One document matched: draft-yang-sipping-voipli-00.txt




SIPPING Working Group                                      Menghui. Yang
Internet-Draft                                              Xiaodong. Li
Intended status: Experimental                                   Wei. Mao
Expires: April 28, 2009                                            CNNIC
                                                            Oct 25, 2008


  Architecture and Practice for VoIP Lawful Interception Using Session
                         Border Controller(SBC)
                    draft-yang-sipping-voipli-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 April 28, 2009.

Abstract

   Recently broadband Voice over IP (VoIP) has a clear trend to
   substitute for traditional telephone services such as wire telephone,
   wireless telephone and mobile telephone service, it has become
   increasingly clear that VoIP services will be expected to provide
   Lawful Intercept to the same level experienced in the Public Switched
   Telephone Network(PSTN)[7].  In an effort to provide lawful
   interception for broadband VoIP, an architecture using session border
   controller was proposed, and a prototype implementation of the
   broadband VoIP lawful interception was created and basic performance
   evaluation was conducted on this prototype system.  This document



Yang, et al.             Expires April 28, 2009                 [Page 1]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


   reports on the prototype implementation and the test results.


Table of Contents

   1.  Introduction . . . . . . . . . . . . . . . . . . . . . . . . .  4
   2.  Terminology  . . . . . . . . . . . . . . . . . . . . . . . . .  5
   3.  A Prototype of Broadband VoIP LI Implementation  . . . . . . .  6
     3.1.  Interfaces . . . . . . . . . . . . . . . . . . . . . . . .  7
       3.1.1.  Internal network interface INI-1 . . . . . . . . . . .  8
       3.1.2.  Internal network interface INI-2 . . . . . . . . . . .  9
       3.1.3.  Internal network interface INI-3 . . . . . . . . . . .  9
       3.1.4.  Internal network interface HI-2  . . . . . . . . . . .  9
       3.1.5.  Internal network interface HI-3  . . . . . . . . . . .  9
       3.1.6.  Internal network interface HI-1  . . . . . . . . . . . 10
     3.2.  Components . . . . . . . . . . . . . . . . . . . . . . . . 10
       3.2.1.  Session border controller  . . . . . . . . . . . . . . 10
       3.2.2.  Delivery Functions . . . . . . . . . . . . . . . . . . 11
       3.2.3.  Collection functions in LEMF . . . . . . . . . . . . . 12
       3.2.4.  Administration Function  . . . . . . . . . . . . . . . 12
   4.  Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
     4.1.  Test environment . . . . . . . . . . . . . . . . . . . . . 13
     4.2.  Test Scenarios and Results . . . . . . . . . . . . . . . . 14
       4.2.1.  SBC Intercepted Maximum Concurrent Session Number
               without media  . . . . . . . . . . . . . . . . . . . . 14
       4.2.2.  SBC Intercepted Maximum Concurrent Session Number
               with media . . . . . . . . . . . . . . . . . . . . . . 16
       4.2.3.  Delivery Function 2 CII Maximum Transactions
               Throughput . . . . . . . . . . . . . . . . . . . . . . 18
       4.2.4.  Delivery Function 2 CII Minimum, Average and
               Maximum Transactions Response Time . . . . . . . . . . 19
       4.2.5.  Delivery Function 3 CC Maximum Transactions
               Throughput . . . . . . . . . . . . . . . . . . . . . . 20
       4.2.6.  Delivery Function 3 Minimum, Average and Maximum
               Transactions Response Time . . . . . . . . . . . . . . 21
       4.2.7.  Collection Function CII Maximum Transactions
               Throughput . . . . . . . . . . . . . . . . . . . . . . 22
       4.2.8.  Collection Function CII Minimum, Average and
               Maximum Transactions Response Time . . . . . . . . . . 23
       4.2.9.  Collection Function CC Maximum Transactions
               Throughput . . . . . . . . . . . . . . . . . . . . . . 24
       4.2.10. Collection Function CC Minimum, Average and
               Maximum Transactions Response Time . . . . . . . . . . 25
   5.  Limitations and Issues . . . . . . . . . . . . . . . . . . . . 26
     5.1.  General Issues . . . . . . . . . . . . . . . . . . . . . . 26
     5.2.  Security Issues  . . . . . . . . . . . . . . . . . . . . . 27
     5.3.  Protocol Details . . . . . . . . . . . . . . . . . . . . . 28
   6.  Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . 28



Yang, et al.             Expires April 28, 2009                 [Page 2]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


   7.  Security considerations  . . . . . . . . . . . . . . . . . . . 29
   8.  Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 29
   9.  References . . . . . . . . . . . . . . . . . . . . . . . . . . 29
     9.1.  Normative References . . . . . . . . . . . . . . . . . . . 29
     9.2.  Informative References . . . . . . . . . . . . . . . . . . 29
   Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 30
   Intellectual Property and Copyright Statements . . . . . . . . . . 31












































Yang, et al.             Expires April 28, 2009                 [Page 3]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


1.  Introduction

   Lawful Interception (LI) is a requirement placed upon service
   providers to provide legally sanctioned official access to private
   communications.  With the existing PSTN, lawful intercept is
   performed by applying a physical 'tap' on the telephone line of the
   target in response to a warrant from a Law Enforcement Agency
   (LEA)[6].  However, VoIP technology has enabled the mobility of the
   end-user, so it is no longer possible to guarantee the interception
   of calls based on tapping a physical line.  So, interception of VoIP
   based on traditional lawful interception regulatory framework, which
   focused on PTSN-based telephone service, is not reasonable.  LI of
   PSTN-based telephone service can be done by wiretapping of the
   switch, with assistance of telecommunications service providers with
   LEA.  But, for LI of VoIP, interception capability should be
   developed and installed or separate interception equipment should be
   deployed.

   Current broadband VoIP services show that there are two types of
   providers: Access providers and Service providers.  Access providers
   are those that actually provide broadband connectivity to the
   premises (cable, DSL, etc.).  Service providers are those providing
   the VoIP services.  In some cases, the Access provider can also be a
   Service provider (e.g., a cable operator that provides VoIP telephony
   services).  The latest large-scale additions to the VoIP offering are
   peer-to-peer connections.  These services bypass the traditional
   centralized call-control mechanisms and work directly with proxies
   and end points.

