; predict exhaustion • develop plan for expansion and capacity relief – split cells, multi-sector expansions, multiple carriersThese steps must be continuously applied to guide needed growth. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 340
  • 341. #6 Indicator: Data Latency IP Data Environment PDSN/Foreign Agent BackboneInternet Network VPNs T SECURE TUNNELS T CDMA RF Environment IP Data Environment PDSN Authentication Home Agent Authorization AAA R-P Interface Accounting •Coverage Holes BTS •Pilot Pollution •Missing Neighbors PSTN v SEL CE •Fwd Pwr Ovld t1 t1 t1 •Rev Pwr Ovld Switch •Search Windows (C)BSC/Access Manager Wireless •Island Cells Mobile Device Traditional Telephony CDMA IOS PPP •Slow Handoff Latency can occur because of RF channel congestion or from IP network causes • RF overload can delay availability of supplemental channels • IP network congestion can delay availability of packets Ping and loopback tests with local PDSN and servers can identify whether problem is in backbone network Does latency correlate with independent evidence of RF congestion? November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 341
  • 342. #7 Indicator: Data Throughput IP Data Environment PDSN/Foreign Agent BackboneInternet Network VPNs T SECURE TUNNELS T CDMA RF Environment IP Data Environment PDSN Authentication Home Agent Authorization AAA R-P Interface Accounting •Coverage Holes BTS •Pilot Pollution •Missing Neighbors PSTN v SEL CE •Fwd Pwr Ovld t1 t1 t1 •Rev Pwr Ovld Switch •Search Windows (C)BSC/Access Manager Wireless •Island Cells Mobile Device Traditional Telephony CDMA IOS PPP •Slow Handoff Throughput can be limited by RF and IP causes • Traditional RF problems limit capacity of the channel • Congestion in the IP network can limit speed of data available Does low throughput correlate with independent RF indicators? Does low throughput correlate with independent IP pings and tests? November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 342
  • 343. Course RF200 System-Side 1xRTT Tools System-Side 1xRTT ToolsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 343
  • 344. Basic Philosophy of System Data Each network manufacturer has its own data sets and counters • Access failures, TCCFs, blocks, drops, failed handoffs • These counters are normally available in 2G-only, 3G-only, and total categories • Additional new statistics are available for IP traffic The basic philosophy of system data analysis is to analyze and discriminate within the available data • Identify and rank existing sectors based on – Traffic levels – raw failures/blocks/drops – percentage failures/blocks/drops • Benchmark and track incremental changes • Investigate all significant problems uncovered – Drive-testing or data testing may be required In-Class activity: view manufacturer documentation and examplesNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 344
