Re-inventing the Telephone System: The Third Generation Henning Schulzrinne Dept. of Computer Science Columbia University ...
Overview <ul><li>1 st  generation: analog 2 nd  generation: digital circuit switched 3 rd  generation: packet-switched </l...
Lifecycle of technologies military corporate consumer traditional technology propagation: opex/capex doesn’t matter; exper...
Internet and networks timeline 1960 1970 1980 1990 2000 2010 theory university prototypes production use in research comme...
Earlier PSTN changes <ul><li>starting in 1980s: </li></ul><ul><ul><li>analog    digital transmission </li></ul></ul><ul><...
Technology evolution of the PSTN SS7: 1987-1997
What is VoIP? <ul><li>Voice-over-IP = Internet telephony </li></ul><ul><ul><li>“ Internet telephony refers to communicatio...
Brief history of packetized voice <ul><li>1969: ARPAnet, predecessor of modern Internet </li></ul><ul><li>1974: real-time ...
How has the industry progressed <ul><li>Softswitch networks carry approximately 2 billion minutes/day vs. 2.3 million in 1...
VoIP penetration Glen Campbell Telecom & Cable Analyst Merrill Lynch Canada May 2004 (CITI VoIP workshop) residential & sm...
Vonage Subscriber Growth *Cable Datacom News Round Up, September 1, 2003 subscribers   in thousands
Why has it taken so long? <ul><li>VoIP technology development since 1995 </li></ul><ul><li>Web: worked on dial-up, motivat...
Technology introductions Source:  OECD, 2003
Conditions for VoIP <ul><li>Multimedia PC with low-latency OS </li></ul><ul><ul><li>earlier Windows versions not suitable ...
Total high-speed lines FCC, 2004
DSL usage DSL Forum, Sept. 2004
DSL penetration DSL Forum, Sept. 2004
Who provides VoIP service? voice service provider (Vonage, Lingo, Packet8) cable providers DSL (ILEC) long-distance carrie...
Motivations for VoIP <ul><li>access fee </li></ul><ul><li>taxes </li></ul><ul><li>monopoly rents </li></ul><ul><li>local-l...
VoIP models: PBX LAN enterprise VoIP gateway call server proxy server “ softswitch” analog telephone adapter IP
VoIP models: IP Centrex LAN enterprise or residence service provider IP
Some differences: VoIP vs. PSTN <ul><li>Separate signaling from media data path </li></ul><ul><li>But, unlike SS7, same ne...
VoIP components <ul><li>Re-uses whole Internet protocol architecture and transmission infrastructure </li></ul><ul><ul><li...
SIP trapezoid SIP trapezoid outbound proxy a@foo.com: 128.59.16.1 registrar 1 st  request 2 nd , 3 rd , … request voice tr...
Example SIP phones about $85
PSTN vs. Internet Telephony Signaling & Media Signaling & Media Signaling Signaling Media PSTN: Internet telephony: China ...
SIP as service enabler <ul><li>Rendezvous protocol </li></ul><ul><ul><li>lets users find each other by only knowing a perm...
Changes caused by VoIP <ul><li>Access independence: single-function network to voice-over-any-network </li></ul><ul><ul><l...
(Early) Adulthood <ul><li>“fully developed and mature” </li></ul><ul><ul><li>Not quite yet, but no longer a teenager </li>...
Emerging technologies <ul><li>Core VoIP technology largely finished </li></ul><ul><ul><li>deployment largely due to cost s...
Near future: Location-based services <ul><li>Finding services based on location </li></ul><ul><ul><li>physical services (s...
Location-based IM & presence
User service creation <ul><li>Tailor a shared infrastructure to individual users </li></ul><ul><li>traditionally, only ven...
Near future: Multimedia <ul><li>Wideband audio </li></ul><ul><ul><li>“ better than phone quality”    lectures, discussion...
Near future: VoIP over WiFi <ul><li>Not fundamentally different from landline VoIP </li></ul><ul><ul><li>combination cellu...
Challenge: Global interconnect <ul><li>Currently, each VoIP “network” largely isolated </li></ul><ul><ul><li>interconnect ...
