One document matched: draft-lee-speermint-use-case-cable-00.txt


Internet-Draft    Session Peering Use Case for Cable     June 19, 2006 
 
 
Network Working Group                                            Y. Lee 
Internet-Draft                                            Comcast Cable 
Expires: December 19, 2006                                    June 2006 
                                                                        
    
    
                    Session Peering Use Case for Cable 
                 draft-lee-speermint-use-case-cable-00.txt 
    
    
Status of this Memo 
    
   By submitting this Internet-Draft, each author represents that any 
   applicable patent or other IPR claims of which he or she is aware 
   have been or will be disclosed, and any of which he or she becomes 
   aware will be disclosed, in accordance with Section 6 of BCP 79. 
    
   Internet-Drafts are working documents of the Internet Engineering 
   Task Force (IETF), its areas, and its working groups.  Note that 
   other groups may also distribute working documents as Internet-
   Drafts. 
    
   Internet-Drafts are draft documents valid for a maximum of six months 
   and may be updated, replaced, or obsoleted by other documents at any 
   time.  It is inappropriate to use Internet-Drafts as reference 
   material or to cite them other than as "work in progress." 
    
   The list of current Internet-Drafts can be accessed at 
   http://www.ietf.org/ietf/1id-abstracts.txt 
   The list of Internet-Draft Shadow Directories can be accessed at 
   http://www.ietf.org/shadow.html. 
    
   This Internet-Draft will expire on December 19, 2006.  
    
Copyright Notice 
    
   Copyright (C) The Internet Society (2006). 
    
    
Abstract 
    
   This document describes a typical use case of session peering in 
   cable industry. Caller Alice makes a VoIP call to Callee Bob. Alice 
   and Bob are served by two different cable operators, mso-o and mso-t. 
   mso-o and mso-t have bi-lateral peering agreement to peer at SIP 
   layer. This document focuses on the SIP layer interactions and 
   discuss some common practices for SIP Peering in cable industry. 
    
    
 
 
Lee                   Expires December 19, 2006              [Page 1] 

Internet-Draft    Session Peering Use Case for Cable     June 19, 2006 
 
 
 
 
Table of Contents 
    
   1. Introduction...................................................3 
   2. Terminology....................................................3 
   3. User Setup.....................................................5 
   4. Network Setup..................................................5 
   5. Call Setup.....................................................6 
   6. User Location Layer............................................9 
   7. Session Routing Layer.........................................10 
      7.1 Topology Hiding Interworking Gateway Function.............10 
      7.2 Network Address Translation Function......................10 
      7.3 IPv4/IPv6 Interworking Function...........................12 
   8. Future Works..................................................13 
      8.1 Peering Policy............................................13 
      8.2 Peering Location Function.................................13 
      8.3 Peering Security..........................................13 
      8.4 Peering QoS...............................................14 
      8.5 Peering Accounting and Billing............................14 
   9. Security Considerations.......................................14 
   10. IANA Considerations..........................................14 
   11. Acknowledgements.............................................15 
   12. References...................................................15 
      12.1 Normative References.....................................15 
      12.2 Informative References...................................16 
   Authors’ Addresses...............................................16 
   Intellectual Property and Copyright Statements..Error! Bookmark not 
   defined. 
    
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Lee                   Expires December 19, 2006              [Page 2] 

Internet-Draft    Session Peering Use Case for Cable     June 19, 2006 
 
 
1. 
  Introduction 
    
   The purpose of this document is to outline the current best practice 
   use case for establishing interconnection of MSO/Cable service 
   Providers for delivery of SIP call termination over those 
   interconnections. These interconnections are to establish real-time 
   sessions between SIP servers at layer 5 network.  While voice calls 
   are the primary motivation for this today, other forms of real-time 
   communications are and will continue to evolve as natural additions 
   to such real-time sessions. This document depicts the network setup 
   and the steps involved in the call flow from a caller in originating 
   MSO network to a callee in another terminating MSO network, by using 
   Call Routing data (CRD) [1] obtained though ENUM services. The 
   scenario is shown in the figure below; Alice calls Bob where Alice 
   and Bob are served by two different cable operators, MSO-o and MSO-t, 
   respectively. Both MSOs connect to an ENUM [1] server that provides 
   ENUM service. Both MSOs have full Layer 3 connectivity. We make no 
   assumption whether they directly peer to each other or through any 
   Layer 3 transit network. This document describes the Layer 5 Peering 
   interactions when Alice calls Bob. 
    
    
2. 
  Terminology 
    
    
   Figure 1 shows the logical entities involved in peering. 
    