   In a Broadband access environment, the access provider doesn!_t know
   what services the subscriber is utilizing because access providers
   only supply a pipe from the subscribers premises to the Internet.  In
   this case, the service provider doesn!_t provide any switching
   equipment and the subscriber can work with any Service provider they
   choose via the pipe to the Internet.  The subscriber can also decide
   to utilize peer-to-peer services over this same broadband pipe.  In
   this environment, the access provider is unable to identify or
   determine certain events or information due to the service provider
   not owning or controlling the switching and routing equipment.

   One of the primary problems that service providers face when managing
   VoIP and multimedia calls is the separation of the signaling and
   media streams.  In other words it is quite possible that the two
   streams may take completely different paths through the network.  In
   addition, even when they do pass through the same device, it may not
   be aware of the relationship between the streams.  Some devices
   within the network are however specifically designed to understand
   and manage the separate signaling and media streams - session border



Yang, et al.             Expires April 28, 2009                 [Page 4]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


   controllers (SBC)[8].

   SBC is a session-aware device that manages VoIP calls at the borders
   of an IP network.  Unlike most network devices, SBC are aware of the
   relationship between the two parts of a VoIP call: signaling and
   media.  Session Border Controllers handle both media and signaling,
   intercept can be performed in a completely undetectable manner.

   SBCs usually sit between two service provider networks in a peering
   environment, or between an access network and a backbone network to
   provide service to residential and/or enterprise customers.  They
   typically are deployed at the border between two VoIP networks, and
   they offer an ideal location to implement a Lawful Intercept
   solution.

   Whilst the detailed requirements for VoIP Lawful Interception may
   differ from one jurisdiction to another, the general requirements are
   the same.  The LI system must provide transparent interception of
   specified traffic only and the subject must not be aware of the
   intercept.  The service provided to other users must not be affected
   during interception.

   As part of the effort in developing a broadband VoIP lawful
   interception architecture, we implemented a prototype and conducted
   evaluation experiments on the prototype system.  In this document, we
   first describe the prototype solutions and then report experimental
   results.  We hope that this document can provide useful information
   to those interested in the subject.

   The prototype implementation and experimental results presented in
   this report serve only as an input, and by no means preempt any
   solution development that may be carried out by future IETF effort.
   Indeed, the presented solutions are experimental approaches and have
   a number of limitations as discussed in Section 5 and 6.


2.  Terminology

   Lawful Intercept (LI) "C The targeted intercept of VoIP services, by
   a service provider on the behalf of Law Enforcement, when authorized
   by a court.

   Session Border Controller (SBC) - provides a variety of functions to
   enable or enhance session-based multi-media services (e.g., Voice
   over IP), provides access to an interception subject!_s.

   Call-Identifying Information (CII) - refers to information that
   identifies the origin, direction, destination, or termination of each



Yang, et al.             Expires April 28, 2009                 [Page 5]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


   call generated or received by a subject.

   Call Content (CC) - audio and video media flow.

   Collection Function (CF) "C A law enforcement facility designated as
   the transmission destination for CC and CII.

   Delivery Function (DF) "C The mechanism responsible for delivering
   intercepted communications form a VoIP network operator/service
   provider to a CF via HI.

   Administration Function - provides provisioning interface for the
   intercept as a result of a court order or warrant delivered by the
   Law Enforcement Agency (LEA)

   Internal Network Interface (INI) "C The VoIP network!_s internal
   interface between SBC and DF.  Consists of INI1,INI2, and INI3
   matching HI1,HI2, and HI3 interfaces.

   Handover Interface (HI) "C Interface for delivering CC and CII from
   DF to CF.


3.  A Prototype of Broadband VoIP LI Implementation

   A new VoIP Lawful intercept architecture based on SBC is proposed, as
   shown in Figure 1.
























Yang, et al.             Expires April 28, 2009                 [Page 6]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


                        +--------------------+     HI-1   +------------+
              INI-1(1)  |                    |            |            |
          ______________|    Adminstration   |------------|            |
         |              |      Function      |            |            |
         |              |                    |            |            |
         |              +--------------------+            |            |
         V                | INI-1(3)    | INI-1 (2)       |            |
   +------------+         |             V                 |            |
   |            |         |          +------------+       |    Law     |
   | Session    |  INI-2  |          |            |  HI-2 | Enforcement|
   | Border     |---------+----------|  Delivery  |-------| Monitoring |
   | Controller |         |          | Function 2 |       |  Facility  |
   |            |         |          |            |       |            |
   |            |         |          +------------+       |            |
   +------------+         |                               |            |
         |                V                               |            |
         |              +--------------------+            |            |
         |              |                    |   HI-3     |            |
         |    INI-3     |                    |------------|            |
         +--------------| Delivery Function 3|            |            |
                        |                    |            |            |
                        +--------------------+            +------------+

   Figure 1 VoIP Lawful Intercept Architecture based on SBC

   This methodology utilizes a TCP/IP interface to the SBC to both
   provision it and receive call data events (originate, ring back, hook
   lash, etc.) from it.  Those interfaces are described as INI-1 for
   provisioning and INI-2 for call data.  A third interface INI-3
   receives a copy of the media stream.  All of this information is sent
   to the Delivery functions from SBC.  The Delivery functions then
   sorts, filters, and formats the information for delivering to Law
   Enforcement.  The interfaces used to deliver the call data events and
   media content to Law Enforcement are TCP/IP interfaces called HI-2
   and HI-3, respectively.  Lawful Interception deals with two products,
   these are: contents of call (CC) and call-identifying information
   (CII).  CC is exactly what it sounds like: the voice, video or
   message contents.  CII refers to information that identifies the
   origin, direction, destination, or termination of each call generated
   or received by a subject.  In the following sections, we describe the
   specific mechanism implemented in detail.

3.1.  Interfaces

   This section provides a brief description of the interfaces in the
   architecture (Figure 1).  One of the objectives in defining these
   interfaces is to keep both the internal interfaces (INI-1, INI-2, and
   INI-3) and the handover interfaces (HI-1,HI-2, and HI-3) the same



Yang, et al.             Expires April 28, 2009                 [Page 7]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


   regardless of country-specific requirements.

