Mobile IPv6 & Cellular Telephony - PowerPoint PPT Presentation

jaden
mobile ipv6 cellular telephony l.
Skip this Video
Loading SlideShow in 5 Seconds..
Mobile IPv6 & Cellular Telephony PowerPoint Presentation
Download Presentation
Mobile IPv6 & Cellular Telephony

play fullscreen
1 / 36
Download Presentation
Mobile IPv6 & Cellular Telephony
326 Views
Download Presentation

Mobile IPv6 & Cellular Telephony

- - - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript

  1. Mobile IPv6 & Cellular Telephony Charles E. Perkins Nokia Research Center Mountain View, CA USA http://www.iprg.nokia.com/~charliep charliep@iprg.nokia.com

  2. Why Mobile IP? • Both ends of a TCP session (connection) need to keep the same IP address for the life of the session. • IP needs to change the IP address when a network node moves to a new place in the network. Mobile IPv4 changes the mobility problem into a routing problem

  3. Mobile IPv4 protocol overview Home Agent Foreign Agent • Advertisement from foreign agent • Seamless Roaming: mobile node keeps home address • Foreign agent offers care-of address • Mobile Node “always on” by way of home agent correspondent node 178.24.9.36

  4. The Mobile IP(v4) solution • Mobile node always uses the same IP address (called the home address) for communication   • The care-of address is used for routing • The home agent manages home network operations for the mobile node while it is away from home: • encapsulation • proxy ARP • Specified in RFCs 2002-2006

  5. Foreign Agents & Triangles • The foreign agent advertises the care-of address, and terminates the tunnel from the home network • All traffic to the mobile node is sent to the mobile node's home address. Traffic from the mobile node does not have to traverse the home network. • This leads to the phenomenon of triangle routing.

  6. Ingress Filtering • Ingress filtering routers prevent packets from entering the Internet unless the source IP address is topologically correct. • This leads to the possibility of reverse tunneling (RFC 2344) from the care-of address to the home agent.

  7. Cellular architectures • Involve SS7 over "control plane" to set up virtual circuits for "user plane" traffic • Are highly optimized for voice traffic (low delay, guaranteed bandwidth), not data • Tend toward "intelligent network" philosophy which for IP is a misplaced locus of control. • Operators want to migrate towards "All-IP" solutions (whatever that means…).

  8. Mobile IPv6 Design Points • Enough Addresses • Enough Security • Address Autoconfiguration • Route Optimization • Destination Options • Reduced Soft-State

  9. Mobile IPv6 protocol overview Home Agent Local Router • Advertisement from local router • Seamless Roaming: mobile node keeps home address • Address autoconfiguration for care-of address • Binding Updates sent to correspondent nodes • Mobile Node “always on” by way of home agent correspondent node charliep@nokia.com

  10. Enough Addresses • 340 undecillion addresses (340,282,366,920,938,463,463,374,607,431,768,211,456) total • Billions of IP-addressable wireless handsets • Address space crunch is already evident • recent unfulfilled request to RIPE • Multi-level NAT unknown/unavailable • Even more addresses for embedded wireless • Especially interesting for China now

  11. Enough Security (almost) • Authentication Header • Needed for Binding Update • Remote Redirect problem • Encapsulating Security Payload • Required from every IPv6 node • Key distribution still poorly understood • PKI? • AAA?

  12. Address Autoconfiguration • A new care-of address onevery link • Stateless Address Autoconfiguration • Link-Local Address  Global Address • Stateful Autoconfiguration (DHCPv6) • Movement Detection Routing Prefix MAC address

  13. Destination Options • Binding Updates without control packets • allows optimal routing • replaces IPv4 Registration Request messages • Home Address option • better interaction with ingress filtering • supported by all IPv6 network nodes • Binding Acknowledgement • replaces Registration Reply

  14. Route Optimization • Reduces network load by ~50% • (depending on your favorite traffic model) • Most Internet devices will be mobile • Route Optimization could double Internet-wide performance levels! • Binding Update SHOULD be part of every IPv6 node implementation

  15. Improved ICMP messages • IPv4 ICMP returns only 8 payload bytes • IPv4 home agents could not relay errors • insufficient inner header information • some data sources might never find out about broken links • IPv6 ICMP messages return enough data • Also used for anycast home agent discovery

  16. Mobile IPv6 status • Interactions with IPsec fully worked out • Mobile IPv6 testing event Sept 15-17 • Bull, Ericsson, NEC, INRIA • Connectathon last month – success! • Internet Draft is ready for Last Call • Another bake-off likely by fall

  17. AAAv4 and Cellular Telephony • Terminology • Protocol overview • Key Distribution • Scalability and Performance • IETF Status • How can we make it work for Mobile IPv6?