    
    
        User Location Layer  
 
             +--------+               \               +--------+  
             | ENUM-o |-----------|   /   |-----------| ENUM-t | 
             +--------+           |   \   |           +--------+ 
                                  |   /   |                  
                                  |   \   |  
             +--------+           |   /   |           +--------+  
             | DNS-o  |--------|  |   \   |  |--------| DNS-t  |  
             +--------+        |  |   /   |  |        +--------+  
                   \           |  |   \   |  |            /  
      --------------\----------|--|-------|--|-----------/------------  
        Session      \         |  |   /   |  |          /  
        Routing Layer \        |  |   \   |  |         /  
                       \       |  |   /   |  |        /  
                    +-------+  |  |   \   |  |  +-------+  
                    | PP-o  |-------------------| PP-t  |  
                    +-------+  |  |   \   |  |  +-------+  
                        |      |  |   /   |  |      |  
                        |      |  |   \   |  |      |  
 
 
Lee                   Expires December 19, 2006              [Page 3] 

Internet-Draft    Session Peering Use Case for Cable     June 19, 2006 
 
 
                    +-------+  |  |   /   |  |  +-------+  
       +-------+    |       |--|  |   \   |  |--|       |    +-------+  
       | UE-o  |----| SM-o  |     |   /   |     | SM-t  |----| UE-t  |  
       +-------+    |       |-----|   \   |-----|       |    +-------+  
                    +-------+         /         +-------+      
                                      \  
                MSO-o                 /                 MSO-t  
    
    
                                   Figure 1 
    
    
   ENUM Server: An ENUM server stores the ENUM information and provides 
   an interface for ENUM query for peering cable operators. The input to 
   server is an E.164 number and the output is the NAPTR record. The 
   ENUM client resolves the NAPTR record to formulate a sip uri 
   associated to the input E.164 number. This ENUM server can be the 
   Public ENUM server that hosts namespace "e164.arpa" [1] or 
   Infrastructure ENUM server that hosts namespace "ie164.arpa" [2]. 
    
   Using Public or Infrastructure ENUM is a business decision. Some 
   cable operators MAY deploy Infrastructure ENUM for peering in the 
   initial stage and migrate to Public ENUM when they see the need. In 
   this document, the only technical requirement for the ENUM server is 
   that it can return the associated NAPTR that can be resolved to a sip 
   uri of the users for peering. 
    
   ENUM-o: The ENUM server in the originating network. 
    
   ENUM-t: The ENUM server in the terminating network. 
    
   DNS Server [3]: DNS resolves the domain part of the sip uri to an IP 
   address so that SM or PP can route the Request and Response to the 
   target. 
    
   DNS-o: The DNS server in the originating network. 
    
   DNS-t: The DNS server in the terminating network. 
    
   Session Manager (SM): A SM is the entity responsible for sending and 
   receiving the SIP messages from or to Peer Proxy (PP). It is also 
   responsible for locating the user home proxy. SM is logical, it MAY 
   contain one functional entity or multiple functional entities. For 
   example, SM can be the P-CSCF, I-CSCF and S-CSCF defined in IMS [20]. 
   SM can also be the Call Manager Server (CMS) defined in PacketCable 
   (PC) 1.5 [19].  
    
   SM-o: The SM originates the call. In this content, it is Alice's SM. 
    
 
 
Lee                   Expires December 19, 2006              [Page 4] 

Internet-Draft    Session Peering Use Case for Cable     June 19, 2006 
 
 
   SM-t: The SM terminates the call. In this content, it is Bob's SM. 
    
   Peer Proxy (PP): A PP is the entity that peers to the external 
   network. It is a sip proxy that enforces peering policy. In most 
   setup, PP has a trusted relationship with the remote PP, so the 
   communication channel between two PPs SHOULD be secured by some sort 
   of security mechanisms such as IPSec [20] or TLS [21]. Optionally, PP 
   MAY provide additional functions such as Topology Hiding Interworking 
   Gateway function (THIG), Network Address Translation (NAT) function, 
   and SIP header normalization. 
    
   PP-o: The PP connects the SM-o and the remote PP. 
    
   PP-t: The PP connects the SM-t and the remote PP. 
    
   User Endpoint (UE): User Endpoint is the client that makes or 
   receives calls. UE can be sip based or non-sip based. For non-sip 
   based UE, SM acts as a signaling gateway and translates the non-sip 
   signaling to sip signaling before sending to PP. 
    
   UE-o: Alice's Originating UE. 
    
   UE-t: Bob's Terminating UE. 
    
    
3. 
  User Setup 
    
   Alice signs up a VoIP service with MSO-o. MSO-o assigns her a 
   globally unique E.164 number +1-215-111-2222. Also, MSO-o assigns her 
   an ENUM entry where +1-215-111-2222 maps to NAPTR record that 
   formulates sip uri <sip:alice@mso-o.com>. For Public ENUM, the E.164 
   number is in namespace e164.arpa. If MSO-o supports only 
   Infrastructure ENUM for peering, the E.164 number is in namespace 
   ie164.arpa. 
    