Challenge: CALEA (lawful intercept) <ul><li>Existing models assume congruence of signaling and voice flows </li></ul><ul><...
Challenge: User-programmable and context-aware services <ul><li>Universal reachability    control reachability in time an...
Challenge: Spam prevention <ul><li>Currently, telemarketing restricted to in-country calling </li></ul><ul><ul><li>With Vo...
Challenge: Service reliability <ul><li>“ QoS”    service availability </li></ul><ul><ul><li>loss of network connection </...
Challenge: Emergency calling <ul><li>911 calling system largely unchanged since 1980s </li></ul><ul><ul><li>call routing t...
Three stages to VoIP 911 yes yes no callback number to PSAP? IP-enabled no (8 or 10 digit) no PSAP modification GNP multim...
Prototype *  gray  features in progress.
Call taker setup SIPc client receives calls GeoLynx software displays caller location
Conclusion <ul><li>VoIP on cusp of widespread deployment: </li></ul><ul><ul><li>commercial-grade VoIP products </li></ul><...
Upcoming SlideShare
Loading in …5
×

CAT-VoIP.ppt

0 views
2,813 views

Published on

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

  • Be the first to like this

No Downloads
Views
Total views
0
On SlideShare
0
From Embeds
0
Number of Embeds
0
Actions
Shares
0
Downloads
27
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide
  • With a few exceptions, in Internet telephony, end systems are the only entities where signaling and media flows converge. Thus, any service that requires interaction with user media is likely to be easier to implement in the end systems. More seperately
  • CAT-VoIP.ppt

    1. 1. Re-inventing the Telephone System: The Third Generation Henning Schulzrinne Dept. of Computer Science Columbia University CAT Forum -- October 12, 2004
    2. 2. Overview <ul><li>1 st generation: analog 2 nd generation: digital circuit switched 3 rd generation: packet-switched </li></ul><ul><li>What is VoIP? Why now? </li></ul><ul><li>A short history </li></ul><ul><li>Wireless VoIP </li></ul><ul><li>Context-aware communications </li></ul><ul><li>VoIP  IM, presence </li></ul><ul><li>Challenges ahead </li></ul>
    3. 3. Lifecycle of technologies military corporate consumer traditional technology propagation: opex/capex doesn’t matter; expert support capex/opex sensitive, but amortized; expert support capex sensitive; amateur Can it be done? Can I afford it? Can my mother use it?
    4. 4. Internet and networks timeline 1960 1970 1980 1990 2000 2010 theory university prototypes production use in research commercial early residential broadband home email ftp DNS RIP UDP TCP SMTP SNMP finger ATM BGP, OSPF Mbone IPsec HTTP HTML RTP 100 kb/s 1 Mb/s 10 Mb/s XML OWL SIP Jabber 100 Mb/s 1 Gb/s port speeds Internet protocols queuing architecture routing cong. control DQDB, ATM QoS VoD p2p ad-hoc sensor
    5. 5. Earlier PSTN changes <ul><li>starting in 1980s: </li></ul><ul><ul><li>analog  digital transmission </li></ul></ul><ul><ul><li>in-band  out-of-band (SS7) signaling </li></ul></ul><ul><li>end systems relatively unaffected </li></ul><ul><li>few additional services </li></ul><ul><ul><li>800# </li></ul></ul><ul><ul><li>CLASS services (caller ID, call waiting) </li></ul></ul><ul><li>customer relationship largely unaffected </li></ul><ul><ul><li>except CLECs and reselling </li></ul></ul>
    6. 6. Technology evolution of the PSTN SS7: 1987-1997
    7. 7. What is VoIP? <ul><li>Voice-over-IP = Internet telephony </li></ul><ul><ul><li>“ Internet telephony refers to communications services—voice, facsimile, and/or voice-messaging applications—that are transported via the Internet , rather than the public switched telephone network (PSTN). The basic steps involved in originating an Internet telephone call are conversion of the analog voice signal to digital format and compression/translation of the signal into Internet protocol (IP) packets for transmission over the Internet; the process is reversed at the receiving end.” (IEC) </li></ul></ul><ul><li>Not a single technology, but combination of Internet technologies </li></ul><ul><li>Now typically voice only, but easily extended to video </li></ul>
