120319 m2m-tutorial-dohler-alonso-ec-2012-final-24630

6,047 views

Published on

Published in: Technology, Business
0 Comments
4 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
6,047
On SlideShare
0
From Embeds
0
Number of Embeds
1,473
Actions
Shares
0
Downloads
312
Comments
0
Likes
4
Embeds 0
No embeds

No notes for slide

120319 m2m-tutorial-dohler-alonso-ec-2012-final-24630

  1. 1. Machine-to-Machine for Smart Grids and Smart Cities Technologies, Standards, and Open Problems Dr. Mischa Dohler, CTTC, Barcelona, Spain Dr. Jesús Alonso-Zárate, CTTC, Barcelona, Spain EW 2012, Poznań, Poland 17 April 2012© 2012 Mischa Dohler and Jesús Alonso-Zárate 1© slide template is copyright of Orange
  2. 2. Disclaimer Besides CTTC’s, many third party copyrighted material is reused within this tutorial under the fair use approach, for sake of educational purpose only, and very limited edition. As a consequence, the current slide set presentation usage is restricted, and is falling under usual copyrights usage. Thank you for your understanding!© 2012 Mischa Dohler and Jesús Alonso-Zárate 2
  3. 3. Machine-to-Machine Definition  Machine-to-Machine (M2M) means no human intervention whilst devices are communicating end-to-end.  This leads to some core M2M system characteristics:  support of a huge amount of nodes  seamless domain inter-operability  autonomous operation  self-organization  power efficiency  reliability  etc, etc© 2012 Mischa Dohler and Jesús Alonso-Zárate © ETSI 3
  4. 4. Machine-to-Machine Vision(s)  Different Visions of M2M:  WWRF [2007-10]: 7 Trillion devices by 2017  Market Study [2009]: 50 Billion devices by 2010  ABI Research [2010]: 225 Million cellular M2M by 2014  Numbers differ significantly, with WWRF prediction implying:  ... 7,000,000,000,000 (7 Trillion) devices by 2017 ...  ... are powered by (in average) AA battery of approx 15kJ ...  ... this requires 100,000,000,000,000,000 (100 Quadrillion) Joules ...  Sanity Check:  1GW nuclear power plant needs to run for more than 3 years to sustain this  1000 devices per person in average at any time  It is important to get this vision and these numbers right!© 2012 Mischa Dohler and Jesús Alonso-Zárate 4
  5. 5. Overview of Tutorial 1. M2M Introduction 1. A Quick Introduction 2. ROI, Markets & Cellular Market Shares 3. M2M in Smart Cities 4. M2M in Smart Grids 2. Capillary M2M 3. Cellular M2M 1. Quick Intro to Capillary M2M 1. Introduction to Cellular M2M 2. Academic WSN Research 2. M2M in Current Cellular Networks 3. Proprietary M2M Solutions 3. M2M Cellular Standardization Activities 4. Standardization Efforts Pertinent to M2M 4. Cellular M2M Business 4. Concluding Observations 1. Conclusions 2. Opportunities & Trends© 2012 Mischa Dohler and Jesús Alonso-Zárate 5
  6. 6. Overview of M2M© 2012 Mischa Dohler and Jesús Alonso-Zárate 6
  7. 7. 1.1 A Quick Introduction© 2012 Mischa Dohler and Jesús Alonso-Zárate 7
  8. 8. Quick Intro  Machine – To – Machine:  device (water meter) which is monitored by means of sensor [in “uplink”]  device (valve) which is instructed to actuate [in “downlink”]  keywords: physical sensors and actuators; cost  Machine – To – Machine:  network which facilitates end-to-end connectivity between machines  composed of radio, access network, gateway, core network, backend server  keywords: hardware; protocols; end-to-end delay and reliability; cost  Machine – To – Machine:  device (computer) which extracts, processes (and displays) gathered information  device (computer) which automatically controls and instructs other machines  keywords: middleware, software, application; cost© 2012 Mischa Dohler and Jesús Alonso-Zárate 8
  9. 9. M2M End-to-End Network  Access Network – connecting the sensors & actuators:  “wired” (cable, xDSL, optical, etc.)  wireless cellular (GSM, GPRS, EDGE, 3G, LTE-M, WiMAX, etc.)  wireless “capillary”/short-range (WLAN, ZigBee, IEEE 802.15.4x, etc.)  Gateway – connecting access and backhaul/core networks:  network address translation  packet (de)fragmentation; etc.  Core/Backend/Internet Network – connecting to computer system:  IPv6-enabled Internet© 2012 Mischa Dohler and Jesús Alonso-Zárate 9
  10. 10. M2M Access Networks [1/2]  Connecting your smart meters through 4 example access methods: CAPILLARY - WIRED CAPILLARY - CELLULAR CELLULAR xDSL GATEWAY© 2012 Mischa Dohler and Jesús Alonso-Zárate 10