   Bob signs up with MSO-t and his globally unique E.164 number is +1-
   212-333-4444. MSO-t assigns him an ENUM entry where +1-212-333-4444 
   maps to a NAPTR record that formulates sip uri <sip:bob@mso-t.com>. 
   For Public ENUM, the E.164 number is in namespace e164.arpa. If MSO-t 
   supports only Infrastructure ENUM for peering, the E.164 number is in 
   namespace ie164.arpa. 
    
    
4. 
  Network Setup 
 
   In Figure 1, we divide the diagram into 2 layers: (1) User Location 
   Layer and (2) Session Routing Layer. User Location Layer is 
   responsible for locating the network serving the terminating UE. It 

 
 
Lee                   Expires December 19, 2006              [Page 5] 

Internet-Draft    Session Peering Use Case for Cable     June 19, 2006 
 
 
   includes ENUM server and DNS server. Each of them provides different 
   services. 
    
   ENUM server accepts an E.164 number as input and returns a NAPTR 
   record to the ENUM client as output. ENUM client parses the regular 
   expression and formulates the sip uri associated to the input E.164 
   number. DNS server accepts a FQDN as input and returns either a SRV 
   record [4] or an A-Record as output. In the diagram, SM has the 
   interface to interact with both ENUM and DNS servers. PP has the 
   interface to interact with DNS server only. 
    
   The actual SIP routing happens in the Session Routing Layer. It 
   includes UE-o, SM-o, PP-o, UE-t, SM-t and PP-t. UE-o and UE-t are sip 
   clients which can make VoIP call. 
    
   SM-o and SM-t are the home SIP proxies to UE-o and UE-t. SM-o and SM-
   t are enable to perform normal SIP routing operations defined in [5]. 
   In addition, it has an interface to access user profile data 
   associated to the registered user for authentication and 
   authorization. They also have ENUM and DNS clients built-in. They can 
   issue ENUM query and formulate uri from the NAPTR records. SM makes 
   routing decision based on the user profile information and the 
   request URI. 
    
   PP-o and PP-t are the peering proxies where the actual peering 
   happens. PP-o connects the SM-o to the remote PP-t. PP-o is the last 
   point in MSO-o's domain. PP-o is responsible for establishing the 
   peering relation to PP-t. MSO-o and MSO-t SHOULD have signed bi-
   lateral agreement. All the necessary peering policies and security 
   measurements such as THIG function and NAT function SHOULD be 
   performed in PP. In the diagram, SIP messages flow between:  
    
           (UE-o)<->(SM-o)<->(PP-o)<->(PP-t)<->(SM-t)<->(UE-t) 
    
   We do not show the media in the diagram. Media can flow from UE-o to 
   UE-t directly or through some media proxy for NAT or media 
   transcoding. 
    
    
5. 
  Call Setup 
    
   Alice is a user served by MSO-o. She has a sip phone registered to 
   SM-o. She has an E.164 number +1-215-111-2222 and a public sip uri 
   <sip:alice@mso-o.com>. She picks up the phone and calls Bob. She 
   enters Bob's TN number +1-212-333-4444 into her key pad. Alice UE-o 
   initiates an INVITE with Bob's global unique tel uri which is 
   <tel:+1-212-333-4444> [6] in the request URI.  
    

 
 
Lee                   Expires December 19, 2006              [Page 6] 

Internet-Draft    Session Peering Use Case for Cable     June 19, 2006 
 
 
   SM-o receiving the SIP INVITE request SHOULD process it according to 
   the following logic: 
    
   1. Perform an ENUM query on the called party in the SIP request URI. 
    
   2. If the ENUM server fails to return the response, SM-o forwards the 
   call to PSTN. 
    
   3. ENUM server returns a NAPTR record. SM-o parses the regular 
   expression and formulates the sip uri of Bob which is <sip:bob@mso-
   t.com>. 
    
   4. SM-o finds out that it does not own "mso-t.com". SM-o has local 
   policies to send the request to PP-o. 
    
   5. SM-o sends a DNS query to locate PP-o’s IP address. 
    
   6. DNS returns PP-o’s IP address to SM-o. SM-o sends the SIP INVITE 
   to PP-o. SM-o MAY choose to record-route to stay on the signaling 
   path. 
 
   7. PP-o receives the SIP INVITE. It examines the request URI and 
   sends a query to DNS server to get the IP address of Bob’s domain 
   "mos-t.com". 
    
   8. PP-o performs all the necessary operations such as sip header 
   normalization and THIG function and sends the INVITE to PP-t. 
   Optionally, PP-o MAY act as a B2BUA. This is necessary when PP-o 
   provides NAT function or NAT-PT function. Section 7.2 and 7.3 
   describes the steps. 
    