3.1.1.  Internal network interface INI-1

   This is an intercept request provisioning interface, through which
   the Administration function provisions warrants on SBC, and activate,
   deactivate or interrogate the Delivery functions.

   Administration function uses this interface to provide the provision
   for the intercept as a result of a court order or warrant delivered
   by the LEMF.  It involves separate provisioning interfaces for SBC
   and Delivery functions.  Administration function takes care of
   provisioning of SBC and Delivery functions with some policy.

   In order to provide a generic interface for intercepting,
   replicating, encapsulating and transporting call information packets
   to Delivery function, the intercept request interface should specify:

   1 the identification of the subject to be intercepted, such as a SIP
   URI;

   2 the destination address and port of the Delivery function or
   Collection function (where to send the intercepted packets);

   3 encapsulation and transport parameters;

   4 start time, end time, and duration.

   INI-1 involves separate provisioning interfaces,INI-1(1),INI-1(2),and
   INI-1(3).  INI-1(1) is an interface between the SBC and
   Administration function for delivering provisioning information to
   point out the intercepted subject identify, when to start and end,
   which delivery function the intercepted information should be sent
   to, etc.  INI-1(2) and INI-1(3) interfaces between the Delivery
   functions and Administration function are used to config the Delivery
   function which SBC will send intercepted information to it, and which
   Collection function the intercepted information should be sent to,
   etc.  Because of the requirement in some laws to limit accessibility
   to authorized personnel, the provisioning interface has to be
   strictly controlled.  In many cases, the identity of the subject
   received from the LEMF has to be translated into an identity that can
   be used by the network to enable the intercept.

   In this prototype implementation, INI-1 is implemented with Abstract
   Syntax Notation One(ASN.1)[1] and Basic Encoding Rule(BER)[2] encoded
   to be binary compatible.





Yang, et al.             Expires April 28, 2009                 [Page 8]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


3.1.2.  Internal network interface INI-2

   This is the VoIP call data (e.g., SIP signaling) intercepting
   interface, over which SBC sends call-identifying information (e.g.,
   SIP signaling) to the DF server.  The intercepted call data is
   encapsulated using DIAMETER protocol [3].  This encapsulation method
   retains all of the information in the original packets(source and
   destination addresses as well as payload) and provides an identifier
   for correlating the CC with the CII.

   Note, however, this interface implemented in prototype is based on
   UDP which is an unreliable and unordered transport protocol (i.e.,
   provides neither retransmission on detection of errors nor ordering
   of data).  The underlying networks should be engineered to meet the
   overall reliability requirements for delivery of content.

3.1.3.  Internal network interface INI-3

   This is the call content (e.g., RTP media stream) intercepting
   interface, over which SBC sends the RTP media stream to the DF
   server.  The transferred information in this interface is
   encapsulated into Diameter event messages[3].

   Since the media path is independent of the signaling path, the media
   may not traverse through the operator!_s network unless the SBC
   modifies the session description exchanged between the user-agents.
   By modifying the session description the SBC can force the media to
   be sent through SBC.  The SBC behaves as a Back-to-Back User
   Agent(B2BUA) and inserts itself in the media path, and modifies the
   session descriptions carried in the SIP messages.  As a result, the
   SBC receives media from one user-agent and relays it to other user-
   agent and performs the identical operation with media traveling in
   the reverse direction.

3.1.4.  Internal network interface HI-2

   The Handover Interface, HI-2, is the interface between DF 2 and LEMF
   for delivering CII.  This interface transmits the result of
   intercepted SIP signaling to the LEMF, and is implemented with
   Abstract Syntax Notation One(ASN.1) and Basic Encoding Rule(BER)
   encoded to be binary compatible.

3.1.5.  Internal network interface HI-3

   The Handover Interface, HI-3, is the interface between DF 3 and LEMF
   for delivering CC.  This interface transmits the result of
   intercepted RTP media stream to the LEMF, and is implemented with
   Abstract Syntax Notation One(ASN.1) and Basic Encoding Rule(BER)



Yang, et al.             Expires April 28, 2009                 [Page 9]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


   encoded to be binary compatible.

3.1.6.  Internal network interface HI-1

   This is an interception administration interface, through which the
   LEMF provides intercept administration information to the service
   provider administration function.

   In this prototype, this interface is under consideration, and is not
   available now.

3.2.  Components

   This section provides a brief description of the key components in
   the architecture (Figure 1).

3.2.1.  Session border controller

   Session border controller handles all SIP messages which are needed
   for interworking between different operators who exchange traffic
   with each other.  It takes care of replicating signaling and delivers
   it to specific delivery function using appropriate format.  SBC
   inserts itself in the media path, and modifies the session
   descriptions carried in the SIP messages to deliver it to specific
   delivery function using appropriate format.

   Example: Consider the following example scenario: the SBC receives an
   INVITE request from the outer network, which in this case is an
   access network.  The received SIP message containing the SDP session
   descriptor is shown as following.

   v=0 o=- 5 2 IN IP4 218.241.111.22 s=CounterPath X-Lite 3.0 c=IN IP4
   218.241.111.22 t=0 0 m=audio 12706 RTP/AVP 107 119 100 106 0 105 98 8
   3 101 a=alt:1 1 : IF5eH1Wn YCfz8oLV 218.241.111.22 12706 a=fmtp:101
   0-15 a=rtpmap:107 BV32/16000 a=rtpmap:119 BV32-FEC/16000 a=rtpmap:100
   SPEEX/16000 a=rtpmap:106 SPEEX-FEC/16000 a=rtpmap:105 SPEEX-FEC/8000
   a=rtpmap:98 iLBC/8000 a=rtpmap:101 telephone-event/8000 a=sendrecv

   In this example, the SBC performs the media intercept function by
   rewriting the !(R)c!_ and !(R)m!_ lines in above SDP.  The following
   shows the session description after the intercept function.

   v=0 o=- 5 2 IN IP4 218.241.111.22 s=CounterPath X-Lite 3.0 c=IN IP4
   218.241.108.108 t=0 0 m=audio 30006 RTP/AVP 107 119 100 106 0 105 98
   8 3 101 a=alt:1 1 : IF5eH1Wn YCfz8oLV 218.241.111.22 12706 a=fmtp:101
   0-15 a=rtpmap:107 BV32/16000 a=rtpmap:119 BV32-FEC/16000 a=rtpmap:100
   SPEEX/16000 a=rtpmap:106 SPEEX-FEC/16000 a=rtpmap:105 SPEEX-FEC/8000
   a=rtpmap:98 iLBC/8000 a=rtpmap:101 telephone-event/8000 a=sendrecv



Yang, et al.             Expires April 28, 2009                [Page 10]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


   SBCs can terminate media streams and SIP dialogs by generating BYE
   requests in a situation where the LI needs.