  11. 11. M2M Access Networks [2/2]  Wired Solution – dedicated cabling between sensor - gateway:  pros: very, very reliable; very high rates, little delay, secure, cheap to maintain  cons: very expensive to roll out, not scalable  Wireless Cellular Solution – dedicated cellular link:  pros: excellent coverage, mobility, roaming, generally secure  cons: expensive rollout, not cheap to maintain, not power efficient, delays  Wireless Capillary Solution – shared short-range link/network:  pros: cheap to roll out, generally scalable, low power  cons: not cheap to maintain, poor range, low rates, weaker security, large delays  (Wireless) Hybrid Solution – short-range until cellular aggregator:  pros: best tradeoff between price, range, rate, power, etc.  cons: not a homogenous and everything-fits-all solution© 2012 Mischa Dohler and Jesús Alonso-Zárate 11
  12. 12. Timeline of M2M  Origin of term “Machine-to-Machine”:  Nokia M2M Platform Family [2002] = Nokia M2M Gateway software + Nokia 31 GSM Connectivity Terminal + Nokia M2M Application Develop. Kit (ADK) past presence near future far future SCADA, >1980 WIRED Maingate, 1998 Nokia M2M, 2002 CELLULAR (also Ericsson) WSN, >1990 CAPILLARY Coronis, 2002 HYBRID© 2012 Mischa Dohler and Jesús Alonso-Zárate 12
  13. 13. Novelty of Wireless M2M  …© 2012 Mischa Dohler and Jesús Alonso-Zárate 13
  14. 14. Challenge of Wireless M2M Today  Challenges for capillary community:  reliability: despite license-exempt bands  range: multihop/mesh is a must  delays: minimize end-to-end delay (due to multihop)  security: suitable security over multiple hops  standards: lack of standardization across layers  Challenges for cellular community:  nodes: management of huge amounts sending small packets  rates: fairly low and rather uplink from small packets  power: highly efficient (must run for years)  delays: quick ramp-up after sleep  application: don’t disturb existing ones  Is this possible?© 2012 Mischa Dohler and Jesús Alonso-Zárate 14
  15. 15. 1.2 ROI, Markets & Cellular Market Shares© 2012 Mischa Dohler and Jesús Alonso-Zárate 15
  16. 16. ROI - Cost of Wireless M2M of Wireless The Promise $ reduced wiring cost wired cost cellular M2M installation, 90%? connection, capillary commissioning M2M sensors computation & communication time© 2012 Mischa Dohler and Jesús Alonso-Zárate 16
  17. 17. Popular M2M Markets Building  Telemetry Automation Smart City Smart Grids Industrial Automation© 2012 Mischa Dohler and Jesús Alonso-Zárate 17
  18. 18. Growing Cellular M2M Market © B. Tournier, Sagemcom, EXALTED Kick-off Meeting, Barcelona, 14 Sept 2010  Predictions on M2M LTE:  minor market until 2014  2.5% (1.7M) of total M2M market  LTE module = twice 3G cost  Predictions on Automotive:  primary market on M2M cellular  unique (short-term) market for M2M LTE© 2012 Mischa Dohler and Jesús Alonso-Zárate 18
  19. 19. 1.3 M2M in Smart Cities© 2012 Mischa Dohler and Jesús Alonso-Zárate 19
  20. 20. Situation Today  These months will be remembered for:  running out of addressing space  a few months ago we run out of IPv4 addresses  running out of living space  as of a few days ago, we are 7bn people  Humanity point of view:  1 out of 2 is living in cities today; impact onto people’s health is enormous  e.g. 2 Million people are estimated to die annually due to pollution  Political point of view:  politicians have hence become very susceptible to this topic  politicians are eyeing ICT technologies as a possible remedy  “smart syndrome”  Market point of view:  >$100bn per year in 2020 with >$20bn annual spendings  Technology point of view:  technology players are hence trying to enter this market (IBM, Cisco, HP, Oracle)© 2012 Mischa Dohler and Jesús Alonso-Zárate 20
  21. 21. Smart City Rollout Phases PHASE 1: Revenue and Useful PHASE 2: Useful to Public PHASE 3: the rest 2010 2015 2020 Smart Efficient City Efficient City Townhall© 2012 Mischa Dohler and Jesús Alonso-Zárate 21
  22. 22. Smart City Rollout Phase Examples PHASE 1: Revenue and Useful: Smart Parking Smart Street Lightening Smart Litter Bins PHASE 2: Useful to Public: Smart Traffic Flow Pollution Monitoring PHASE 3: the rest: AR (gaming), etc© 2012 Mischa Dohler and Jesús Alonso-Zárate 22
  23. 23. Smart City Stakeholders (abstraction)© 2012 Mischa Dohler and Jesús Alonso-Zárate © PPP OUTSMART 23
  24. 24. ICT Technologies as Enablers  ICT arena is changing very quickly:  about a decade ago, headlines were dominated by companies like Vodafone, Orange, Telefonica, Nokia, Ericsson, Siemens, etc  today, headlines are only dominated by companies like Google, Apple, Facebook  ICT infrastructures and technologies have been sidelined to facilitators/enablers!  A few additional observations from these latest trends:  whoever provides only hardware & infrastructure is loosing out on the long run  being close to the user (or to the problem) is paramount (see Apple’s iPhone)  allowing for true scalability via ability for 3rd parties to capitalize on entire system, i.e. hardware and software and services, is key (see App Store concept which Steve Jobs by the way was against)  having a hand on the data is absolute key (see Google who search, but also IBM who store, Cisco who route, etc)© 2012 Mischa Dohler and Jesús Alonso-Zárate 24