   9. PP-t receives the INVITE. It examines the request URI to verify 
   the domain is one of its serving domains. If it is, PP-t will forward 
   the INVITE to some proxy that has access to Bob's user data to locate 
   Bob’s home proxy. In the diagram, SM-t is logical to provide the user 
   location function. Based on the user profile information, SM-t MAY 
   re-write the request URI to something more location specific. For 
   example, SM-t knows that Bob's home proxy is the San Jose proxy, so 
   it re-writes the request URI to <sip:bob@sanjose-proxy.mso-t.com> to 
   the INVITE and deliver the message to the San Jose proxy directly. 
   This location service is internal to the domain. MSO-t MAY use 
   internal DNS or some other proprietary methods to retrieve the 
   location information. MSO-t chooses the method best fit to the 
   internal architecture.  
    
   10. SM-t receives the SIP INVITE. SM-t contains the registration 
   information of Bob’s UE-t. This is the home proxy which hosts the 
   contact information of Bob’s UE-t. SM-t forwards the SIP INVITE 
   request to UE-t. 
 
 
Lee                   Expires December 19, 2006              [Page 7] 

Internet-Draft    Session Peering Use Case for Cable     June 19, 2006 
 
 
    
   11. Bob's UE-t receives the SIP INVITE request. Bob accepts the call. 
   UE-t sends the 200OK and Alice acknowledges it. 
    
   12. Alice and Bob starts 2-way conversation.  
    
    
   Figure 2 illustrates the message interactions: 
    
    
    
      UE-o  SM-o  PP-o   DNS-o   ENUM  DNS-t  PP-t   SM-t   UE-t 
      |      |      |      |      |      |      |      |      |  
      |INVITE|      |      |      |      |      |      |      |  
      |----->|      |      |      |      |      |      |      |  
      |      |     ENUM Query     |      |      |      |      |  
      |      |------------------->|      |      |      |      |  
      |      |     ENUM Response  |      |      |      |      |  
      |      |<-------------------|      |      |      |      |  
      |      |  DNS Query  |      |      |      |      |      |  
      |      |------------>|      |      |      |      |      |  
      |      | DNS Response       |      |      |      |      |  
      |      |<------------|      |      |      |      |      |  
      |      |INVITE|      |      |      |      |      |      |  
      |      |----->|      |      |      |      |      |      |  
      |      |      DNS Query     |      |      |      |      |  
      |      |      |----->|      |      |      |      |      |  
      |      |    DNS Response    |      |      |      |      |  
      |      |      |<-----|      |      |      |      |      |  
      |      |      |      |  INVITE     |      |      |      |  
      |      |      |-------------------------->|      |      |  
      |      |      |      |      |      |      |INVITE|      |  
      |      |      |      |      |      |      |----->|      |  
      |      |      |      |      |      |      |      |INVITE|  
      |      |      |      |      |      |      |      |----->|  
      |      |      |      |      |      |      |      |200OK |  
      |      |      |      |      |      |      |      |<-----|  
      |      |      |      |      |      |      | 200OK|      |  
      |      |      |      |      |      |      |<-----|      |  
      |      |      |      |    200OK    |      |      |      |  
      |      |      |<--------------------------|      |      |  
      |      | 200OK|      |      |      |      |      |      |  
      |      |<-----|      |      |      |      |      |      |  
      | 200OK|      |      |      |      |      |      |      |  
      |<-----|      |      |      |      |      |      |      |  
      | ACK  |      |      |      |      |      |      |      |  
      |----->|      |      |      |      |      |      |      |  
      |      |      |      |      |      |      |      |      |  
      |      | ACK  |      |      |      |      |      |      |  
 
 
Lee                   Expires December 19, 2006              [Page 8] 

Internet-Draft    Session Peering Use Case for Cable     June 19, 2006 
 
 
      |      |----->|      |      |      |      |      |      |  
      |      |      |      |     ACK     |      |      |      |  
      |      |      |-------------------------->|      |      |  
      |      |      |      |      |      |      | ACK  |      |  
      |      |      |      |      |      |      |----->|      |  
      |      |      |      |      |      |      |      | ACK  |  
      |      |      |      |      |      |      |      |----->|  
      |      |      |      |      |      |      |      |      |  
      |      |      |      |2-Way Media  |      |      |      |  
      |<=====================================================>|  
      |      |      |      |      |      |      |      |      |  
      |      |      |      |      |      |      |      |      |  
      |      |      |      |      |      |      |      |      |  
    
                               Figure 2 
    
6. 
  User Location Layer 
      
   In the call flow shown in Figure 2, when PP-o receives the SIP INVITE 
   request from SM-o, PP-o queries DNS to resolve the IP address of the 
   domain "mso-t.com". PP-o MAY choose not to query DNS server to 
   resolve "mso-t.com". By examining the domain part of Bob's sip uri, 
   SM-o knows that "msoB.com" is one of its trusted peer. In many cases, 
   PP-o's configuration will have static configuration pointing to a 
   static IP address associated to PP-t. There is number of reasons to 
   have this setup. Most common reason is security such that PP-o only 
   peers to the pre-configured IP address. In this setup, PP-o MAY skip 
   querying DNS to resolve the domain name of the remote target. That 
   said, it does not stop PP-o to use DNS to resolve the domain name. 
    