   Note, if an SBC directs many media streams through SBC itself, it is
   likely to cause a significant amount of additional traffic to a path
   to that SBC.  This might create possible bottleneck in the path.

   In this prototype, we used an Open Source software, OpenSBC, written
   by Joegen Baclor, Solegy LCC.  To enhance the OpenSBC to replicate
   RTP packet, we added media intercept function by using PCAPLIB.  We
   also added some code in OpenSBC to encapsulate SIP and RTP using
   Diameter AVP.

3.2.2.  Delivery Functions

   Delivery function implemented the INI-2, INI-3, HI-2 and HI-3
   interfaces specification.

   Delivery function performs the delivery functions.  It converts the
   information from the INI-2 and INI-3 interfaces to the corresponding
   information format on the HI-2 and HI-3 interfaces respectively.

   It receives the data from the SBC, packets them into the correct
   format and delivers them to the correct LEMF.  In the case where
   multiple law enforcement agencies are intercepting the same subject,
   the delivery function may replicate the information many times, and
   forward them to multiple LEMF interfaces simultaneously.

   Signaling messages and RTP media streams are using different delivery
   functions.  Delivery function 2 is used for signaling messages
   processing and delivering, and Delivery function 3 is used for RTP
   media streams processing and delivering.

   Any information stored within the SBC and Delivery functions are
   stored in volatile memory, so that these information are erased if a
   component of the network node is removed or powered down.  Only the
   encrypted database of the ADMF should be maintained during power-down
   situations.  In the event of a link failure between the DF and the
   LEMF the intercept products may be buffered for a short time in
   memory only.  Any long term failure of the interface will result in
   intercept products being lost - this information must not be spooled
   to permanent storage.

   In this prototype, we implemented Delivery functions using C++.







Yang, et al.             Expires April 28, 2009                [Page 11]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


3.2.3.  Collection functions in LEMF

   Collection function in LEMF implemented the HI-2 and HI-3 interfaces
   specification, and can receive intercepted VoIP call signaling and
   RTP media stream on these two interfaces.

   It records and stores the intercepted traffic, and provides analysis
   tools to track, correlate and interpret intercepted traffic.

   Also, collection function can recover the original VoIP call based on
   the received SIP signaling and RTP media.

   In this prototype, we implemented collection function using C++.

3.2.4.  Administration Function

   Administration function provides the provisioning interface for the
   intercept as a result of a court order or warrant delivered by the
   LEMF.  This function implemented INI-1 and HI interfaces
   specification.  It is used for receiving warrants from HI-1
   interface, and delivering provisioning information to SBC and
   Delivery functions to point out the intercepted subject identify,
   when to start and end, which delivery function the intercepted
   information should be sent to, etc.

   It supports user interface, maintains all warrant information from
   LEMF using encrypted database, creates shared memory image of
   intercept information.

   In this prototype, this component is under consideration, and is not
   available now.


4.  Test

   The objective of test is to evaluate SBC behaviors while it performs
   LI, and the intercepting capability of developed LI interfaces and
   functions.

   The test cases covered in this document provide performance
   evaluation metrics for this prototype, including:

   1.concurrent number of users that completed their run successfully;
   2.number of transactions that passed, failed, stopped, or ended with
   errors; 3.average response time taken to perform transactions during
   each second; 4.total of number of completed transactions(both
   successful and unsuccessful) performed during each second; 5.minimum,
   average, and maximum response time for all the transactions in the



Yang, et al.             Expires April 28, 2009                [Page 12]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


   load test.

4.1.  Test environment

   Test environment is shown in Figure 2.  SBC is using the OpenSBC
   which is enhanced with RTP media intercept and supports INI
   interfaces; The calling and called parties are using SIPP to simulate
   concurrent calls.


                               +-------+ INI-2 +-------+ HI-2  +-------+
                               | SBC   |-------|   DF  |-------|  CF   |
                               |       |.......|       |.......|       |
                             .'+-------+`.INI-3+-------+ HI-3  +-------+
                           .'     /\      `.
                         .'      /  \       `.
                      RTP     SIP    \SIP     RTP
                     .'        /      \         `.
                   .'         /        \          `.
                 .'   +-------+       +-------+     `.
          +------+    | SIP   |       | SIP   |    +------+
          |      |----| Proxy |       | Proxy |----|      |
          +------+    +-------+       +-------+    +------+
    Target Subscriber                             Target Subscriber


   Figure 2 Test Topology for VoIP Lawful Interception Using SBC

   VoIP call flows is shown in Figure 3.






