  25. 25. Smart City Technology Platform Internet Improve Efficiency Smart City Offer New Services Crowdsourcing Operating System Sensor Streams Power Applications© 2012 Mischa Dohler and Jesús Alonso-Zárate 25
  26. 26. Wireless M2M Technologies Machine-To-Machine (M2M) Smart City Technologies Low Cost Capillary M2M Low Energy Cellular M2M Low Env. Footprint© 2012 Mischa Dohler and Jesús Alonso-Zárate 26
  27. 27. Yesterday’s M2M Smart City Vision © Northstream© 2012 Mischa Dohler and Jesús Alonso-Zárate 27
  28. 28. Today’s M2M Smart City Reality© 2012 Mischa Dohler and Jesús Alonso-Zárate 28
  29. 29. 1.4 M2M in Smart Grids© 2012 Mischa Dohler and Jesús Alonso-Zárate 29
  30. 30. Smart Grid Vision  Historical Smart Grid Developments:  EU initiated the smart grid project in 2003  Electric Power Research Institute, USA, around 2003  US DOE had a Grid 2030 project, around 2003  NIST is responsible as of 2007  Obama’s “National Broadband Plan” [March 2010]  Mission of ICT in Smart Grids:  enable energy efficiency  keep bills at both ends low  minimize greenhouse gas emissions  automatically detect problems and route power around localized outages  accommodate all types and volumes of energy, including alternative  make the energy system more resilient to all types of failures© 2012 Mischa Dohler and Jesús Alonso-Zárate 30
  31. 31. Reduce Waste & Dependency ... [National Broadband]© 2012 Mischa Dohler and Jesús Alonso-Zárate 31
  32. 32. ... with Smart Grids Macro Storage Hydro Power Plant Nuclear Power Plant Re- newable Power Power Usage Micro Storage Macro Micro Smart Grid Smart Grid Solar Field Power Plant Macro Storage© 2012 Mischa Dohler and Jesús Alonso-Zárate 32
  33. 33. Future Energy Landscape [© CEN-CENELEC-ETSI 2011] 33© 2012 Mischa Dohler and Jesús Alonso-Zárate
  34. 34. Microgrids Play Central Role [© CENER] 34© 2012 Mischa Dohler and Jesús Alonso-Zárate
  35. 35. Smart Grid Taxonomy [1/2]© 2012 Mischa Dohler and Jesús Alonso-Zárate 35
  36. 36. Smart Grid Taxonomy [2/2]© 2012 Mischa Dohler and JesúsCOMMUNICATION NETWORKS FOR POWER ENGINEERS,” [© Fabrizio Granelli, et al. “C4P: Alonso-Zárate 36Tutorial at SmartGridComms 2011, Brussels, Belgium.]
  37. 37. Smart Grid Comms Standards  Most relevant standard:  IEEE P2030 Smart Grid Interoperability of Energy Technology and Information Technology Operation with the Electric Power System (EPS), End- Use Applications, and Loads  Power System Control and Monitoring:  IEEE C37.238™-2011 - IEEE Standard Profile for Use of IEEE 1588 Precision Time Protocol in Power System Applications:  IEEE C37.239™-2010 - IEEE Standard for Common Format for Event Data Exchange (COMFEDE) for Power Systems  IEEE 1031™-2011 - IEEE Guide for the Functional Specification of Transmission Static Var Compensators  IEEE 1250™-2011 - IEEE Guide for Identifying and Improving Voltage Quality in Power Systems  IEEE 1808™-2011 - IEEE Guide for Collecting and Managing Transmission Line Inspection and Maintenance Data© 2012 Mischa Dohler and JesúsCOMMUNICATION NETWORKS FOR POWER ENGINEERS,” [© Fabrizio Granelli, et al. “C4P: Alonso-Zárate 37Tutorial at SmartGridComms 2011, Brussels, Belgium.]
  38. 38. IEEE P2030 Interoperability Concept© 2012 Mischa Dohler and JesúsCOMMUNICATION NETWORKS FOR POWER ENGINEERS,” [© Fabrizio Granelli, et al. “C4P: Alonso-Zárate 38Tutorial at SmartGridComms 2011, Brussels, Belgium.]
  39. 39. IEEE P2030 Task Force 3  Task Force 3 defines communication requirements between devices in the Smart Grid:  neutral to PHY/MAC standards used in the Smart Grid  deals with layers above PHY/MAC and below Layer 6  leave exact choice to designer to pick what is best for the application depending on geography, scalability, requirements, latency, etc.  Specific role of TF3  develop suitability matrix for various PHY/MACs and list PHY/MAC layers that can be used for devices interfacing to the Smart Grid (e.g. IEEE 802.15.4g/k or IEEE 802.11ah)  use IP for a large set of reasons (scalability, security, well understood, etc)© 2012 Mischa Dohler and JesúsCOMMUNICATION NETWORKS FOR POWER ENGINEERS,” [© Fabrizio Granelli, et al. “C4P: Alonso-Zárate 39Tutorial at SmartGridComms 2011, Brussels, Belgium.]