  • 345. Information on System-Side Statistics Lucent • Technical Reference: Watchmark Prospect for Lucent, v17.0 Nortel • 411-2131-814 DMS-MTX Operational Measurements Reference Manual version v. 12.02 June, 2001 • 411-2131-900 DMS-MTX Operational Measurements Quick Reference Guide Motorola • “Performance Analysis 2.16.0” v O , Motorola Inc., January 2002. • “1x network Performance Matrix” v. 0.1, Motorola Inc., April 2001. • “CDMA 2000 – 1x Voice and Data – Cellular Application Note” , v. 1.1 – Draft; Motorola Inc. • “Impact on CDL and CFC in Version 2.16.0” v.1.4, Part No. 8700SCRP20GCDLCFC-D, Motorola Inc., August 2001 • “CFC Resolution Document” v. 1.3, Motorola Inc Performance Analysis 2.16.0” v O , Motorola Inc., January 2002November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 345
  • 346. Course RF200 Mobile-Side 1xRTT Tools Mobile-Side 1xRTT ToolsNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 346
  • 347. Sources of CDMA Data and Tools for Processing CDMA NETWORK EQUIPMENT HANDSET Switch CBSC BTS SLM CM GPSR IS-95/J-STD-8 GPSR Messages BSM CDSU CDSU DISCO TFU1 TFU1Switch Data DMS-BUS DISCO 1 CDSU Ch. Card ACC CDSU CDSU pegs, logsLPP ENET LPP CDSU System DISCO 2 Σα Txcvr A Internal Messages B CDSU RFFE A DTCs CDSU Σβ Txcvr RFFE B Handset SBS CDSU Σχ Txcvr C RFFE C Vocoders Messages PC-based IOC Selectors Mobile Data Capture Tools IS-95/J-STD-008 Messages Unix-based, Various PC-based PC-based External Data Analysis Mobile Data Analysis Post-Processing Post-Processing Tools Tools Tools CDMA optimization data flows from three places: • Switch • CDMA peripherals (CBSC & BTS) • Handset Each stream of data has a family of software and hardware tools for collection and analysis November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 347
  • 348. CDMA Field Test Tools Field Collection Tools using Handset Data PN Scanners Motorola Qualcomm Agilent Berkeley (formerly HP) Varitronics Grayson Agilent SAFCO (formerly HP) Grayson Qualcomm Comarco LCC Safeco Willtech There are many commercial CDMA field test tools Characteristics of many test tools: • capture data from data ports on commercial handsets • log data onto PCs using proprietary software • can display call parameters, messaging, graphs, and maps • store data in formats readable for post-processing analysis • small and portable, easy to use in vehicles or even on foot A few considerations when selecting test tools: • does it allow integration of network and mobile data? • Cost, features, convenience, availability, and support • new tools are introduced every few months - investigate!November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 348
  • 349. Grayson’s new Invex3G Tool In 1Q2001 Grayson introduced its new Invex3G tool, with new features • 100 MB ethernet connection to PC • the eight card slots can hold receivers or dual-phone cards • there’s also room for two internal PN scanners • Multiple Invex units can be cascaded for multi-phone load-test applications • Cards are field-swappable - Users can reconfigure the unit in the field for different tasks without factory assistanceNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 349
  • 350. Grayson Invex Playback Example 76.8 kb/sThis mobile is in a 4-way soft handoff(four green FCH walsh codesassigned) in the middle of a downlinkSCH burst. Notice walsh code #2, 8chips long, is assigned as an SCHbut only on one sector, and thedownlink data speed is 76.8kb/s. November, 2004 (c) 2004 Scott BaxterTechnical Introduction to Wireless -- ©1997 Scott Baxter - V0.0 RF200 v4.0 350