Yang, et al.             Expires April 28, 2009                [Page 13]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


    A               B              SBC               DF               CF
    |               |               |                |                |
    |               |               |                |                |
    | Request:Invite|               |                |                |
    |---------------+-------------->|INI-2 CII:Invite|                |
    |     Status:100|Trying         |--------------->|HI-2 CII:Invite |
    |---------------+---------------|                |--------------->|
    |               | Request:Invite|                |                |
    |               |---------------|                |                |
    |               Status:180Ringing                |                |
    |               |-------------->| INI-2 CII:200OK|                |
    |Status:180Ringing              |----------------| HI-2 CII:200 OK|
    |---------------+---------------|                |--------------->|
    |               |Status:200 OK  |                |                |
    |               |-------------->|                |                |
    |Status:200 OK  |               |                |                |
    |---------------+---------------| INI-2 CII:ACK  |                |
    |               |  Request:ACK  |--------------->+  HI-3 CII:ACK  |
    |---------------+-------------->|                |--------------->|
    |               |               |                |                |
    |               | Request:ACK   |                |                |
    |               |---------------|                |                |
    |     RTP       |               |                |                |
    ################################# INI-3 CC:RTP   |                |
    |               |      RTP      |--------------->|  HI-3 CC:RTP   |
    |               |################                |--------------->+
    | Request:BYE   |               |                |                |
    |---------------+-------------->| INI-2 CII:BYE  |                |
    |               |  Request:BYE  |--------------->|  HI-2 CII:BYE  |
    |               |---------------|                |--------------->+
    |               |Status:200 OK  |                |                |
    |               |-------------->|                |                |
    | Status:200 OK |               |                |                |
    |---------------+---------------|                |                |
    |               |               |                |                |


   Figure 3 Call flow of VoIP in testing

4.2.  Test Scenarios and Results

4.2.1.  SBC Intercepted Maximum Concurrent Session Number without media

   VoIP call flows is shown in Figure 3, but there are no RTP meida
   stream sent.

   Procedure:




Yang, et al.             Expires April 28, 2009                [Page 14]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


   1.  Configure SIP UDP with an attempted session rate =10 sessions per
   second, session duration=20ms, and media streams per session =0.

   2.  Start to initiate SIP Session establishment with the SBC.

   3.  Measure failed session attempts and total session established at
   the SBC.

   4.  If no failed session attempt is recorded then increase the
   attempted session rate configured on the Tester by 50%.

   5.  If a failed session attempt is recorded then reduce the attempted
   session rate configured on the Tester by 50%.

   6.  Repeat steps 2 through 5 until the maximum concurrent session
   establishment rate is obtained.

   Test result is shown in Figure 4.  From result in our test, the
   maximum concurrent session number is about 80.  The average response
   time for SIP messages is within 30ms when LI was not carried out in
   SBC.  The average response time for SIP messages is within 80ms when
   LI was carried out in SBC.





























Yang, et al.             Expires April 28, 2009                [Page 15]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


                           Maximum Concurrent SIP Sessions
               ________________________________________________________
              |                                                        |
           160|                                                        |
              |                                                        |
              |                                                        |
           140|                                         /|             |
              |                                        ,'`             |
              |                                      ,'   \            |
   Average 120|                                   ,-'      |           |
   Respinse   |                                  /         \.        | |
   Time for   |                                 /           \       |` |
   SIP     100|                                 |            L.-|   J  |
   Messages   |                            ,.__/                \  /   +
              |                          _/                  _/  b'    |
            80|       _..______,,... .--'                 +-' |        |
              |`-----'             '`'                   |    |        |
              |                                          |     |       |
            60|                                          /     `b    ,.|
              |                                         /        | _/ `|
              |                                        /         `o'   |
            40|                                       /i               |
              |                        |-_   _.  _...-'                |
              |                       ,' '`''  `'                      |
            20|..______/\         _/-'                                 |
              |_________iiiiiiiiii_____________________________________|
                  10   20   30   40   50   60   70   80   90   100   110
                          Number of Concurrent SIP Sessions


   Figure 4 Maximum Concurrent Session Number

4.2.2.  SBC Intercepted Maximum Concurrent Session Number with media

   VoIP call flows is shown in Figure 3.

   Procedure:

   1.  Configure SIP UDP with an attempted session rate =10 sessions per
   second, session duration=20 sec.

   2.  Start to initiate SIP Session establishment with the SBC and
   transmit media through the SBC by modifying the SDP.

   3.  Measure failed session attempts and total session established,
   and Packet Loss of the media.

   4.  If no failed session attempt or packet loss is recorded then



Yang, et al.             Expires April 28, 2009                [Page 16]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


   increase the attempted session rate configured on the Tester by 50%.

   5.  If a failed session attempt or packet loss is recorded then
   reduce the attempted session rate configured on the Tester by 50%.

   6.  Repeat steps 2 through 5 until the maximum concurrent session
   number is obtained.

   Test result is shown in Figure 5 and Figure 6.  From Figure, we can
   see the maximum concurrent session number is 80.

   The average response time for SIP messages is within 30ms when LI was
   not carried out in SBC.  The average response time for SIP messages
   is within 90ms when LI was carried out in SBC.


            200 _______________________________________________________
               |                                                       |
               |                                                       |
            180|                                                       |
               |                                                       |
               |                                                |.     |
            160|                                                |\     |
               |                                               ,' \    |
               |                                               /   \ ,\|
   Average  140|                                              /     - ||
   Response    |                                             ,'       \|
   Time for    |                                             '         |
   SIP      120|                                            /          |
   Messages    |                                           |'          |
               |                                           /           |
            100|                                   ,Y'''`.|  o         |
               |                                 ,/       '  ||        |
               |_/`.b       ...._,,'''`-.......,'            ||        |
             80|/    `-----'                                / \        |
               |                                           /   .       |
               |                                          /    \.      |
             60|                                         ,'     |      |
               |                                        /       `.+   ,|
               |                                     .-`         `o  ,'|
             40|                       ,+..     __,.-'             `/  |
               |       .._            /    `...'                    '  |
               |      /   -.         /                                 |
             20+---L,'      \       ,'                                 |
               |_____________\..===i___________________________________|
                   10   20   30   40   50   60   70   80   90   100  110
                          Number of Concurrent SIP Sessions




Yang, et al.             Expires April 28, 2009                [Page 17]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


   Figure 5 Maximum Concurrent Session Number

   In our SBC implementation, SIP signaling and RTP media are processed
   in different modules.  In our tester, SIP signaling flows are same in
   above two test cases.  From Figure, we can see the RTP media
   processing reached the peak value at the 80 Concurrent SIP sessions.
   The throughput of RTP media streams reaches about 600 kilo bytes per
   seconds.