  40. 40. ETSI M2M Smart Grid Concept [© ETSI M2M]© 2012 Mischa Dohler and Jesús Alonso-Zárate 40
  41. 41. Today’s M2M Smart Grid Reality [© EDP Portugal and FP7 STREP WSAN4CIP]© 2012 Mischa Dohler and Jesús Alonso-Zárate 41
  42. 42. Capillary M2M© 2012 Mischa Dohler and Jesús Alonso-Zárate 42
  43. 43. 2.1 Quick Intro to Capillary M2M© 2012 Mischa Dohler and Jesús Alonso-Zárate 43
  44. 44. History of WSN  M2M© 2012 Mischa Dohler and Jesús Alonso-Zárate 44
  45. 45. Characteristics of Capillary M2M  What is “Capillary M2M”:  mostly embedded design  short-range communication systems  power consumption is major headache (go harvesting?)  ought to be standards compliant to facilitate “universal” connectivity  What is it not:  cellular system (cellular connectivity only possible via gateway)  pure wireless sensor networks (since not guaranteeing universal connectivity)  Conclusion:  Whilst many insights from academic research on WSNs can be used, the capillary M2M will be dominated by industry-driven standardized low-power solutions.© 2012 Mischa Dohler and Jesús Alonso-Zárate 45
  46. 46. Barriers in Capillary M2M Reliability Standards Ease of use Power consumption Development cycles Node size 0% 20% 20% 60% 80% 100% * source: OnWorld, 2005© 2012 Mischa Dohler and Jesús Alonso-Zárate 46
  47. 47. Design of Capillary M2M  Each node typically consists of these basic elements:  sensor  radio chip  microcontroller  energy supply  These nodes should be:  low – cost  low – complexity  low – size  low – energy© 2012 Mischa Dohler and Jesús Alonso-Zárate 47
  48. 48. Off-The-Shelf Hardware – Today?© 2012 Mischa Dohler and Jesús Alonso-Zárate 48
  49. 49. Hardware Differs Significantly© 2012 Mischa Dohler and Jesús Alonso-Zárate 49
  50. 50. 2.2 From Academia To Practice© 2012 Mischa Dohler and Jesús Alonso-Zárate 50
  51. 51. Experimentation – Surprise, Surprise! http://people.csail.mit.edu/jamesm/ http://senseandsensitivity.rd.francetelecom.com/© 2012 Mischa Dohler and Jesús Alonso-Zárate 51
  52. 52. Important Practical Challenges  External Interference:  often neglected in protocol design  however, interference has major impact on link reliability  Wireless Channel Unreliability:  MAC and routing protocols were often channel agnostic  however, wireless channel yields great uncertainties  Position Uncertainty:  (mainly geographic) routing protocols assumed perfect location knowledge  however, a small error in position can cause planarization techniques to fail© 2012 Mischa Dohler and Jesús Alonso-Zárate 52
  53. 53. First Challenge: External Interference IEEE802.11 (Wi-Fi) IEEE802.15.1 (Bluetooth) IEEE802.15.4 (ZigBee)© 2012 Mischa Dohler and Jesús Alonso-Zárate 53
  54. 54. First Challenge: External Interference Typical Tx power  IEEE802.11-2007: 100mW  IEEE802.15.4-2006: 1mW 2.4 GHz PHY Channels 11-26 5 MHz 2.4 GHz 2.4835 GHz© 2012 Mischa Dohler and Jesús Alonso-Zárate 54
  55. 55. First Challenge: External Interference 868 MHz 2.4 GHz 5 GHz 433 MHz IEEE802.11b/g/n IEEE802.11a/n IEEE802.15.4© 2012 Mischa Dohler and Jesús Alonso-Zárate 55
  56. 56. First Challenge: External Interference  45 motes*  50x50m office environment  12 million packets exchanged, equaly over all 16 channels *data collected by Jorge Ortiz and David Culler, UCB Publicly available at wsn.eecs.berkeley.edu© 2012 Mischa Dohler and Jesús Alonso-Zárate 56
  57. 57. Second Challenge: Multipath Fading© 2012 Mischa Dohler and Jesús Alonso-Zárate 57
  58. 58. Second Challenge: Multipath Fading 0% reliability 100% reliability ch.11 ch.12© 2012 Mischa Dohler and Jesús Alonso-Zárate 58
  59. 59. 2.3 From Practice to Proprietary M2M Solutions© 2012 Mischa Dohler and Jesús Alonso-Zárate 59
  60. 60. Key Embedded M2M Companies The Internet© 2012 Mischa Dohler and Jesús Alonso-Zárate 60
  61. 61. 2.4 Standardization Efforts Pertinent to M2M© 2012 Mischa Dohler and Jesús Alonso-Zárate 61
  62. 62. Interoperability Issues Why Standardization?© 2012 Mischa Dohler and Jesús Alonso-Zárate 62
  63. 63. Standardization Bodies  Standards Developing Organization bodies can be  international (e.g. ITU-T, ISO, IEEE),  regional (e.g. ANSI, ETSI), or  national (e.g. CCSA)  Standardization efforts pertinent to capillary M2M are:  IEEE (physical and link layer protocols)  IETF (network and transport protocols)  ISA (regulation for control systems)  ETSI (complete M2M solutions)  in cellular part© 2012 Mischa Dohler and Jesús Alonso-Zárate 63
  64. 64. Standardized Capillary M2M StackStack Protocol Zigbee-like Low-Power Wifi Application IETF CORE HTTP, etc Transport (Lightweight TCP), UDP TCP, UDP, etc IETF IETF ROLL (routing) Networking IPv4/6, etc IETF 6LoWPAN (adapt.) MAC IEEE 802.15.4e IEEE IEEE 802.11 PHY IEEE 802.15.4-2006© 2012 Mischa Dohler and Jesús Alonso-Zárate 64
  65. 65. 2.4.1 IEEE-Pertinent Capillary M2M Standards© 2012 Mischa Dohler and Jesús Alonso-Zárate 65