   Only SM has an interface to ENUM server to resolve the E.164 number 
   to sip uri. When SM-o queries the ENUM server and realizes that Bob 
   resides in a different domain, SM-o will re-writes the request URI 
   from Bob's tel uri to Bob's sip uri before sending the request to PP-
   o. 
    
   When PP-o sends a query to the DNS for "mso-t.com", it MAY return an 
   A-record or a SRV record of PP-t. Hence, PP-o MUST prepare to accept 
   a SRV record and try to reach the available PP-t in the returned 
   list. Once PP-o selects a PP-t, it SHOULD stick with the same PP-t 
   for the duration of the call. This is important because peering 
   policies MAY vary from session to session. So, PP-t will contain the 
   peering state of that particular session. 
    
    




 
 
Lee                   Expires December 19, 2006              [Page 9] 

Internet-Draft    Session Peering Use Case for Cable     June 19, 2006 
 
 
7. 
  Session Routing Layer 
 
   Session Routing Function performs generic SIP routing function. With 
   regard to session peering in cable environment, there are few 
   specific functions that cable operators MAY consider to support. 
    
    
7.1 
   Topology Hiding Interworking Gateway Function 
    
   In the case PP-o performs THIG. PP-o SHOULD remove the proxies 
   written in Via and Record-Route headers and replace itself to the Via 
   and Record-Route headers. When PP-o sends a message to PP-t, it will 
   look the same as PP-o is the only proxy in MSO-o. Similarly, when PP-
   t sends a message to PP-o, the message will look the same as PP-t is 
   the only proxy in MSO-t. Alternately, PP-o MAY act as B2BUA such that 
   it is the UAC to the peer. 
    
    
7.2 
   Network Address Translation Function 
 
   In Figure 2, we assume that the UE-o and UE-t use public routable IP 
   addresses so that they can establish direct peer-to-peer 2-way 
   conversation. However, many cable operators use RFC 1918 [16] 
   addresses for their UEs. Since those addresses are not routable 
   outside its domain, UE-o and UE-t require some way to perform NAT 
   function. NAT is problematic in SIP. Detailed description can be 
   found in [7]. The NAT function can happen in two places, it can 
   happen in the edge layer or in the network layer. Either way, the 
   network MUST pass the NAT information to the session layer. This 
   requires some form of communications between the session layer and 
   network layer. There are several protocols [7,8,9] being worked out 
   in IETF. 
    
   If UE is aware of NAT, it will be responsible for putting the public 
   transport address in the SIP/SDP. UE MAY use ICE 8] to discover the 
   best possible way such as STUN [7] or TURN [9] to overcome NAT. 
   However, this requires both UEs to support ICE. ICE runs a STUN 
   server per transport address, this adds significant load to UE. In 
   today cable environment, the most common UE is the Embedded Media 
   Termination Adaptor (eMTA), they have limited memory and processing 
   power, so they MAY require hardware upgrade to support ICE. 
    
   If UE is unaware any NAT, it will simply put its RFC 1918 address in 
   the SIP/SDP and sends the SIP message to SM. It relies on the network 
   to perform the NAT function. Consider a UE-o wants to make a call to 
   UE-t, UE-o uses RFC 1918 address. In this setup, the originating MSO-
   o is responsible for NAT function. The NAT function MAY happen in the 
   access network or at the network border. Regardless where it happens, 
   MSO-o MUST replace the RFC 1918 address in the session layer before 
 
 
Lee                   Expires December 19, 2006             [Page 10] 

Internet-Draft    Session Peering Use Case for Cable     June 19, 2006 
 
 
   sending the SIP message to MSO-t. MSO-t also needs to relay the media 
   packets before sending the traffic to UE-t. Since it is not well 
   defined how to pass the NAT information between network layer and 
   session layer, most cable operators chooses PP to perform the NAT 
   function. Figure 3 shows the network setup.  
    