  • 351. Grayson Invex Playback Example 153.6 kb/sThis mobile is in a 2-way soft handoff(two green FCH walsh codesassigned) in the middle of a downlinkSCH burst. Notice walsh code #3, 4chips long, is assigned as an SCHbut only on one sector, and thedownlink data speed is 153.6kb/s. November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 351
  • 352. Grayson Invex Playback Example F-SCH rates 153.6 kbps; R-SCH 76.8kbps CDMA Status PN Scanner DataCurrent Data Task Status Layer-3 Messages November, 2004 (c) 2004 Scott BaxterTechnical Introduction to Wireless -- ©1997 Scott Baxter - V0.0 RF200 v4.0 352
  • 353. WillTech Tools Blue Rose platform can manage multiple phones and collect data • Internal processor manages test operations independently for stand- alone operation • Internal PCMCIA flash card provides storage • An external PC can display collected data during or after data collectionNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 353
  • 354. Agilent Drive-Test Tools Agilent offers Drive-Test tools • Serial interfaces for up to four CDMA phones • A very flexible digital receiver with several modes PN Scanner • Fast, GPS-locked, can scan two carrier frequencies Spectrum Analyzer • Can scan entire 800 or 1900 mHz. Bands Base-Station Over-Air Tester (BOAT) • Can display all walsh channel activity on a specific sector • Useful for identifying hardware problems, monitoring instantaneous traffic levels, etc. Post-Processing tool: OPAS32November, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 354
  • 355. IS-95 Busy Sector Snapshot of Walsh UsageNovember, 2004 RF200 v4.0 (c) 2004 Scott Baxter RF200 - 355
  • 356. 153,600 307200 76,800 38,400 19,200 9,600 4,800 2,400 Code# Code# Code# 31 Code# Code# Code# sps sps sps sps sps sps 127 63 19.2k 63 38.4k 95 ksps 76.8 31 19.2k RF200 - 356 15 31 153.6 ksps 111 47 19.2k 15 47 F-SCH 38.4k 79 15 19.2k 15 7 119 38.4k 55 19.2k 23 55 87 ksps 76.8 23 19.2k 23 7 307.2 ksps 103 Walsh code’s parents and children. Remember, we cannot use any Walsh code if 39 19.2k 39 F-SCH 38.4k 7 This way of arranging Walsh codes is called “bit reversal order”. It shows eachForward Link Walsh Codes in 1xRTT 71 7 Paging 7 7 3 123 59 19.2k 27 59 38.4k 91 ksps 76.8 27 19.2k 11 27 153.6 ksps 107 43 19.2k 11 43 F-SCH 38.4k 75 11 19.2k 11 3 115 51 19.2k 19 51 38.4k 83 ksps 76.8 19 19.2k 19 3 another Walsh code directly above it or below it is in use. 99 35 19.2k 35 38.4k 3 67 3 Paging 3 3 125 61 19.2k 29 61 38.4k 93 ksps 76.8 29 19.2k 13 29 153.6 ksps 109 45 19.2k 13 45 F-SCH 38.4k 77 13 19.2k 13 5 117 53 19.2k 21 53 38.4k 85 RF200 v4.0 (c) 2004 Scott Baxter ksps 76.8 21 19.2k 21 5 101 37 19.2k 37 38.4k 5 69 5 Paging 5 5 1 121 57 19.2k 25 57 38.4k 89 ksps 76.8 25 19.2k 25 9 105 41 19.2k 41 38.4k 9 73 9 19.2k 9 1 113 49 19.2k 17 49 38.4k 81 17 19.2k 18 1 97 33 19.2k 33 1 65 1 Paging 1 126 62 19.2k 30 62 38.4k 94 ksps 76.8 30 19.2k 14 30 153.6 ksps 110 46 19.2k 14 46 F-SCH 38.4k 78 14 19.2k 14 6 118 54 19.2k 22 54 38.4k 86 ksps 76.8 22 19.2k 22 6 307.2 ksps 102 38 19.2k 38 F-SCH 38.4k 6 70 6 PCH 6 6 2 122 58 19.2k 26 58 38.4k 90 ksps 76.8 26 19.2k 10 26 153.6 ksps 106 42 19.2k 10 42 F-SCH 38.4k 74 10 19.2k 10 2 114 50 19.2k 18 50 38.4k 82 ksps 76.8 18 19.2k 18 2 98 34 19.2k 34 38.4k 2 66 2 PCH 2 2 124 60 19.2k 28 60 38.4k 92 ksps 76.8 28 19.2k 12 28 153.6 ksps 108 44 19.2k 12 44 F-SCH 38.4k 76 12 19.2k 12 November, 2004 4 116 52 19.2k 20 52 38.4k 84 ksps 76.8 20 19.2k 20 4 100 36 19.2k 36 38.4k 4 68 4 PCH 4 4 0 120 56 19.2k 24 56 38.4k 88 ksps 24 19.2k 76.8 24 8 104 40 19.2k 40 38.4k 8 72 8 19.2k 8 0 112 QPCH 48 16 48 QPCH 80 QPCH 16 16 TX Div PIlot 0 96 32 Sync 32 Code# 0 64 0 Pilot 0 Code# Code# Code# Code# Code# 4 chips 8 chips 16 chips 32 chips 64 chips 128 chips