  66. 66. IEEE – Embedded Standards  The IEEE usually standardizes:  PHY layer of the transmitter  MAC protocol rules  The following IEEE standards are applicable to M2M:  IEEE 802.15.1 (technology used e.g. by Bluetooth/WiBree)  IEEE 802.15.4 (technology used e.g. by ZigBee and IETF 6LowPan)  IEEE 802.11 (technology used by WiFi)  Some facts and comments:  ultra-low power (ULP) IEEE 802.15.1 (WiBree) is competing … but to be seen  IEEE 802.15.4 was dormant and only with .15.4e seems to become viable  low power IEEE 802.11 solutions are becoming reality (origins with Ozmo Devices)© 2012 Mischa Dohler and Jesús Alonso-Zárate 66
  67. 67. APPL COAP TRAN UDP NET RPL IEEE 802.15.4e – Overview adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4  Standards history:  2006: TSCH approach emerges in the proprietary Time Synchronized Mesh Protocol (TSMP) of Dust Networks  2008/2009: TSMP is standardized in ISA100.11a and WirelessHART  2011: 802.15.4e Sponsor Ballot opened on 27 July 2011 and closed on 28 August with 96% of votes being affirmative  Aim of amendment:  define a MAC amendment to the existing standard 802.15.4-2006  to better support industrial markets  3 different MACs for 3 different types of applications:  LL: Low Latency  CM: Commercial Application  PA: Process Automation© 2012 Mischa Dohler and Jesús Alonso-Zárate 67
  68. 68. APPL COAP TRAN UDP NET RPL PA - Process Automation [1/2] adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4  Slotframe structure = sequence of repeated time slots:  time slot can be used by one/multiple devices (dedicated/shared link) or empty  multiple slotframes with different lengths can operate at the same time  SlotframeCycle: every new slotframe instance in time  Slotframe size: # slots in a slotframe slotframe time slot TS0 TS1 TS2 TS0 TS1 TS2 TS0 TS1 TS2 TS0 TS1 TS2 CYCLE N -1 CYCLE N CYCLE N+1 CYCLE N+2© 2012 Mischa Dohler and Jesús Alonso-Zárate 68
  69. 69. APPL COAP TRAN UDP NET RPL PA - Process Automation [2/2] adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4  Link = (time slot, channel offset)  CHANNEL HOPPING  Dedicated link assigned to:  dedicated link: 1 node for Tx; 1 or more for Rx  shared link: 1 or more for Tx  Prime aim to help:  channel impairments  system capacity© 2012 Mischa Dohler and Jesús Alonso-Zárate 69
  70. 70. APPL COAP TRAN UDP NET RPL PA - Channel Hopping adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4 A B© 2012 Mischa Dohler and Jesús Alonso-Zárate 70
  71. 71. APPL COAP TRAN UDP NET RPL PA - Slotted Structure adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4 A  A super-frame repeats over time  Number of slots in a superframe is tunable C B  Each cell can be assigned to a pair of motes, in a given direction E D16 channel offsets F G I H J© 2012 Mischa Dohler and Jesús Alonso-Zárate e.g. 31 time slots (310ms) 71
  72. 72. APPL COAP TRAN UDP NET RPL PA - Slot Structure adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4 9.976ms 2.120ms < 4.256ms 0.800ms 0.400ms 2ms 2.400ms© 2012 Mischa Dohler and Jesús Alonso-Zárate 72
  73. 73. APPL COAP TRAN UDP NET RPL PA - Energy Consumption adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4 9.976ms 2.120ms < 4.256ms 0.800ms 0.400ms 2ms 2.400ms Type of slot Transmitter Receiver “OFF” - - transmission w. ACK 6.856ms 7.656ms Transmission w.o. ACK 4.256ms 5.256ms Listening w.o. reception - 2.000ms© 2012 Mischa Dohler and Jesús Alonso-Zárate 73
  74. 74. APPL COAP TRAN UDP NET RPL PA - Slotted Structure adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4  Cells are assigned according to A application requirements C B E D16 channel offsets F G I H J© 2012 Mischa Dohler and Jesús Alonso-Zárate e.g. 33 time slots (330ms) 74
  75. 75. APPL COAP TRAN UDP NET RPL PA - Trade-Off [1/3] adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4 A  Cells are assigned according to application requirements  Tunable trade-off between  packets/second C B …and energy consumption E D16 channel offsets F G I H J© 2012 Mischa Dohler and Jesús Alonso-Zárate e.g. 33 time slots (330ms) 75
  76. 76. APPL COAP TRAN UDP NET RPL PA - Trade-Off [2/3] adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4 A  Cells are assigned according to application requirements  Tunable trade-off between  packets/second C B  Latency …and energy consumption E D16 channel offsets F G I H J© 2012 Mischa Dohler and Jesús Alonso-Zárate e.g. 33 time slots (330ms) 76
  77. 77. APPL COAP TRAN UDP NET RPL PA - Trade-Off [3/3] adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4 A  Cells are assigned according to application requirements  Tunable trade-off between  packets/second C B  Latency …and energy consumption  Robustness E D16 channel offsets F G I H J© 2012 Mischa Dohler and Jesús Alonso-Zárate e.g. 33 time slots (330ms) 77
  78. 78. APPL COAP TRAN UDP NET RPL PA - Synchronization adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4 clocks drift Periodic realignment (10ppm typical) (within a clock tick) ∆t© 2012 Mischa Dohler and Jesús Alonso-Zárate 78
  79. 79. APPL COAP TRAN UDP NET RPL PA – Lifetime adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4 Type of slot Transmitter Receiver “OFF” - - transmission w. ACK 6.856ms 7.656ms  Assumptions Transmission w.o. ACK 4.256ms 5.256ms  2400mAh (AA battery) Listening w.o. reception - 2.000ms  14mA when radio on (AT86RF231)  If my radio is on all the time  171 hours of time budget (7 days of lifetime)  If I only want to keep synchronization (theoretical lower limit)  7.656ms from a time budget of 171 hours I can resync. 80x106 times  76 years of lifetime (» battery shelf-life)  A duty cycle of 1%  2 years of lifetime© 2012 Mischa Dohler and Jesús Alonso-Zárate 79