    
                                       / 
                   +-------+ call-leg-2\       +-------+  
                   | PP-o  |-------------------| PP-t  |  
                   +-------+           \       +-------+  
         call-leg-1   |   \            /           |  
                      |    \undefined  \           |  
                   +-------+\          /       +-------+  
      +-------+    |       | \         \       |       |    +-------+  
      | UE-o  |----| SM-o  |  \        /       | SM-t  |----|  UE-t |  
      +-------+    |       |   |       \       |       |    +-------+  
          ||       +-------+   |       /       +-------+        || 
          ||                   |       \                        || 
          ||         Priv +-------+ Pub/                        || 
          ||==============| Media |=============================|| 
                  RTP     | Relay |    \          RTP 
                          |  GW   |    / 
                          +-------+    \ 
                                       / 
                         MSO-o         \           MSO-t 
                                       / 
    
                                   Figure 3 
    
    
   In this setup, PP-o acts as a B2BUA. When PP-o receives the SIP 
   INVITE request, it terminates the INVITE (Call-Leg-1) and creates a 
   new INVITE (Call-Leg-2) to relay the header information to MSO-t. PP-
   o creates the Private-to-Public address binding between the internal 
   and external networks and perform any necessary address translation 
   in the SIP header. The address translation of signaling happens in 
   PP-o, the address translation of media MAY happen in a different 
   physical entity. To allow this, PP-o and the Media Relay Gateway 
   require to exchange Private-to-Public address binding information. 
   UE-o sees PP-o the UAS and forwards all the SIP messages to PP-o. UE-
   t sees PP-o the UAC and forwards all the SIP messages to PP-o. Media 
   passes through the Media Relay Gateway in MSO-o for NAT binding for 
   the media stream. 
    
    
    
    

 
 
Lee                   Expires December 19, 2006             [Page 11] 

Internet-Draft    Session Peering Use Case for Cable     June 19, 2006 
 
 
7.3 
   IPv4/IPv6 Interworking Function 
    
   Some cable operators are actively working on IPv6 [10]. This allows 
   an IPv6 device to register to SM. Many UEs in the market support 
   IPv4/IPv6 dual stacks. During provisioning, the cable operator MAY 
   offer IPv4, IPv6 or both addresses to it. For the discussion here, we 
   restrict that a UE can choose to register with either an IPv4 or an 
   IPv6 address [11]. In other words, a UE can only register to SM with 
   one IP address, either an IPv4 or an IPv6 address. During IPv4/IPv6 
   transition [12], the cable operator which runs IPv4/IPv6 dual stacks 
   (MSO6) will probably peer with many IPv4 only peers. When setting up 
   sessions with them, MSO6 MUST perform all the necessary translations 
   inside the MSO6’s network. IPv4 peer cable operator (MSO4) does not 
   understand IPv6 address. From the MSO4 point of view, it sees MSO6 an 
   IPv4 network. 
    
   Consider an example, an IPv6 device (UE6-o) wants to make a call to 
   an IPv4 device (UE4-t). UE6-o registers to a cable operator which 
   runs dual stacks (MSO6-o). UE4t registers to an IPv4 cable operator 
   (MSO4-t). Figure 4 shows the network setup.  
    
    
                                       / 
                   +-------+ call-leg-2\       +-------+  
                   | PP-o  |-------------------| PP-t  |  
                   +-------+ IPv4      \       +-------+  
           Call-leg-1 |   \            /           |  
              IPv6    |    \undefined  \           |  
                   +-------+\          /       +-------+  
      +-------+    |       | \         \       |       |    +-------+  
      | UE6-o |----| SM-o  |  \        /       | SM-t  |----| UE4-t |  
      +-------+    |       |   |       \       |       |    +-------+  
          ||       +-------+   |       /       +-------+        || 
          ||                   |       \                        || 
          ||         IPv6 +-------+ IPv4                        || 
          ||==============| Media |=============================|| 
                  RTP     | Relay |    \          RTP 
                          |  GW   |    / 
                          +-------+    \ 
                                       / 
                         MSO6-o        \          MSO4-t 
                      (mso-o.com)      /       (mso-t.com) 
    
                                   Figure 4 
    
    
   To form a session between UE6o and UE4t, MSO6-o MUST translate UE6-
   o’s IPv6 address to an IPv4 address. This translation (NAT-PT) [17] 
   is similar to NAT function discussed in Section 7.2. PP-o performs 
 
 
Lee                   Expires December 19, 2006             [Page 12] 

Internet-Draft    Session Peering Use Case for Cable     June 19, 2006 
 
 
   any necessary IPv6-to-IPv4 address translation. When PP-o receives 
   the INVITE from SM-o, it sends a DNS query for domain "mso-t.com". 
   Since MSO4-t supports only IPv4, the DNS will return an IPv4 address 
   to PP-o. Upon receiving the response, PP-o realizes that it needs to 
   perform IPv4/IPv6 interworking function. PP-o allocates IPv4 
   addresses and ports from its IPv4 address pool and creates the IPv6-
   to-IPv4 address binding. It also instructs the Media Relay Gateway to 
   do the same for media relay. 
    
    
8. 
  Future Works 
    
   This document illustrates a simple use case for session peering in 
   cable industry. We describe the major entities that participate the 
   peering. We also outline the high-level interactions between these 
   entities. From the interactions, we see some areas for future work. 
    