    8. 8. Brief history of packetized voice <ul><li>1969: ARPAnet, predecessor of modern Internet </li></ul><ul><li>1974: real-time packetized voice (early Internet) </li></ul><ul><li>1990: primitive version used for transatlantic calls (G.764) </li></ul><ul><li>1991: DARTnet (test network) audio experiments using Sun workstations </li></ul><ul><li>1992: first IETF multicast audiocast </li></ul><ul><li>1992: RTP (transport) draft </li></ul><ul><li>1995: first commercial PC-to-PC voice software (Vocaltec) </li></ul><ul><li>1995/1996: first PC-to-PC and PC-to-phone services (Net2Phone, DialPad, Vocaltec, …) </li></ul><ul><li>1996: first version of SIP and H.323 standards </li></ul><ul><li>~2000: first service providers </li></ul><ul><li>~2002: first large-scale consumer services </li></ul><ul><li>2002: 3G wireless specifies Internet multimedia subsystem </li></ul>
    9. 9. How has the industry progressed <ul><li>Softswitch networks carry approximately 2 billion minutes/day vs. 2.3 million in 1999 </li></ul><ul><li>Services, such as IP-Centrex, are quickly being adopted by enterprises </li></ul><ul><li>10% of all international voice traffic to/from U.S. carried on IP </li></ul><ul><ul><li>primarily prepaid calling cards </li></ul></ul><ul><li>IP-enabled handset sales over 4.5 million units in 2002 </li></ul><ul><ul><li>35% of all total premise sales are IP – enabled </li></ul></ul><ul><ul><li>IP handset costs drop from $600 in 2001 to $99 today </li></ul></ul>TIA, 2003 Telecommunications Market Review and Forecast Gartner Dataquest, 2001-2007, United States: Fixed Public Network Services Gartner Dataquest, 2002 Premises Based Equipment Sales Jack Waters Fall VON 2003
    10. 10. VoIP penetration Glen Campbell Telecom & Cable Analyst Merrill Lynch Canada May 2004 (CITI VoIP workshop) residential & small business
    11. 11. Vonage Subscriber Growth *Cable Datacom News Round Up, September 1, 2003 subscribers in thousands
    12. 12. Why has it taken so long? <ul><li>VoIP technology development since 1995 </li></ul><ul><li>Web: worked on dial-up, motivated broadband </li></ul><ul><ul><li>deployment from 1992 to 2000 </li></ul></ul><ul><li>VoIP: not usable on dial-up, spurred by residential broadband </li></ul><ul><li>More than just protocols needed: </li></ul><ul><ul><li>eco-system (management, configuration, OSS, …) </li></ul></ul><ul><ul><li>interoperability </li></ul></ul><ul><ul><li>spectrum of products – low to high end </li></ul></ul><ul><ul><li>interoperation with legacy equipment </li></ul></ul>
    13. 13. Technology introductions Source: OECD, 2003
    14. 14. Conditions for VoIP <ul><li>Multimedia PC with low-latency OS </li></ul><ul><ul><li>earlier Windows versions not suitable </li></ul></ul><ul><li>Broadband access for residence </li></ul><ul><ul><li>modem adds significant delay </li></ul></ul><ul><li>High-speed switched LANs for businesses </li></ul><ul><li> only feasible since mid-1990’s </li></ul>
    15. 15. Total high-speed lines FCC, 2004
    16. 16. DSL usage DSL Forum, Sept. 2004
    17. 17. DSL penetration DSL Forum, Sept. 2004
    18. 18. Who provides VoIP service? voice service provider (Vonage, Lingo, Packet8) cable providers DSL (ILEC) long-distance carrier (e.g., AT&T, MCI) service + gateways + IP network service + gateways service + gateways + IP network + access