  80. 80. APPL COAP TRAN UDP NET RPL PA – Lifetime adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4  Looking at node D A  “normal” case • 1 reception, 1 transmission (15ms) every 3.3 seconds • .45% duty cycle  4 years lifetime C B E D16 channel offsets F G I H J© 2012 Mischa Dohler and Jesús Alonso-Zárate e.g. 330 time slots (3.3s) 80
  81. 81. APPL COAP TRAN UDP NET RPL PA – Lifetime adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4  Looking at node D A  “normal” case  Triple data rate • 3 receptions, 3 transmissions (45ms) every 3.3 seconds C B • 1.36% duty cycle  17 months lifetime E D16 channel offsets F G I H J© 2012 Mischa Dohler and Jesús Alonso-Zárate e.g. 33 time slots (330ms) 81
  82. 82. APPL COAP TRAN UDP NET RPL PA – Network Schedule adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4  IEEE 802.15.4e draft defines how the MAC executes a schedule but it does not specify how such schedule is built:  centralized approach: a manager node (i.e., the gateway connecting the network to Internet) is responsible for building and maintaining the network schedule; provide efficient schedules for static network  distributed approach: nodes decide locally on which links to transmit with which neighbors; suitable for mobile networks with many gateways  Traffic Aware Scheduling Algorithm (TASA): Figure. Matching and Coloring phases of TASA scheme [ M.R. Palattella, N. Accettura, M. Dohler, L.A. Grieco, G. Boggia, “Traffic Aware Scheduling Algorithm for the Emerging IEEE 802.15.4e Standard”, submitted to ACM© 2012 Mischa Dohler and Jesús Alonso-Zárate Transaction on Sensor Networks.] 82
  83. 83. APPL COAP TRAN UDP NET RPL PA – Network Schedule adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4 50 % 80 % gain Performance of IEEE 802.15.4e networks implementing TASA [M.R. Palattella, N. Accettura, M. Dohler, L.A. Grieco, G. Boggia, “Traffic Aware Scheduling Algorithm for the Emerging IEEE 802.15.4e Standard”, submitted to ACM Transaction on© 2012 Mischa Dohler and Jesús Alonso-Zárate Sensor Networks.] 83
  84. 84. 2.4.2 IETF-Pertinent Capillary M2M Standards© 2012 Mischa Dohler and Jesús Alonso-Zárate 84
  85. 85. IETF – Overview  Internet Engineering Task Force:  not approved by the US government; composed of individuals, not companies  quoting the spirit: “We reject kings, presidents and voting. We believe in rough consensus and running code.” D. Clark, 1992  more than 120 active working groups organized into 8 areas  General scope of IETF:  above the wire/link and below the application  TCP/IP protocol suite: IP, TCP, routing protocols, etc.  however, layers are getting fuzzy (MAC & APL influence routing)  hence a constant exploration of "edges“  IETF developments pertinent to Capillary M2M:  6LoWPAN (IPv6 over Low power WPAN)  ROLL (Routing Over Low power and Lossy networks)  CORE (Constrained RESTful Environments)© 2012 Mischa Dohler and Jesús Alonso-Zárate 85
  86. 86. APPL COAP TRAN UDP NET RPL Internet Protocol (IP) version 6 adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4 # of protocols  Every host on the Internet has a unique Internet Protocol (IP) address HTTP, XML, etc.  A packet with an IP header is routed to its destination over the Internet TCP, UDP  IP is the narrow waist of the Internet  “If you speak IP, you are on the Internet” IP  Evolution of the Internet Protocol  IPv4 (1981) is currently used • 32-bit addresses • “third-party toolbox”: ARP, DHCP IEEE802.3  IPv6 (1998) is being deployed IEEE802.11 • “toolbox” integrated • 128-bit addresses IEEE802.15.4© 2012 Mischa Dohler and Jesús Alonso-Zárate 86
  87. 87. APPL COAP TRAN UDP NET RPL IETF 6LoWPAN adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4  6LoWPAN has the following key properties:  IPv6 for very low power embedded devices using IEEE 802.15.4  provision of neighborhood discovery protocol  header compression with up to 80% compression rate  packet fragmentation (1260 byte IPv6 frames -> 127 byte 802.15.4 frames)  direct end-to-end Internet integration (but no routing)© 2012 Mischa Dohler and Jesús Alonso-Zárate 87
  88. 88. APPL COAP TRAN UDP NET RPL IETF 6LoWPAN adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4  Typical architecture:© 2012 Mischa Dohler and Jesús Alonso-Zárate 88
  89. 89. APPL COAP TRAN UDP NET RPL IETF ROLL – Status adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4  IETF WG “Routing Over Low power and Lossy networks”  Design a routing protocol for Wireless Mesh Network  Final stage of standardization  Since LLNs are application specific, 4 scenarios are dealt with:  home applications: draft-brandt-roll-home- routing-reqs  industrial applications: draft-pister-roll-indus- routing-reqs  urban applications: draft-dohler-roll-urban- routing-reqs  vehicular applications: draft-wakikawa-roll- invehicle-reqs© 2012 Mischa Dohler and Jesús Alonso-Zárate 89
