MOBOPTS Research Group A. Dutta Internet-Draft Telcordia Expires: August 14, 2005 Y. Ohba (Ed.) K. Taniuchi TARI H. Schulzrinne Columbia Univ. February 13, 2005 A Framework of Media-Independent Pre-Authentication (MPA) draft-ohba-mobopts-mpa-framework-00 Status of this Memo This document is an Internet-Draft and is subject to all provisions of section 3 of RFC 3667. 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 become aware will be disclosed, in accordance with RFC 3668. 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 August 14, 2005. Copyright Notice Copyright (C) The Internet Society (2005). Abstract This document describes a framework of Media-independent Pre-Authentication (MPA), a new handover optimization mechanism that has a potential to address issues on existing mobility management Dutta, et al. Expires August 14, 2005 [Page 1] Internet-Draft MPA Framework February 2005 protocols and mobility optimization mechanisms. MPA is a mobile-assisted, secure handover optimization scheme that works over any link-layer and with any mobility management protocol. This document also shows an initial implementation of MPA in our testbed and some performance results to show how existing protocols could be leveraged to realize the functionalities of MPA. Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.1 Performance Requirements . . . . . . . . . . . . . . . . . 5 2. Existing Work Fast-handover . . . . . . . . . . . . . . . . . 7 3. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 9 4. MPA Framework . . . . . . . . . . . . . . . . . . . . . . . . 11 4.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . 11 4.2 Functional Elements . . . . . . . . . . . . . . . . . . . 12 4.3 Basic Communication Flow . . . . . . . . . . . . . . . . . 12 5. Detailed Issues . . . . . . . . . . . . . . . . . . . . . . . 17 5.1 Discovery . . . . . . . . . . . . . . . . . . . . . . . . 17 5.2 Proactive IP address acquisition . . . . . . . . . . . . . 18 5.2.1 PANA-assisted proactive IP address acquisition . . . . 19 5.2.2 IKEv2-assisted proactive IP address acquisition . . . 19 5.2.3 Proactive IP address acquisition using DHCP only . . . 19 5.3 Address resolution issues . . . . . . . . . . . . . . . . 20 5.3.1 Proactive duplicate address detection . . . . . . . . 20 5.3.2 Proactive address resolution update . . . . . . . . . 21 5.4 Tunnel management . . . . . . . . . . . . . . . . . . . . 22 5.5 Binding Update . . . . . . . . . . . . . . . . . . . . . . 23 5.6 Preventing packet loss . . . . . . . . . . . . . . . . . . 23 5.7 Link-layer security and mobility . . . . . . . . . . . . . 24 5.8 Authentication in initial network attachment . . . . . . . 25 6. Initial Implementation and Results . . . . . . . . . . . . . . 26 6.1 Network structure . . . . . . . . . . . . . . . . . . . . 26 6.2 MPA Scenario . . . . . . . . . . . . . . . . . . . . . . . 27 6.3 Non-MPA Scenario . . . . . . . . . . . . . . . . . . . . . 29 6.4 The evaluation and the results . . . . . . . . . . . . . . 31 6.5 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . 32 7. Security Considerations . . . . . . . . . . . . . . . . . . . 33 8. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 34 9. References . . . . . . . . . . . . . . . . . . . . . . . . . . 35 9.1 Normative References . . . . . . . . . . . . . . . . . . . . 35 9.2 Informative References . . . . . . . . . . . . . . . . . . . 36 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . 38 Intellectual Property and Copyright Statements . . . . . . . . 39 Dutta, et al. Expires August 14, 2005 [Page 2] Internet-Draft MPA Framework February 2005 1. Introduction As wireless technologies including cellular and wireless LAN are popularly used, supporting terminal handovers across different types of access networks, such as from a wireless LAN to CDMA or to GPRS is considered as a clear challenge. On the other hand, supporting terminal handovers between access networks of the same type is still more challenging, especially when the handovers are across IP subnets or administrative domains. To address those challenges, it is important to provide terminal mobility that is agnostic to link-layer technologies in an optimized and secure fashion without incurring unreasonable complexity. In this document we discuss terminal mobility that provides seamless handovers with low-latency and low-loss. Seamless handovers are characterized in terms of performance requirements as described in Section 1.1. The basic part of terminal mobility is accomplished by a mobility management protocol that maintains a binding between a locator and an identifier of a mobile terminal, where the binding is referred to as the mobility binding. The locator of the mobile node may dynamically change when there is a movement of the mobile terminal. The movement that causes a