    19. 19. Motivations for VoIP <ul><li>access fee </li></ul><ul><li>taxes </li></ul><ul><li>monopoly rents </li></ul><ul><li>local-loop access </li></ul><ul><li>separate wiring plant </li></ul><ul><li>cheaper services </li></ul><ul><li>(caller ID, etc.) </li></ul><ul><li>higher network </li></ul><ul><li>efficiency </li></ul><ul><li>better voice quality </li></ul><ul><ul><li>possible </li></ul></ul><ul><li>user-defined services </li></ul><ul><li>video and app. sharing </li></ul><ul><li>integration of presence </li></ul><ul><li>abundance of </li></ul><ul><li>identifiers </li></ul><ul><li>mobility </li></ul><ul><li>media encryption </li></ul><ul><li>signaling encryption </li></ul><ul><li>user authentication </li></ul>financial services security
    20. 20. VoIP models: PBX LAN enterprise VoIP gateway call server proxy server “ softswitch” analog telephone adapter IP
    21. 21. VoIP models: IP Centrex LAN enterprise or residence service provider IP
    22. 22. Some differences: VoIP vs. PSTN <ul><li>Separate signaling from media data path </li></ul><ul><li>But, unlike SS7, same network  lower call setup delay </li></ul><ul><li>Avoid CTI complexity of &quot;remote control&quot; </li></ul><ul><li>Mobile and wireline very similar </li></ul><ul><li>Any media as session: </li></ul><ul><ul><li>any media quality (e.g., TV and radio circuits) </li></ul></ul><ul><ul><li>interactive games </li></ul></ul><ul><li>No need for telephone company </li></ul>voice service provider (RTP, SIP) ISP (IP, DHCP, DNS) dark fiber provider Yahoo MCI NYSERNET
    23. 23. VoIP components <ul><li>Re-uses whole Internet protocol architecture and transmission infrastructure </li></ul><ul><ul><li>IP, UDP for transport </li></ul></ul><ul><ul><li>TLS and S/MIME for security </li></ul></ul><ul><ul><li>HTTP for configuration </li></ul></ul>SIP/SDP H.248 MGCP H.323 signaling provide URI provide destination address codecs (G.7xx, H.26x)  ENUM H.350 directories RTP transport
    24. 24. SIP trapezoid SIP trapezoid outbound proxy a@foo.com: 128.59.16.1 registrar 1 st request 2 nd , 3 rd , … request voice traffic RTP destination proxy (identified by SIP URI domain)
    25. 25. Example SIP phones about $85
    26. 26. PSTN vs. Internet Telephony Signaling & Media Signaling & Media Signaling Signaling Media PSTN: Internet telephony: China Belgian customer, currently visiting US Australia
    27. 27. SIP as service enabler <ul><li>Rendezvous protocol </li></ul><ul><ul><li>lets users find each other by only knowing a permanent identifier </li></ul></ul><ul><li>Mobility enabler: </li></ul><ul><ul><li>personal mobility </li></ul></ul><ul><ul><ul><li>one person, multiple terminals </li></ul></ul></ul><ul><ul><li>terminal mobility </li></ul></ul><ul><ul><ul><li>one terminal, multiple IP addresses </li></ul></ul></ul><ul><ul><li>session mobility </li></ul></ul><ul><ul><ul><li>one user, multiple terminals in sequence or in parallel </li></ul></ul></ul><ul><ul><li>service mobility </li></ul></ul><ul><ul><ul><li>services move with user </li></ul></ul></ul>
    28. 28. Changes caused by VoIP <ul><li>Access independence: single-function network to voice-over-any-network </li></ul><ul><ul><li>separation of transport and services </li></ul></ul><ul><li>Transition from “polling” service (try until user happens to be available) to “presence” service </li></ul><ul><li>Voice special  voice just one media among many </li></ul>
    29. 29. (Early) Adulthood <ul><li>“fully developed and mature” </li></ul><ul><ul><li>Not quite yet, but no longer a teenager </li></ul></ul><ul><ul><li>probably need another 6 years to be grown up… </li></ul></ul><ul><li>Responsibilities: </li></ul><ul><ul><li>Dealing with elderly relatives  POTS </li></ul></ul><ul><ul><li>Financial issues  payments, RADIUS </li></ul></ul><ul><ul><li>Family emergencies  911 </li></ul></ul>