  • 357. 153,600 307200 76,800 38,400 19,200 9,600 4,800 2,400 Code# Code# Code# 31 Code# Code# Code# sps sps sps sps sps sps 127 63 19.2k 63 38.4k ??????? 95 ksps 76.8 31 19.2k RF200 - 357 15 31 153.6 ksps But if the users are highly mobile, forward power may exhaust at typically 30-40 users. 111 47 19.2k 15 47 F-SCH 38.4k 79 15 19.2k 15 7 In fixed-wireless or “stadium” type applications, all walsh codes may be usable. 119 38.4k 55 19.2k 23 55 87 ksps 76.8 23 19.2k 23 7 307.2 ksps 103 39 19.2k 39 F-SCH 38.4k 7 71 7 19.2k 7 3 123 59 19.2k Pilot, Paging Sync, up to 61 Voice Users 27 59 38.4k 91 ksps 76.8 27 19.2k 11 27 153.6 ksps 107 43 19.2k 11 43 F-SCH 38.4k 75 11 19.2k 11 3 115 51 19.2k 19 51 38.4k 83 ksps 76.8IS-95 Today Typical Usage: 19 19.2k 19 3 99 35 19.2k 35 38.4k 3 67 3 19.2k 3 125 61 19.2k 29 61 38.4k 93 ksps 76.8 29 19.2k 13 29 153.6 ksps 109 45 19.2k 13 45 F-SCH 38.4k 77 13 19.2k 13 5 117 53 19.2k 21 53 38.4k 85 RF200 v4.0 (c) 2004 Scott Baxter ksps 76.8 21 19.2k 21 5 101 37 19.2k 37 38.4k 5 69 5 19.2k 5 1 121 57 19.2k 25 57 38.4k 89 ksps 76.8 25 19.2k 25 9 105 41 19.2k 41 38.4k 9 73 9 19.2k 9 1 113 49 19.2k 17 49 38.4k 81 17 19.2k 18 1 97 33 19.2k 33 1 65 1 Paging 1 126 62 19.2k 30 62 38.4k 94 ksps Traffic Channels 76.8 30 19.2k 14 30 153.6 ksps 110 46 19.2k Voice or Data 14 46 F-SCH 38.4k 78 14 19.2k 14 6 9.6k/14.4k 118 54 19.2k 22 54 38.4k 86 ksps 76.8 22 19.2k 22 6 307.2 ksps 102 38 19.2k 38 F-SCH 38.4k 6 70 6 19.2k 6 2 122 58 19.2k 26 58 38.4k 90 ksps 76.8 26 19.2k 10 26 153.6 ksps 106 42 19.2k 10 42 F-SCH 38.4k 74 10 19.2k 10 2 114 50 19.2k 18 50 38.4k 82 ksps 76.8 18 19.2k 18 2 98 34 19.2k 34 38.4k 2 66 2 19.2k 2 124 60 19.2k 28 60 38.4k 92 ksps 76.8 28 19.2k 12 28 153.6 ksps 108 44 19.2k 12 44 F-SCH 38.4k 76 12 19.2k 12 November, 2004 4 116 52 19.2k 20 52 38.4k 84 ksps 76.8 20 19.2k 20 4 100 36 19.2k 36 38.4k 4 68 4 19.2k 4 0 120 56 19.2k 24 56 38.4k 88 ksps 24 19.2k 76.8 24 8 104 40 19.2k 40 38.4k 8 72 8 19.2k 8 0 112 QPCH 48 19.2k 16 48 QPCH 38.4k 16 19.2k 80 QPCH 16 TX Div PIlot 0 96 32 Sync 32 Code# 0 64 0 Pilot 0 Code# Code# Code# Code# Code# 4 chips 8 chips 16 chips 32 chips 64 chips 128 chips
  • 358. 153,600 307200 76,800 38,400 19,200 9,600 4,800 2,400 Code# Code# Code# 31 Code# Code# Code# sps sps sps sps sps sps 127 63 19.2k 63 38.4k ?? 95 ksps 76.8 31 19.2kMixed IS-95 / 1xRTT RC3 Voice Typical Usage: IS-95. The BTS will probably have enough forward power to carry calls on all 61 walsh codes! RF200 - 358 15 31 153.6 ksps 111 47 19.2k 15 47 F-SCH 38.4k FCHs of 1xRTT RC3 users consume less power, so more total users are possible than in 79 15 19.2k 15 7 119 38.4k 55 19.2k 23 55 87 ksps 76.8 23 19.2k 23 7 307.2 ksps 103 39 19.2k 39 F-SCH 38.4k 7 71 7 19.2k 7 3 123 59 19.2k 27 59 38.4k 91 ksps 76.8 Pilot, Paging Sync, up to 61 Voice Users 27 19.2k 11 27 153.6 ksps 107 43 19.2k 11 43 F-SCH 38.4k 75 11