                   ____________________________________________________
               600|                                      __............|
                  |                                    ,'              |
                  |                                 ,''                |
                  |                               ,'                   |
               500|                              /                     |
                  |                            ,'                      |
                  |                           /                        |
                  |                          |                         |
   Throuthput  400|                          |                         |
   of RTP         |                         /                          |
   media streams  |                        ,'                          |
   (kilo          |                       /                            |
    Byte       300|                      /                             |
    per           |                    ,'                              +
    second)       |                  _/                                |
                  |                 ,'                                 |
                  |              _Y'                                   |
               200|           _/'                                      |
                  |         ,'                                         |
                  |       ,'                                           |
                  |     ,'                                             |
               100|    /                                               |
                  |   /                                                |
                  | ,'                                                 |
                  +/                                                   |
                  |i___________________________________________________|
                    0   20   30   40   50   60   70   80   90   100  110
                           Number of Concurrent SIP Sessions



   Figure 6 Maximum Concurrent RTP Media Streams

4.2.3.  Delivery Function 2 CII Maximum Transactions Throughput

   We use the test tool Load Running to evaluate the Delivery function
   performance.  We create 200 concurrent transactions, and it is the



Yang, et al.             Expires April 28, 2009                [Page 18]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


   maximum that can be obtained in this tool.

   We can obtain the result of maximum concurrent SIP signaling
   transactions throughput of Delivery Function 2 is 110 transactions,
   and average concurrent transactions per second is 32.  Figure 7 shows
   the total number of successful CII transactions performed during each
   second of a load test.



                               Totol Transaction per Second
                 +-----------------------------------------------------+
              120|                                                     |
                 |                                           |         |
                 |                                          ||         |
              100|                                          |`.        |
                 |                                         ,' |        |
   Transactions  |                                         /  |        |
   Number      80|                                         |  |        |
                 |                                         |  |        |
                 |                                         |  |        |
               60|                                         |  |        |
                 |                                        |   |        |
                 |                                        |   |        |
               40|                _                       |   |        |
                 |   _          _/'". _.,`.               |   |        |
                 | ,' \b_     ,/     "     \.     /''''o |     `.      |
               20|-      `'"./               `.__-'     `|       `ob___|
                 |                                                     |
                 |                                                     |
                 +-----------------------------------------------------+
                 00:00          00:05           00:10            00:15
                              Elapsed scenario time mm:ss

   Figure 7 Total Passed CII Transactions per Second

   This graph helps to determine the actual transaction load on VoIP
   Lawful Intercept system at any given moment.

4.2.4.  Delivery Function 2 CII Minimum, Average and Maximum
        Transactions Response Time

   Figure 8 shows the minimum, average and maximum CII transactions
   response time for SIP signaling transactions.  The minimum response
   time is 0.002 seconds, the maximum response time is 0.269 second, and
   the average response time is 0.018 second.





Yang, et al.             Expires April 28, 2009                [Page 19]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


                         Average Transaction Response Time
            0.3+-------------------------------------------------------+
               |                                                       |
               |                                            +|         |
               |                                            ||         |
               |                                            |\         |
               |                                            | |        |
               |                                            / |        |
            0.2|                                           |  |        |
               |                                           |  \        |
    Average    |                                           |   |       |
    Response   |                                           |   |       |
    Time       |                                           |   \       |
    (seconds)  |                                           /    |      |
               |                                          |     |      |
               |                                          |     |      |
               |                                          |     \      |
            0.1|                                          |      |     |
               |                                          /      |     |
               |                                         |       \     |
               |                                         |        |    |
           0.05|                                         |        |    |
               |                                         |        |    |
               | ......................................../        \....|
               |_______________________________________________________|
                00:00          00:05          00:10            00:15
                             Elapsed scenario time mm:ss


   Figure 8 CII Transaction Response Time

4.2.5.  Delivery Function 3 CC Maximum Transactions Throughput

   Figure 9 shows the total number of successful CC transactions
   performed during each second.  We can obtain the result of maximum
   concurrent RTP transactions throughput of Delivery Function 3 is 25
   transactions, and average concurrent RTP transactions per second is
   7.













Yang, et al.             Expires April 28, 2009                [Page 20]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


                              Totol Transaction per Second
               24`+----------------------------------------------------+
                 | \                                                   |
                 |  `.                                                 |
               21|    \*                                               |
                 |     |                                               |
                 |     |                                               |
               18|     \                                               |
                 |      |                                              |
                 |      |                                              |
   Transaction 15|      |                                              |
   Number        |      \                                              |
                 |       |                                             |
               12|       |                                             |
                 |       \                                             |
                 |        |                                            |
               9+|        |               /|                           |
                 |        \              / \                           |
                 |         |            /   \                          |
               6+|         |           |     |                         |
                 |         |           /     \                         |
                 |         \          /       |                        |
               3+|          |     ,'*/        \           /\           |
                 |          |    /             \        ,'  \          |
                 |          \  ,'               |      /     \         |
                0|___________*/_________________\____,i_______\*______+
               00:00       00:02             00:05           00:08
                             Elapsed scenario time mm:ss

   Figure 9 Total Passed CC Transactions per Second

4.2.6.  Delivery Function 3 Minimum, Average and Maximum Transactions
        Response Time

   Figure 10 show the minimum, average and maximum CC response time for
   RTP transactions.  The minimum response time is 0.02 seconds, the
   maximum response time is 9.05 second, and the average response time
   is 3.46 second.













Yang, et al.             Expires April 28, 2009                [Page 21]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


                           Average Transaction Response Time
            9+---------------------------------------------------------'
             |                                                         |
            8|                                                         |
             |                                                         |
            7|                                                      .-'|
             |                                                   .-'   |
   Average  6|                                              .-:-'      |
   Response  |                                           .-'           |
   Time     5|                                        .-'              |
   (seconds) |                                      .'                 |
            4|                                   .-'                   |
             |                                .-'                      |
            3|                   .'+------+.-'                         |
             |                .-'                                      |
            2|              .'                                         |
             |           .-'                                           |
            1|         .'                                              |
             +------.-                                                 |
             +------+-------+------+------+------+------+------+------+
           00.00  00.01 00.02 00.03 00.04  00.05 00.06  00.07  00.08
                             Elapsed scenario time mm:ss

   Figure 10 CC Transaction Response Time

4.2.7.  Collection Function CII Maximum Transactions Throughput

   From Figure 11, we can obtain the result of maximum concurrent SIP
   signaling transactions throughput of collection function in LEMF is
   105 transactions, and average concurrent transactions per second is
   38.




