    30. 30. Emerging technologies <ul><li>Core VoIP technology largely finished </li></ul><ul><ul><li>deployment largely due to cost savings, not new services </li></ul></ul><ul><ul><li>toll and fee bypass </li></ul></ul><ul><ul><li>integrated infrastructure (LAN & WAN) </li></ul></ul><ul><ul><li>extend “PBX” reach to home and branch offices </li></ul></ul><ul><li>Presence  from “polling” to “status report” </li></ul><ul><ul><li>special case of event notification </li></ul></ul><ul><ul><li>events as common infrastructure for services </li></ul></ul><ul><ul><li>location-based services </li></ul></ul><ul><li>Integration of IM and VoIP </li></ul><ul><ul><li>often used in same conference (side channel) </li></ul></ul><ul><ul><li>IM as initiator of real-time voice/video </li></ul></ul>
    31. 31. Near future: Location-based services <ul><li>Finding services based on location </li></ul><ul><ul><li>physical services (stores, restaurants, ATMs, …) </li></ul></ul><ul><ul><li>electronic services (media I/O, printer, display, …) </li></ul></ul><ul><ul><li>not covered here </li></ul></ul><ul><li>Using location to improve (network) services </li></ul><ul><ul><li>communication </li></ul></ul><ul><ul><ul><li>incoming communications changes based on where I am </li></ul></ul></ul><ul><ul><li>configuration </li></ul></ul><ul><ul><ul><li>devices in room adapt to their current users </li></ul></ul></ul><ul><ul><li>awareness </li></ul></ul><ul><ul><ul><li>others are (selectively) made aware of my location </li></ul></ul></ul><ul><ul><li>security </li></ul></ul><ul><ul><ul><li>proximity grants temporary access to local resources </li></ul></ul></ul>
    32. 32. Location-based IM & presence
    33. 33. User service creation <ul><li>Tailor a shared infrastructure to individual users </li></ul><ul><li>traditionally, only vendors (and sometimes carriers) </li></ul><ul><li>learn from web models </li></ul><ul><ul><li>not one “killer application” </li></ul></ul><ul><ul><li>grass-roots applications not foreseen by carriers </li></ul></ul>VoiceXML (voice), LESS VoiceXML end system CPL SIP servlets, sip-cgi network servers end user programmer, carrier
    34. 34. Near future: Multimedia <ul><li>Wideband audio </li></ul><ul><ul><li>“ better than phone quality”  lectures, discussions, speaker phone </li></ul></ul><ul><ul><li>better codecs  same bandwidth as existing NB codecs </li></ul></ul><ul><li>Video phone itself remains niche application </li></ul><ul><ul><li>given low incremental cost, may be viable </li></ul></ul><ul><ul><li>useful for sign language </li></ul></ul><ul><li>Video for group meetings </li></ul><ul><ul><li>capture whiteboard </li></ul></ul><ul><li>Shared applications (WebEx, etc.) </li></ul><ul><ul><li>still requires standardization </li></ul></ul><ul><li>Instant messaging </li></ul><ul><ul><li>side channel </li></ul></ul><ul><li>Better means of coordination (floor control) </li></ul>wideband audio
    35. 35. Near future: VoIP over WiFi <ul><li>Not fundamentally different from landline VoIP </li></ul><ul><ul><li>combination cellular + WiFi = wide-area + “cordless” phone </li></ul></ul><ul><li>Small packet sizes make VoIP over WiFi far less efficient than nominal data rate </li></ul><ul><li>Hand-off delay between different base stations  interruptions  CU modified hand-off algorithm </li></ul><ul><li>Delay jitter with high loads  new scheduling algorithms </li></ul><ul><li>L3 hand-off across different network types </li></ul>
    36. 36. Challenge: Global interconnect <ul><li>Currently, each VoIP “network” largely isolated </li></ul><ul><ul><li>interconnect via PSTN even if both endpoints are on IP </li></ul></ul><ul><ul><li>interconnect via few peering points even if neighbors </li></ul></ul><ul><li>Long-term solution: ENUM DNS listing </li></ul><ul><ul><li>administration appears difficult </li></ul></ul><ul><li>Short-term for pure-IP (FWD, etc.): special number prefixes </li></ul>GW GW VSP A Enterprise B
    37. 37. Challenge: CALEA (lawful intercept) <ul><li>Existing models assume congruence of signaling and voice flows </li></ul><ul><li>Challenges: </li></ul><ul><ul><li>voice service providers outside the US </li></ul></ul><ul><ul><li>signaling-only providers or no voice providers </li></ul></ul><ul><ul><li>end-to-end media and signaling encryption (Skype, SRTP) </li></ul></ul><ul><li> Intercept IP traffic, not application </li></ul><ul><li>Assume that long-term, all application traffic (except browsing of public web pages) will have strong encryption </li></ul>