   - Peering Policy 
    
   - User Location Service 
    
   - Peering Security 
    
   - Peering QoS 
    
   - Peering Accounting and Billing 
    
    
8.1 
   Peering Policy 
    
   Currently, most of the peering policies are local to the domain and 
   statically configured. There MAY be needs for the two trusted peers 
   to exchange peering policies. These need further investigation in the 
   working group. 
    
    
8.2 
   Peering Location Function 
    
   ENUM and DNS provide a way to locate the peering point of a peer 
   domain. Once the request enters the home domain, SM uses [13] to 
   locate the next-hop proxy of the target. There MAY be needs to 
   provide more sophisticated information than what ENUM and DNS provide 
   today. This is future item for the working group. 
    
    
8.3 
   Peering Security 
    
   There are existing security mechanisms today to ensure peer 
   authentication. Most current peering deployments use TLS or other 
 
 
Lee                   Expires December 19, 2006             [Page 13] 

Internet-Draft    Session Peering Use Case for Cable     June 19, 2006 
 
 
   similar mechanism to ensure security channel. This MAY not scale well 
   when an operator tries to peer with few hundred peers. This happens 
   for cable operators provide peering service to large numbers of 
   enterprise customers. Peering security is a working item for the 
   working group. 
    
    
8.4 
   Peering QoS 
    
   Even thought we do not discuss media QoS in the use case, media QoS 
   most impacts the user experience. For some critical services, 
   guaranteed media QoS is a MUST. SIP has defined a framework for pre-
   condition in SIP [14,15]. This framework is for the UA to request 
   end-to-end QoS for media. But, it is unclear how to propagate the 
   session information to the lower network layer when a QoS media 
   session is needed. This requires collaborate effort between working 
   groups to identify the requirements. 
    
    
8.5 
   Peering Accounting and Billing 
    
   In today PSTN peering model, two cable operators compare the outbound 
   minutes for accounting. For Internet peering, they compare the total 
   bandwidth of outbound traffic for accounting. For session peering, it 
   is unclear what is the right model for accounting and billing. 
   Session peering is similar to Internet service, the PSTN peering 
   accounting model MAY not fit very well. Today, most cable operators 
   do not charge users for per minute usage for Internet. Instead, they 
   charge them for bandwidth usage. For the Internet peering accounting 
   model, since signaling and media can possibly travel in two different 
   paths, signaling itself does not necessary convey the accurate 
   bandwidth usage to the cable operators. 
    
    
9. 
  Security Considerations 
    
   Security is a major area for session peering. We MUST prevent 
   unauthenticated peer from making calls to the network and protect the 
   network from DoS attack at session layer. A lot of security work has 
   been done on other working groups to ensure channel security and user 
   authentication. We SHOULD evaluate them and develop some 
   recommendations to the working group. 
    
    
10. 
   IANA Considerations 
    
   This document has no IANA considerations. 
    

 
 
Lee                   Expires December 19, 2006             [Page 14] 

Internet-Draft    Session Peering Use Case for Cable     June 19, 2006 
 
 
11. 
   Acknowledgements 
 
   Special thanks go to Gaurav Khandpur, Tom Creighton, and Jason 
   Livingood for their valuable input to this documents 
    
    
12. 
   References 
    
12.1 
    Normative References 
    
   [1]  Meyer, D., "SPEERMINT Requirements and Terminology ", I-D draft-
   ietf-speermint-reqs-and-terminology-01.txt, February 2006. 
    
   [2]  Mealling, M., "Dynamic Delegation Discovery System (DDDS) Part 
   Three: The Domain Name System (DNS) Database", RFC 3403, October 
   2002.  
    
   [3]  Livingood, J., Pfautz, P. and Stastny, R., "The E.164 to Uniform 
   Resource Identifiers (URI) Dynamic Delegation Discovery System (DDDS) 
   Application for Infrastructure ENUM", I-D draft-ietf-enum-
   infrastructure-00, February 2006. 
    
   [4]  Gulbrandsen, A., Vixie, P. and Esibov, L., "A DNS RR for 
   COecifying the location of services (DNS SRV)", RFC 2782, February 
   2000. 
    
   [5]  Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston, A., 
   Peterson, J., COarks, R., Handley, M., and E. Schooler, "SIP: Session 
   Initiation Protocol", RFC 3261, June 2002. 
    
   [6]  Schulzrinne, H., "The tel URI for Telephone Numbers", RFC 3966, 
   December 2004. 
    
   [7]  Rosenberg, J., Weinberger, J., Huitema, C. and Mahy, R., "STUN - 
   Simple Traversal of User Datagram Protocol (UDP) Through Network 
   Address Translators (NATs)", RFC 3489, March 2003. 
    