change of the locator may occur not only physically but also logically. A mobility management protocol may be defined at any layer. In the rest of this document, the term "mobility management protocol" refers to a mobility management protocol which operates at network layer or higher. There are several mobility management protocols at different layers. Mobile IP [RFC3344] and Mobile IPv6 [RFC3775] are mobility management protocols that operate at network-layer. There are several ongoing activities in the IETF to define mobility management protocols at layers higher than network layer. For example, MOBIKE (IKEv2 Mobility and Multihoming) [I-D.ietf-mobike-design] is an extension to IKEv2 that provides the ability to deal with a change of an IP address of an IKEv2 end-point. HIP (the Host Identity Protocol) [I-D.ietf-hip-base] defines a new protocol layer between network layer and transport layer to provide terminal mobility in a way that is transparent to both network layer and transport layer. Also, SIP-Mobility is an extension to SIP to maintain the mobility binding of a SIP user agent [SIPMM]. While mobility management protocols maintain mobility bindings, using them solely in their current form is not sufficient to provide seamless handovers. An additional optimization mechanism that works in the visited network of the mobile terminal to prevent loss of outstanding packets transmitted while updating the mobility binding is needed to achieve seamless handovers. Such a mechanism is referred to as a mobility optimization mechanism. For example, Dutta, et al. Expires August 14, 2005 [Page 3] Internet-Draft MPA Framework February 2005 mobility optimization mechanisms [I-D.ietf-mobileip-lowlatency-handoffs-v4] and [I-D.ietf-mipshop-fast-mipv6] are defined for Mobile IPv4 and Mobile IPv6, respectively, by allowing neighboring access routers to communicate to carry information on mobile terminals. There are protocols that are considered as "helpers" of mobility optimization mechanisms. The CARD (Candidate Access Router Discovery Mechanism) protocol [I-D.ietf-seamoby-card-protocol] is designed to discover neighboring access routers. The CTP (Context Transfer Protocol) [I-D.ietf-seamoby-ctp] is designed to carry state that is associated with the services provided for the mobile terminal, or context, among access routers. There are several issues in existing mobility optimization mechanisms. First, existing mobility optimization mechanisms are tightly coupled with specific mobility management protocols. For example, it is not possible to use mobility optimization mechanisms designed for Mobile IPv4 or Mobile IPv6 for MOBIKE. What is strongly desired is a single, unified mobility optimization mechanism that works with any mobility management protocol. Second, there is no existing mobility optimization mechanism that easily supports handovers across administrative domains without assuming a pre-established security association between administrative domains. A mobility optimization mechanism should work across administrative domains in a secure manner only based on a trust relationship between a mobile node and each administrative domain. Third, a mobility optimization mechanism needs to support not only multi-interface terminals where multiple simultaneous connectivity through multiple interfaces can be expected, but also single-interface terminals. This document describes a framework of Media-independent Pre-Authentication (MPA), a new handover optimization mechanism that has a potential to address all those issues. MPA is a mobile-assisted, secure handover optimization scheme that works over any link-layer and with any mobility management protocol including Mobile IPv4, Mobile IPv6, MOBIKE, HIP, SIP mobility, etc. In MPA, the notion of IEEE 802.11i pre-authentication is extended to work at higher layer, with additional mechanisms to perform early acquisition of IP address from a network where the mobile terminal may move as well as proactive handover to the network while the mobile terminal is still attached to the current network. Since this document focuses on the MPA framework, it is left to the future work to choose actual set of protocols for MPA and define detailed operations. Nevertheless, the document also shows an initial implementation of MPA in our testbed and some performance results to show how existing protocols could be leveraged to realize the functionalities of MPA. Dutta, et al. Expires August 14, 2005 [Page 4] Internet-Draft MPA Framework February 2005 1.1 Performance Requirements In order to provide desirable quality of service for interactive VoIP and streaming traffic, one needs to limit the value of end-to-end delay, jitter and packet loss to a certain threshold level. ITU-T and ITU-E standards define the acceptable values for these parameters. For