  90. 90. APPL COAP TRAN UDP NET RPL IETF ROLL – Overview adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4  Gradient-Based & Distance Vector Routing Protocol:  topology organized as a Direct Acyclic Graph (DAG, i.e., a gradient)  Destination Oriented DAG (DODAG) is built per sink or LBR (Local Border Router) • redundant paths from each leaf to the DODAG root are included • nodes acquire a “rank” based on the distance to the sink • message follow the gradient of ranks  Separation of packet processing and forwarding (RPL core functionalities) from the routing optimization (Objective Function and Routing Metrics/Constraints) rank = 1 rank = 2 rank = 3 [© Maria Rita, UL presentation, 2012]© 2012 Mischa Dohler and Jesús Alonso-Zárate 90
  91. 91. IETF RPL – Packet Forwarding (UL) @0 @0 cost=10 0028 cost=20 cost=30 @255@20 @255@20 @255@10 cost=10 cost=15 @255@25 cost=25 0068 0051 008a cost=30 002b cost=10 cost=30 @255@50 cost=20 @255@35 cost=20 @255@40 @255@40 006e cost=15 0045 cost=15 007e 0063 cost=10 1. Each node heartbeats its rank @255@45 • Initially 0 for the OpenLBR • Initially 255 (max value) for others 2. Nodes evaluate the link cost (ETX) to their neighbors 0072 • In our case 10*(1/packet delivery ratio) • Perfect link: cost=10 • Link with 50% loss: cost=20 3. Nodes update their rank as min(rank neighbor+link cost) over all neighbors • The chosen neighbor is preferred routing parent 4. Continuous and Jesús Alonso-Zárate© 2012 Mischa Dohler updating process 91
  92. 92. APPL COAP TRAN UDP NET RPL IETF ROLL – Routing Opt. adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4  Objective Functions (OFs):  translate key metrics and constraints into a rank  allow the selection of a DODAG to join  parent selection at a node could be triggered in response to several events  Metrics and Constraints:  node energy, hop count, throughput, latency, link reliability and encryption  possibility to timely adapt the topology to changing network conditions  need to keep under control the adaptation rate of routing metrics in order to avoid path instabilities© 2012 Mischa Dohler and Jesús Alonso-Zárate [© Maria Rita, UL presentation, 2012] 92
  93. 93. APPL COAP TRAN UDP NET RPL IETF– UDP adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4  Transport layer is responsible of providing end-to-end reliability over IP-based networks  Why TCP (Transmission Control Protocol) is not (so) suitable?  support for unicast communications only  reacts badly to e.g. wireless packet loss  not all protocols require total reliability  3-way handshake TCP connection not suitable for very short transactions  The User Datagram Protocol (UDP) (RFC 768)  datagram oriented protocol (i.e., asynchronous)  used to deliver short messages over IP  unreliable, connectionless protocol  can be used with broadcast and multicast  to be integrated with retransmission control mechanisms at application layer© 2012 Mischa Dohler and Jesús Alonso-Zárate [© Maria Rita, UL presentation, 2012] 93
  94. 94. APPL COAP TRAN UDP NET RPL IETF CORE – COAP adaptation MAC PHY 6LoWPAN IEEE802.15.4e IEEE802.15.4  2008: The IETF Constrained RESTful Environments (CORE) working group defines the Constrained Application Protocol (CoAP) o it translates to “HTTP (HyperText Transfer Protocol) for integration with the web, meeting specialized LLNs requirements: multicast support, very low overhead, and simplicity for constrained environments.”  Asynchronous request/response protocol over UDP  The CoAP architecture is divided into two layers: 1. a message layer in charge of reliability and sequencing 2. a request/response layer in charge of mapping requests to responses and their semantics  Main features addressed by CoAP: o constrained web protocol specialized to M2M requirements o stateless HTTP mapping through the use of proxies or direct mapping of HTTP interfaces to CoAP o UDP transport with application layer reliable unicast and best-effort multicast support o asynchronous message exchanges o low header overhead and parsing complexity o URI and Content-type support o simple proxy and caching capabilities o optional resource discovery© 2012 Mischa Dohler and Jesús Alonso-Zárate [© Maria Rita, UL presentation, 2012] 94
  95. 95. 2.4.3 Low Power Wifi & Comparisons with .15.4© 2012 Mischa Dohler and Jesús Alonso-Zárate 95
  96. 96. Wifi-Enabled Sensors! “Low-power Wi-Fi provides a significant improvement over typical Wi-Fi on both latency and energy consumption counts.” “LP-Wifi consumes approx the same as 6LoWPAN for small packets but is much better for large packets.”© 2012 Mischa Dohler and Jesús Alonso-Zárate [© IEEE, from “Feasibility of Wi-Fi Enabled Sensors for Internet of Things,” by Serbulent Tozlu 96
  97. 97. Cellular M2M© 2012 Mischa Dohler and Jesús Alonso-Zárate 97
  98. 98. Cellular M2M: Outline  3.1 Introduction to Cellular M2M  3.1.1 Fundamentals of Cellular Systems  3.1.2 Motivating Cellular Systems for M2M  3.2 M2M in Current Cellular Systems  3.2.1 GSM family: GSM, GPRS, EDGE  3.2.2 3GPP family: UMTS, LTE, LTE-A  3.3 Cellular M2M Standardization Activities  3.3.1 Overview of Standardization in Cellular Communications  3.3.2 M2M Activities in ETSI  3.3.3 M2M Activities in 3GPP  3.3.4 M2M Activities in IEEE  3.4 Concluding Remarks© 2012 Mischa Dohler and Jesús Alonso-Zárate 98