Yang, et al.             Expires April 28, 2009                [Page 22]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


                              Totol Transaction Response Time
                   ____________________________________________________
                  |                                                    |
                  |                                        \           |
               100|                                        |\          |
                  |                                        | `.        |
                90|                                        /   \   .|  |
                  |                       /|              |     \.' |  |
                80|                      / \              |         \  |
                  |                     |   |             /          | |
                70|                     /   |            |           | |
   Transactions   |                    /    \            |           \ |
   Numberf      60|                   /      |           |            ||
                  |                   +      \           /            ||
                50|                   /       |         |             ||
                  |                  /        \         |             \|
                40|                 |          |        /              |
                  |                 /          |        /              |
                30|                /           \       |               |
                  |               |             |      /               |
                20|    /`.        /             \     /                |
                  |   /   `-...../              +---+/                 |
                10|  /                                                 |
                  | /                                                  |
                  |/___________________________________________________|
                  00:01               00:05                00:10
                               Elapsed scenario time mm:ss


   Figure 11 Total Passed CII Transactions per Second

4.2.8.  Collection Function CII Minimum, Average and Maximum
        Transactions Response Time

   Figure 12 shows the CF minimum, average and maximum CII transaction
   response time for SIP signaling transactions.  The minimum response
   time is 0.003 seconds, the maximum response time is 3.029 second, and
   the average response time is 0.669 second.













Yang, et al.             Expires April 28, 2009                [Page 23]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


                         Average Transaction Response Time
           3 +----------------------------------------------------------
             |                                                        .|
             |                                                       / |
             |                                                      /  |
             |                                                     |   |
           2 |                                                     /   |
             |                                                    /    |
   Average   |                                                   /     |
   Response  |                                                  /      |
   Time      |                                X                |       |
   (seconds) |                               / \               /       |
           1 +                             .'   \             /        |
             +                            /      |           /         |
             |                           /       \          |          |
             |                          .'        \         /          |
             |                         /           \_     ./           |
           0 |,,,,,,,,,,,,,,,,,,,,,,,,;           ,++-,,,;+            |
             +---+---+---+---+---+---+---+----+----+---+----+---+----+-+
            00:00              00:05                00:10          00:14
                             Elapsed scenario time mm:ss


   Figure 12 CII Transaction Response Time

4.2.9.  Collection Function CC Maximum Transactions Throughput

   Figure 13 shows the total number of successful CC transactions in CF
   performed during each second of a load test.  We can obtain the
   result of maximum concurrent RTP transactions throughput of CF is 63
   transactions, and average concurrent RTP transactions per second is
   13.



















Yang, et al.             Expires April 28, 2009                [Page 24]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


                                  Totol Transaction per Second
                   ____________________________________________________
                  |                          +                         |
                60|                          +                         |
                  |                         |\                         |
                  |                         / |                        |
                50|                        |  |                        |
                  |                        /  \                        |
                  |                       |    |                       |
   Transactions 40+                       /    |                       |
   Number         \                      |     \                       |
                  ||                     /      |                      |
                30|\                    |       |                      |
                  | \                   /       |                      |
                  |  \                 |        \                      |
                20+   |                /         |                     |
                  |   \               |          |                     |
                  |    \              /          \                     |
                10+     |            |            |          .'.       |
                  |     \            /            |        .'   `-.    |
                  |      \`-.      .-'            \       /        `.._|
                  |__________i=..=i_______________+......i_____________|
                  00:00 00:01 00:02 00:03  00:04 00:05 00:06 00:07 00:08
                                   Elapsed scenario time mm:ss

   Figure 13 Total Passed CC Transactions per Second

4.2.10.  Collection Function CC Minimum, Average and Maximum
         Transactions Response Time

   Figure 14 shows the minimum, average and maximum CC response time for
   RTP transactions in CF.  The minimum response time is 0.013 seconds,
   the maximum response time is 7.083 second, and the average response
   time is 3.285 second.

















Yang, et al.             Expires April 28, 2009                [Page 25]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


                             Average Transaction Response Time
              _________________________________________________________
            7|                                                         |
             |                                                       ,"|
             |                                                     ,"  |
            6|                                                   ,"    |
             |                                                 ,"      |
             |                                               ,"        |
            5|                                            ,-"          |
             |                                          ,"             |
   Average   |                                        ,"               |
   Response 4|                                      ,"                 |
   Time      |                                    ,"                   |
   (seconds) |                                  ,"                     |
            3|                      /.........,"                       |
             |                    ,"                                   |
             |                   /                                     |
            2|                 ,"                                      |
             |                /                                        |
             |               /                                         |
            1|             ,"                                          |
             |            /                                            |
             |__________,                                              |
             |_________________________________________________________|
           00:00   00:01   00:02   00:03   00:04   00:05   00:06   00:07
                                  Elapsed scenario time mm:ss


   Figure 14 CC Transaction Response Time


5.  Limitations and Issues

   There are several issues both within this overall VoIP LI problem
   area and with the particular approach taken in the prototype.

5.1.  General Issues

   It seems not to be difficult to decide whether technically mature and
   widely provided services such as PSTN-based voice services, cellular
   services and 3G services are to be intercepted or not.  However, the
   situation for recently developed services such as internet telephony
   service, internet access services and messenger is difficult.  The
   necessary of lawful interception for these services is increasing
   because of the increase of the users, but LI of the services can be
   controversial because the boundary of services is not clear,
   government!_s policy for IP services is not fully established, and so
   on[4].



Yang, et al.             Expires April 28, 2009                [Page 26]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


   For the internet telephone service, the followings can be considered
   to decide whether to intercept the service or not:

   1.  Whether it substitutes for traditional telephone services
   considering the number of the service users or communication time or
   sales;

   2.  Telecommunication service management regulations, and so on.

   Because VoIP has a clear trend to substitute for traditional
   telephone services such as wire telephone, wireless telephone and
   mobile telephone service, it has become increasingly clear that VoIP
   services will be expected to provide Lawful Intercept to the same
   level experienced in the PSTN.  The FCC [9] in North America for
   example has mandated that both emergency calls and Lawful Intercept
   must be available for VoIP.  Whilst not all countries mandate this
   capability, any network operator building a publicly available voice
   or multimedia over IP service today will need to plan a network which
   is flexible enough to implement these regulatory services in the
   future.