example for one-way delay, ITU-T G.114 recommends 150 ms as the upper limit for most of the applications, and 400 ms as generally unacceptable delay. One way delay tolerance for video conferencing is in the range of 200 to 300 ms. Also if an out-of-order packet is received after a certain threshold it is considered lost. References [RFC2679], [RFC2680] and 2681 [RFC2681] describe some of the measurement techniques for delay and jitter. Also if an out-of-order packet is received after a certain threshold it is considered lost. An end-to-end delay consists of several components, such as network delay, OS delay, CODEC delay and application delay. Network delay consists of transmission delay, propagation delay, queueing delay in the intermediate routers. Operating System related delay consists of scheduling behavior of the operating system in the sender and receiver. CODEC delay is generally caused due to packetization and depacketization at the sender and receiver end. Application delay is mainly attributed to playout delay that helps compensate the delay variation within a network. End-to-end delay and jitter values can be adjusted using proper value of the playout buffer at the receiver end. In case of interactive VoIP traffic end-to-end delay affects the jitter value and is an important thing to consider. During a mobile's frequent handover, transient traffic cannot reach the mobile and this contributes to the jitter as well. If the end system has a playout buffer, then this jitter is subsumed by the playout buffer delay, but otherwise this adds to the delay for interactive traffic. Packet loss is typically caused by congestion, routing instability, link failure, lossy links such as wireless links. During a mobile's handover a mobile is subjected to packet loss because of its change in attachment to the network. Thus for both streaming traffic and VoIP interactive traffic packet loss will contribute to the service quality of the real-time application. Number of packets lost is proportional to the delay during handover and rate of traffic the mobile is receiving. Lost packets contribute to congestion in case of TCP traffic because of re-transmission, but it does not add to any congestion in case of streaming traffic that is RTP/UDP based. Thus it is essential to reduce the packet loss and effect of handover delay in any mobility management scheme. In Section 2, we describe some of the fast-handover scheme that have tried to reduce the handover. According to ETSI TR 101 [ETSI] a normal voice conversation can Dutta, et al. Expires August 14, 2005 [Page 5] Internet-Draft MPA Framework February 2005 tolerate up to 2% packet loss. If a mobile is subjected to frequent handoff during a conversation, each handoff will contribute to packet loss for the period of handoff. Thus maximum loss during a conversation needs to be reduced to a level that is acceptable. There is no clear threshold value for packet loss for streaming application, but it needs to be reduced as much as possible to provide better quality of service to a specific application. Dutta, et al. Expires August 14, 2005 [Page 6] Internet-Draft MPA Framework February 2005 2. Existing Work Fast-handover While basic mobility management protocols such as Mobile IP [RFC3344], Mobile IPv6 [RFC3775], SIP-Mobility [SIPMM] offer solution to provide continuity to TCP and RTP traffic, these are not optimized to reduce the handover latency during mobile's frequent movement between subnets and domains. In general these mobility management protocols suffer from handover delays incurred at several layers such as layer 2, layer 3 and application layer for updating the mobile's mobility binding. There have been several optimization techniques that apply to currently mobility management schemes that try to reduce handover delay and packet loss during a mobile's movement between cells, subnet and domain. There are few micro-mobility management schemes [CELLIP], [HAWAII], and intra-domain mobility management schemes such as [IDMP], [I-D.ietf-mobileip-reg-tunnel] that provide fast-handover by limiting the signaling updates within a domain. Fast Mobile IP protocols for IPv4 and IPv6 networks [I-D.ietf-mobileip-lowlatency-handoffs-v4], [I-D.ietf-mipshop-fast-mipv6] provide fast-handover techniques that utilize mobility information made available by the link layer triggers. Yokota et al. [YOKOTA] proposes joint use of access point and dedicated MAC bridge to provide fast-handover without altering MIPv4 specification. [MACD] scheme reduces the delay due to MAC layer handoff by providing a cache-based algorithm. Some of the mobility management schemes use dual interfaces thus providing make-before-break scenario [SUM]. In a make-before-break situation communication usually continues with one interface, when the secondary interface