   [8]  Rosenberg, J., "Interactive Connectivity Establishment (ICE): A 
   Methodology for Network Address Translator (NAT) Traversal for 
   Offer/Answer Protocols", I-D draft-ietf-mmusic-ice-06, March 2006. 
    
   [9]  Rosenberg, J., Mahy, R. and Huitema, C., "Obtaining Relay 
   Addresses from Simple Traversal of UDP Through NAT (STUN)", I-D 
   draft-ietf-behave-turn-00, February 2006. 
    
   [10] Deering, S. and Hinden, R., "Internet Protocol, Version 6 (IPv6) 
   Specification", RFC 1883, December 1995. 
    

 
 
Lee                   Expires December 19, 2006             [Page 15] 

Internet-Draft    Session Peering Use Case for Cable     June 19, 2006 
 
 
   [11] Draves, R., "Default Address Selection for Internet Protocol 
   version 6 (IPv6)”, RFC 3483, February 2003. 
    
   [12] Gilligan, R., "Transition Mechanisms for IPv6 Hosts and 
   Routers", RFC 2893, August 2000. 
 
   [13] Rosenberg, J. and H. Schulzrinne, "Session Initiation Protocol 
   (SIP): Locating SIP Servers", RFC 3263, June 2002. 
    
   [14] Camarillo, G., Marshall, W. and Rosenberg., J., "Integration of 
   Resource Management and Session Initiation Protocol (SIP)", RFC 3312, 
   October 2002. 
    
   [15] Camarillo, G. and Kyzivat, P., "Update to the Session Initiation 
   Protocol (SIP) Preconditions Framework", RFC 4032, March 2005. 
    
   [16] Rekhter, Y., Moskowitz, B., Karrenberg, D., de Groot, G. J. 
   and Lear E., "Address Allocation for Private Internets", RFC 1918, 
   February 1996. 
    
   [17] Tsirtsis, G. and Srisuresh, P., "Network Address Translaton – 
   Protocol Translation (NAT-PT)", RFC 2766, February 2000. 
    
    
12.2 
    Informative References 
    
   [18] 3GPP TS 23.228 V7.3.0, "IP Multimedia Subsystem (IMS); Stage 2 
   (Release 7)", March, 2006. 
    
   [19] CableLabs, "PacketCable 1.5 Architecture Framework Technical 
   Report" PKT-TR-ARCH1.5-V01-050128, January, 2005. 
    
   [20] Dierks, T. and Allen, C., "The TLS Protocol Version 1.0", RFC 
   2246, January 1999. 
    
   [21]  Kent, S. and Seo, K. "Security Architecture for the Internet 
   Protocol", RFC 4301, December 2005. 
    
    
Authors’ Addresses 
    
   Yiu L. Lee 
   Comcast Cable Communications 
   1500 Market Street, 
   Philadelphia, PA 19102 
   US 
    
   Phone: +1-215-320-5894 
   Email: yiu_lee@cable.comcast.com 
 
 
Lee                   Expires December 19, 2006             [Page 16] 

Internet-Draft    Session Peering Use Case for Cable     June 19, 2006 
 
 
Intellectual Property and Copyright Statements  
    
   Intellectual Property Statement  
    
   The IETF takes no position regarding the validity or scope of any 
   Intellectual Property Rights or other rights that might be claimed to 
   pertain to the implementation or use of the technology described in 
   this document or the extent to which any license under such rights 
   might or might not be available; nor does it represent that it has 
   made any independent effort to identify any such rights.  Information 
   on the procedures with respect to rights in RFC documents can be 
   found in BCP 78 and BCP 79.  
    
   Copies of IPR disclosures made to the IETF Secretariat and any 
   assurances of licenses to be made available, or the result of an 
   attempt made to obtain a general license or permission for the use of 
   such proprietary rights by implementers or users of this 
   specification can be obtained from the IETF on-line IPR repository at 
   http://www.ietf.org/ipr.  
    
   The IETF invites any interested party to bring to its attention any 
   copyrights, patents or patent applications, or other proprietary 
   rights that may cover technology that may be required to implement 
   this standard.  Please address the information to the IETF at ietf-
   ipr@ietf.org. 
     
   Disclaimer of Validity  
    
   This document and the information contained herein are provided on an 
   "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS 
   OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET 
   ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, 
   INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE 
   INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED 
   WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. 
    
    
   Copyright Statement  
    
   Copyright (C) The Internet Society (2006). This document is subject 
   to the rights, licenses and restrictions contained in BCP 78, and 
   except as set forth therein, the authors retain all their rights.  
    
    
   Acknowledgment  
    
   Funding for the RFC Editor function is currently provided by the 
   Internet Society.  

 
 
Lee                   Expires December 19, 2006             [Page 17] 


PAFTECH AB 2003-20262026-04-23 16:43:48