  18. Terminology • Authentication – verifying a node’s identity • Authorization – for access to resources • according to authentication and policy • Accounting – measuring utilization • Network Access Identifier (NAI) – user@realm • Challenge – replay protection from local attendant • AAAF for foreign domain • AAAH for home domain

  19. AAA & Mobile IP protocol overview AAAF AAAH • Advertisement from local attendant (e.g., router) • Connectivity request w/MN-NAI from Mobile Node • Local Attendant asks AAAF for help • AAAF looks at realm to contact AAAH • AAAH authenticates & authorizes, starts accounting • AAAH, optionally, allocates a home address • AAAH contacts & initializes Home Agent Local Attendant Home Agent charliep@nokia.com

  20. Key Distribution • New security model • just one SA: mobile node  AAAH • So, association needed HA  mobile node • TR45.6, others, want also: • local attendant  mobile node • AAAH can allocate the keys for this

  21. AAAH AAAF Home Agent Local Attendant Brokers • Needed when there are 1000’s of domains • NAI is perfect to enable this • AAAF decides whether to use per realm • may prefer bilateral arrangement • iPASS, GRIC

  22. Scalability and Performance • Single Internet Traversal • Brokers • Eliminate all unnecessary AAA interaction • Handoff between local attendants (routers) • can use keys from previous router • Regional Registration • HA can use single care-of address per domain

  23. Mobile IP/AAA Status • AAA working group has been formed • Taking a page from the existing model with RADIUS • Mobile IP (v4) AAA requirements draft • Last Call “over” • Several 3G requirements documents online • Mobile IP/AAA extensions draft • AAAv6 Internet Draft(s) submitted • stateless and stateful variations

  24. Other features (needed for IPv6) • Routers used instead as mobility agents • Regional registration • eliminates most location update traffic • GGSNs/border routers are candidates • UDP Lite • Robust Header Compression • AAA  HLR adaptation layer • Challenge generation (not from HLR?) • Privacy considerations

  25. Hierarchical Foreign Agents GFA Home Agent LFA Home Agent stores GFA address as the Care-of Address Mobile Node registers only once with Home Agent Mobile node registers locally with GFA Often, only one level of hierarchy is being considered

  26. HSS 3GPP with GPRS Evolution from cellular packet/GPRS Mobility agent At GGSN Subscription andLocation Directory BSS PSTN CPS/GK GGSN GW BSC/RNC SGSN GPRS Internet Call Processing Server/Gatekeeper Traditional BSS withpacket data QoS enhancements

  27. Evolution from general IP networks Subscriberdatabase AAA Server AAA Server HA HA(mobility within serving ntw) "Slim RNC/BSC" FW CPS FW Internet, Intranets GW One (of many) “ALL-IP” visions PSTN

  28. Subscriberdatabase AAA Server HA CDMA2000 3G micromobility AAA Server RNN PDSN

  29. CDMA2000 3G micromobility • Terminate physical layer distant from “FA” • Protected, private n/w between FA and MN • PDSN (Packet Data Serving Node) ~ GFA • RNN (Radio Network Node) ~ LFA • RNN manages the physical layer connection to the mobile node

  30. CDMA2000 3G Requirements • GRE encapsulation (but will it survive?) • Reverse Tunneling (RFC 2344) • Registration Update • Registration Acknowledge • Session-specific registration extension • contains MN-ID, type, MN Connection-ID • contains Key field for GRE

  31. CDMA2000 Registration Update • Used for handovers to new RNN • Acknowledgement required • allows PDSN/old RNN to reclaim resources • New authentication extension required • Home address  0 • Home agent  PDSN • Care-of address  RNN

  32. IMT-2000/UMTS/EDGE reqt’s • Independent of access technology • so should work for non-GSM also • Interoperation with existing cellular • Privacy/encryption (using IPsec) • QoS for Voice/IP and videoconferencing • particular concern during handover • Fixed/mobile convergence desired

  33. IMT-2000 reqt’s, continued • Charge according to QoS attribute request • Roaming to diverse access technologies • e.g., Vertical IP • Route optimization • Identification/authorization based on NAI • Proxy registration for legacy mobile nodes • Signaling for firewall traversal

  34. IMT-2000 reqt’s, continued • Reverse tunneling • Private networks • but, still allow access to networks other than the mobile node’s home network • Dynamic home address assignment • Dynamic home agent assignment • even in visited network • even when roaming from one visited network to another

  35. IPv6 status for cellular telephony • Seems highly likely to be mandated for 3GPP, with IPv4 support optional only • MWIF recommendation for IPv6 • 3GPP2 study group favorable towards IPv6 • Seems difficult to make a phone call to a handset behind a NAT (not impossible, just expensive and cumbersome and protocol-rich)

  36. Summary and Conclusions • Future Internet is largely wireless/mobile • IPv6 needed for billions of wireless devices • Mobile IPv6 is far better and more efficient • Autoconfiguration suitable for the mobile Internet • Security is a key component for success • AAA has a big role to play for cellular rollout • Leverage from current cellular interest