is in the process of getting connected. The IEEE 802.21 working group is discussing these scenarios in details. Providing fast-handover using a single interface needs more careful design techniques than for a client with multiple interfaces. [SIPFAST] provides an optimized handover scheme for SIP-based mobility management, where the transient traffic is forwarded from the old subnet to the new one by using an application layer forwarding scheme. [MITH] provides a fast handover scheme for a single interface case that uses mobile initiated tunneling between the old foreign agent and new foreign agent. [MITH] defines two types of handover schemes such as Pre-MIT and Post-MIT. Our MPA scheme is very similar in nature to MITH's predictive scheme where the mobile communicates with the foreign agent before actually moving to the new network. However the proposed MPA scheme described in this document is not limited to MIP type mobility protocol only and in addition this scheme takes care of movement between domains and performs pre-authentication in addition to proactive handover. Thus the proposed scheme reduces the overall delay to close to link-layer Dutta, et al. Expires August 14, 2005 [Page 7] Internet-Draft MPA Framework February 2005 handover delay. Dutta, et al. Expires August 14, 2005 [Page 8] Internet-Draft MPA Framework February 2005 3. Terminology Mobility Binding: A binding between a locator and an identifier of a mobile terminal. Mobility Management Protocol (MMP): A protocol that operates at network layer or higher to maintain a binding between a locator and an identifier of a mobile terminal. Binding Update: A procedure to update a mobility binding. Media-independent Pre-Authentication Mobile Node (MN): A mobile terminal of media-independent pre-authentication (MPA) which is a mobile-assisted, secure handover optimization scheme that works over any link-layer and with any mobility management protocol. An MPA mobile node is an IP node. In this document, the term "mobile node" or "MN" without a modifier refers to "MPA mobile node". An MPA mobile node usually has a functionality of a mobile node of a mobility management protocol as well. Candidate Target Network (CTN): A network to which the mobile may move in the near future. Target Network (TN): The network to which the mobile has decided to move. The target network is selected from one or more candidate target network. Proactive Handover Tunnel (PHT): A bidirectional IP tunnel that is established between the MPA mobile node and an access router of the candidate target network. In this document, the term "tunnel" without a modifier refers to "proactive handover tunnel". Point of Attachment (PoA): A link-layer device (e.g., a switch, an access point or a base station, etc.) that functions as a link-layer attachment point for the MPA mobile node to a network. Dutta, et al. Expires August 14, 2005 [Page 9] Internet-Draft MPA Framework February 2005 Care-of Address (CoA): An IP address used by a mobility management protocol as a locator of the MPA mobile node. Dutta, et al. Expires August 14, 2005 [Page 10] Internet-Draft MPA Framework February 2005 4. MPA Framework 4.1 Overview Media-independent Pre-Authentication (MPA) is a mobile-assisted, secure handover optimization scheme that works over any link-layer and with any mobility management protocol. With MPA, a mobile node is not only able to securely obtain an IP address and other configuration parameters from a candidate target network, but also able to send and receive IP packets using the obtained IP address and other configuration parameters, before it attaches to the candidate target network when the candidate target network becomes the target network. This makes it possible for the mobile node to complete the binding update of any mobility management protocol and use the new care-of address before performing a handover at link-layer. This functionality is provided by allowing a mobile node, which has a connectivity to the current network but is not yet attached to a candidate target network, to (i) establish a security association with the candidate target network to secure the subsequent protocol executions, then (ii) securely execute a configuration protocol to obtain an IP address and other configuration parameters from the candidate target network as well as a tunnel management protocol to establish a bidirectional tunnel between the mobile node and an access router of the candidate target network, then (iii) send and receive IP packets, including signaling messages for binding update of a mobility management protocol and data packets transmitted after completion of binding update, over the tunnel using the obtained IP address as the tunnel inner address, and finally (iv) deleting or disabling the tunnel immediately before attaching to the