  99. 99. 3.1 Introduction to Cellular M2M© 2012 Mischa Dohler and Jesús Alonso-Zárate 99
  100. 100. 3.1.1 Fundamentals of Cellular Systems© 2012 Mischa Dohler and Jesús Alonso-Zárate 100
  101. 101. Data Traffic Evolution - PAST AT&T traffic evolution Source: AT&T© 2012 Mischa Dohler and Jesús Alonso-Zárate 101
  102. 102. Data Traffic Evolution - FUTURE Total mobile traffic (EB per year) 140.00 120.00 100.00 Yearly traffic in EB Europe 80.00 Americas Asia 60.00 Rest of the world W orld 40.00 20.00 - 2010 2015 2020 Source: IDATE Exabyte = 10^18© 2012 Mischa Dohler and Jesús Alonso-Zárate 102
  103. 103. Cellular Evolution ITU-R req. for IMT-Advanced Means to achieve higher data rates: More spectrum, more efficient RRM, smaller cells 2G 2.5G 3G 3.5G 4G B4G? Exabyte = 10^18 Source: NEC – Andreas Maeder, Feb 2012© 2012 Mischa Dohler and Jesús Alonso-Zárate 103
  104. 104. Cellular Generation Salad [1/2]  2G Networks:  GSM (Global System for Mobile Communications), 1990, worldwide  IS95 (Interim Standard 95), mainly US  2.5G Network:  GPRS (General Packet Radio System), worldwide  3G Networks:  EDGE (Enhanced Data Rates for GSM Evolution), GSM evolution  UMTS (Universal Mobile Telecommunication System) (3GPP-Release 4)  CDMA2000 (based on 2G CDMA Technology) (3GPP2), discontinued in 2008  WiMAX, IEEE 802.16 technology  3.5G Network:  HDxPA (High Data Packet Access), 3GPP evolution (Release 5 and 6), 2007  HDPA+ with complementary EDGE (Release 7)© 2012 Mischa Dohler and Jesús Alonso-Zárate 104
  105. 105. Cellular Generation Salad [2/2]  “3.9G” Network:  December 2008: LTE (Long Term Evolution), UMTS evolution/revolution, worldwide, Release 8, small enhancements in Release 9  4G Networks:  LTE-A (LTE Advanced), LTE evolution/revolution, worldwide, 2009 first submission to be considered for IMT-advanced, Release 10  WiMAX II, IEEE 802.16j/m high capacity networks Note that both LTE and WiMAX are regarded as beyond 3G (B3G) systems but are strictly speaking not 4G since not fulfilling the requirements set out by the ITU for 4G next generation mobile networks (NGMN). NGMN requires downlink rates of 100 Mbps for mobile and 1 Gbps for fixed-nomadic users at bandwidths of around 100 MHz which is the prime design target of LTE Advanced and WiMAX II. Therefore, even though LTE is (somehow wrongly but understandably) marketed as 4G, it is not and we still need to wait for LTE-A.© 2012 Mischa Dohler and Jesús Alonso-Zárate 105
  106. 106. 3GPP Release 11 (LTE-A) – Timeline  What’s next? Release 11  Green Activities / Energy Efficiency: ICT 2% of global emissions (telecom 0,5%)  Standardization for M2M applications  Technical Specification Group (TSG) Service and Systems Aspects (SA) Report – September 2010:  http://www.3gpp.org/ftp/tsg_sa/TSG_SA/TSGS_49/Report/  Tentative Freeze Dates:  Stage 1 freeze (no further additional functions added) date: September 2011  Stage 2 freeze date: March 2012  Stage 3 freeze target: September 2012  RAN ASN.1 freeze target: 3 months after Stage 3 Freeze.  equivalent CT formal interface specification freeze: 3 months after Stage 3 Freeze  More info at: http://www.3gpp.org/ftp/Information/WORK_PLAN/Description_Releases/© 2012 Mischa Dohler and Jesús Alonso-Zárate 106
  107. 107. However, things are changing…© 2012 Mischa Dohler and Jesús Alonso-Zárate 107
  108. 108. Vision of the Network of Things Presented by Interdigital: Globecom’11 – IWM2M, Houston© 2012 Mischa Dohler and Jesús Alonso-Zárate 108
  109. 109. 3.1.2 Motivating Cellular for M2M Applications© 2012 Mischa Dohler and Jesús Alonso-Zárate 109
  110. 110. A Simple Motivation: Numbers Source: “The revenue opportunity for mobile connected devices in saturated markets,” Northstream White Paper, February 2010© 2012 Mischa Dohler and Jesús Alonso-Zárate 110
  111. 111. A Simple Motivation: Initiatives  Global Initiatives:  ETSI, GSMA, TIA TR-50 Smart Device Communications  Modules & Modems:  Anydata, CalAmp, Cinterion, DiGi, Enfora, Ericsson, eDevice, Inside M2M, Iwow, Laird Technologies, Maestro, Moxa, Multitech, Motorola, Mobile Devices, Owasys, Quectel Industry, Sagem, Sierra Wireless, SimCom, Telit, Teltonika, uBlox  Network Connectivity/Services:  AT&T Inc., KORE Telematics, KPN, Numerex Corp., Orange SA, Rogers Business Solutions, Sprint, TIM (Brasil), Telcel  System Integrators:  Accenture Ltd., Atos Origin, IBM, inCode  Sim Cards:  Gemalto, Giesecke & Devrient, Oberthur, Sagem Orga© 2012 Mischa Dohler and Jesús Alonso-Zárate 111
  112. 112. Reality  THE advantage of cellular M2M:  Ethernet/WiFi/etc only provides local coverage  Users already familiar with and proven infrastructure  Easier configuration: suitable for short-term deployments  Cellular networks provide today ubiquitous coverage & global connectivity  Mobility and High-Speed Data Transmission  … and, above all, interference can be managed© 2012 Mischa Dohler and Jesús Alonso-Zárate 112

×