    38. 38. Challenge: User-programmable and context-aware services <ul><li>Universal reachability  control reachability in time and space by context </li></ul><ul><ul><li>allow callee to decide reachability (defer and decline communication) </li></ul></ul><ul><ul><li>choose appropriate media (text, automated data response) </li></ul></ul>presence activity/availability not yet, but similar to location data sensor data (mood, bio) location-based call routing location events location caller preferences capabilities Call Processing Language (CPL), sip-cgi, … time
    39. 39. Challenge: Spam prevention <ul><li>Currently, telemarketing restricted to in-country calling </li></ul><ul><ul><li>With VoIP, few economical constraints on automated calls from anywhere </li></ul></ul><ul><ul><li>Also, SPIM (instant message spam) </li></ul></ul><ul><li>Cannot use content-based filtering </li></ul><ul><li>Public key infrastructure (PKI) for individual verification has never scaled </li></ul><ul><ul><li>provide domain-level verification (~ TLS) in signaling </li></ul></ul><ul><ul><li>blacklists and whitelists </li></ul></ul><ul><ul><ul><li>may depend on local domain policies for user verification </li></ul></ul></ul><ul><ul><li>reputation-based systems </li></ul></ul>
    40. 40. Challenge: Service reliability <ul><li>“ QoS”  service availability </li></ul><ul><ul><li>loss of network connection </li></ul></ul><ul><ul><li>loss of infrastructure components </li></ul></ul><ul><ul><ul><li>DNS, SIP servers, DHCP, … </li></ul></ul></ul><ul><ul><li>bursts of packet loss  cannot be repaired at end system </li></ul></ul><ul><ul><li>sustained high packet loss (> 10-15%) </li></ul></ul><ul><li>Current service availability probably around 99.5% </li></ul><ul><ul><li>realistic goal: 99.9% (10h/year) to 99.99% (1h/year) </li></ul></ul>
    41. 41. Challenge: Emergency calling <ul><li>911 calling system largely unchanged since 1980s </li></ul><ul><ul><li>call routing to appropriate destination </li></ul></ul><ul><ul><li>deliver caller location information </li></ul></ul><ul><li>Fundamental differences for VoIP: </li></ul><ul><ul><li>may not have phone number </li></ul></ul><ul><ul><li>may be no “phone company” </li></ul></ul><ul><ul><li>identifier does not describe location </li></ul></ul><ul><ul><li>location determination more difficult </li></ul></ul><ul><li>Also use solution for “311” and other location-based call routing systems </li></ul>
    42. 42. Three stages to VoIP 911 yes yes no callback number to PSAP? IP-enabled no (8 or 10 digit) no PSAP modification GNP multimedia international calls ALI not needed MSAG replaced by DNS location in-band yes stationary nomadic mobile no late 2004 I3 none update yes stationary nomadic no Dec. 2004 I2 none no no stationary allowed now I1 new services ALI (DB) modification caller location to PSAP? mobility use 10-digit admin. number? spec. available?
    43. 43. Prototype * gray features in progress.
    44. 44. Call taker setup SIPc client receives calls GeoLynx software displays caller location
    45. 45. Conclusion <ul><li>VoIP on cusp of widespread deployment: </li></ul><ul><ul><li>commercial-grade VoIP products </li></ul></ul><ul><ul><li>mature standards for key components </li></ul></ul><ul><ul><li>widespread broadband availability </li></ul></ul><ul><ul><li>better Internet QoS </li></ul></ul><ul><li>Focus may shift from “bare-bones” VoIP to context-aware communications </li></ul><ul><li>Operational and technical challenges </li></ul><ul><ul><li>911, CALEA, network reliability, user-defined services, multimedia </li></ul></ul><ul><li>Thus, roughly where PSTN was in 1980  </li></ul>

    ×