5.2.  Security Issues

   The lawful intercept solution should have the ability to provide
   stringent security measures to combat threats such as impersonation
   of delivery function!_s, privacy and confidentiality breaches, as
   well as message forgery and replay attacks.

   Illegal interception may be done using facilities, equipment,
   functions, human resources and procedures prepared for lawful
   interception.  Therefore, it is very important to prepare technical
   and managerial countermeasures to prevent illegal interception,
   unauthorized access to call identifying information, call content, or
   to interception log data.

   In general, all interfaces could have the capability of providing
   strong cryptographic authentication to establish the identity of the
   principals, and be able to correlate the identity of the principle
   with the action they are attempting to perform.  All interfaces could
   be capable of performing some sort of cryptographic message integrity
   checking such as, for example, HMAC-MD5, if needed[5].

   While this document doesn!_t discuss issues of physical security,
   operating system, or application hardening within the principals of
   the lawful intercept solution, they are clearly important.






Yang, et al.             Expires April 28, 2009                [Page 27]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


5.3.  Protocol Details

   In the current VoIP lawful interception prototype, we defined the
   HI-2 and HI-3 interface specification with Abstract Syntax Notation
   One(ASN.1) and Basic Encoding Rule(BER).  And also provide methods to
   encapsulate SIP and RTP using Diameter AVP in INI-2 and INI-3
   interfaces specification.

   The broadband VoIP lawful intercept solution in this document is
   merely one option among many.  Solutions appear to depend highly on
   SBC.  In any case, the prototyped solution has a number of
   limitations.


6.  Conclusion

   Several conclusions can be drawn from the practice.

   First, the practice is a proof that a prototype can be built that is
   deployable on border between two VoIP networks to carry out both
   interworking and intercepting for VoIP services.  The practice also
   proved a proof of concept for the SBC-based lawful intercept for VoIP
   service.

   SBC is deployed at the border between two VoIP networks to execute a
   number of access, router, switch, security and quality management
   roles, and they offer an ideal location to implement a Lawful
   Intercept solution.  Carriers don!_t need to learn the LI functions
   of multiple devices, reduces costs for training, maintenance, etc.

   Nevertheless, as discussed in the previous section, there are a
   number of limitations, issues, and questions in the prototype
   designs.

   Future work in this area should attempt to answer some of the issues
   raised in section 5.  Some of the key issues going forward include:

   1.  Scalability for SBC and Delivery functions.

   2.  Interfaces specification design issues, such as INI-1 and HI
   interfaces, etc.

   3.  Security considerations, all interfaces could be capable of
   performing some sort of cryptographic message integrity checking such
   as, for example, HMAC-MD5, if needed.






Yang, et al.             Expires April 28, 2009                [Page 28]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


7.  Security considerations

   The purpose of the document is to report broadband VoIP LI
   architecture prototype implementation and experimental results.  Some
   security considerations of the solution mechanisms of the prototype
   are mentioned in section 5, but are not the main problem to be
   described in this document.


8.  Acknowledgements

   The authors thank Yuanmin Chen, Bin Guo, Wei Zhang, and Hui Chen for
   their assistance in developing this prototype system.  The authors
   thank their contribution to this document.


9.  References

9.1.  Normative References

   [1]  CCITT, "Specification of Abstract Syntax Notation One (ASN.1)",
        Recommendation x. 208, November 1988.

   [2]  CCITT, "Specification of Basic Encoding Rules for Abstract
        Syntax Notation One (ASN.1)", Recommendation x. 209, November
        1988.

   [3]  Calhoun, P., Loughney, J., Guttman, E., Zorn, G., and J. Arkko,
        "Diameter Base Protocol", RFC 3588, Setpember 2003.

   [4]  Bradner, S., "IETF Policy on Wiretapping", RFC 2804, May 2000.

   [5]  Baker, F., Foster, B., and C. Sharp, "Cisco Architecture for
        Lawful Intercept in IP Networks", RFC 3924, October 2004.

   [6]  ANS,  ., "Lawfully Authorized Electronic Surveillance(LAES) for
        Voice over Packet Technologies in Wireline Telecommunications
        Networks", T1 678,  2004.

   [7]  ETSI, "Telecommunications security: Lawful Interception (LI):
        Requirements for network functions", ETSI 201 158, April 2004.

9.2.  Informative References

   [8]  Hautakorpi, J., Camarillo, G., Penfield, R., Hawrylyshen, A.,
        and M. Bhatia, "Requirements form SIP (Session Initiation
        Protocol) Session Border Control Deployments", June 2008.




Yang, et al.             Expires April 28, 2009                [Page 29]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


   [9]  Federal Communication Commission, "Second Report and Order and
        Memorandum Option and order, Washinton,D.C", May 2006.


Authors' Addresses

   Menghui Yang
   CNNIC
   4 South 4th Street,Zhongguancun,Haidian District
   Beijing, Beijing  100190
   China

   Phone: +86 10 5881 3204
   Email: yangmenghui@cnnic.cn


   Xiaodong Li
   CNNIC
   4 South 4th Street,Zhongguancun,Haidian District
   Beijing, Beijing  100190
   China

   Phone: +86 10 5881 3020
   Email: lee@cnnic.cn


   Wei Mao
   CNNIC
   4 South 4th Street,Zhongguancun,Haidian District
   Beijing, Beijing  100190
   China

   Phone: +86 10 5881 2230
   Email: mao@cnnic.cn

















Yang, et al.             Expires April 28, 2009                [Page 30]

Internet-Draft    Architecture and Practice for VoIP LI         Oct 2008


Full Copyright Statement

   Copyright (C) The IETF Trust (2008).

   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.

   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, THE IETF TRUST 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.


Intellectual Property

   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.











Yang, et al.             Expires April 28, 2009                [Page 31]



PAFTECH AB 2003-20262026-04-24 01:33:20