candidate target network when it becomes the target network and then re-assigning the inner address of the deleted or disabled tunnel to its physical interface immediately after the mobile node is attached to the target network through the interface. Instead of deleting or disabling the tunnel before attaching to the the target network, the tunnel may be deleted or disabled immediately after being attached to the target network. Especially, the third procedure makes it possible for the mobile to complete higher-layer handover before starting link-layer handover. This means that the mobile is able to send and receive data packets transmitted after completion of binding update over the tunnel, while it is still able to send and receive data packets transmitted before completion of binding update outside the tunnel. In the above four basic procedures of MPA, the first procedure is referred to as "pre-authentication", the second procedure is referred to as "pre-configuration", the combination of the third and fourth Dutta, et al. Expires August 14, 2005 [Page 11] Internet-Draft MPA Framework February 2005 procedures are referred to as "secure proactive handover". The security association established through pre-authentication is referred to as an "MPA-SA". The tunnel established through pre-configuration is referred to as a "proactive handover tunnel". 4.2 Functional Elements In the MPA framework, the following functional elements are expected to reside in each candidate target network to communicate with a mobile node: Authentication Agent (AA), Configuration Agent (CA) and Access Router (AR). Some or all of those elements can be placed in a single network device or in separate network devices. An authentication agent is responsible for pre-authentication. An authentication protocol is executed between the mobile node and the authentication agent to establish an MPA-SA. The authentication protocol MUST be able to derive a key between the mobile node and the authentication agent and SHOULD be able to provide mutual authentication. The authentication protocol SHOULD be able to interact with a AAA protocol such as RADIUS and Diameter to carry authentication credentials to an appropriate authentication server in the AAA infrastructure. The derived key is used for further deriving keys used for protecting message exchanges used for pre-configuration and secure proactive handover. Other keys that are used for bootstrapping link-layer and/or network-layer ciphers MAY also be derived from the MPA-SA. A protocol that can carry EAP [RFC3748] would be suitable as an authentication protocol for MPA. A configuration agent is responsible for one part of pre-configuration, namely securely executing a configuration protocol to securely deliver an IP address and other configuration parameters to the mobile node. The signaling messages of the configuration protocol MUST be protected using a key derived from the key corresponding to the MPA-SA. An access router is a router that is responsible for the other part of pre-configuration, i.e., securely executing a tunnel management protocol to establish a proactive handover tunnel to the mobile node, and secure proactive handover using the proactive handover tunnel. The signaling messages of the configuration protocol MUST be protected using a key derived from the key corresponding to the MPA-SA. IP packets transmitted over the proactive handover tunnel SHOULD be protected using a key derived from the key corresponding to the MPA-SA. 4.3 Basic Communication Flow Assume that the mobile node is already connected to a point of Dutta, et al. Expires August 14, 2005 [Page 12] Internet-Draft MPA Framework February 2005 attachment, say oPoA (old point of attachment), and assigned a care-of address, say oCoA (old care-of address). The communication flow of MPA is described as follows. Throughout the communication flow, data packet loss should not occur except for the period during the switching procedure in Step 5, and it is the responsibility of link-layer handover to minimize packet loss during this period. Step 1 (pre-authentication phase): The mobile node finds a candidate target network through some discovery process and obtains the IP addresses, an authentication agent, a configuration agent and an access router in the candidate target network by some means. The mobile node performs pre-authentication with the authentication agent. If the pre-authentication is successful, an MPA-SA is created between the mobile node and the authentication agent. Two keys are derived from the MPA-SA, namely an MN-CA key and an MN-AR key, which are used to protect subsequent signaling messages of a configuration protocol and a tunnel management protocol, respectively. The MN-CA key and the MN-AR key are then securely delivered to the configuration agent and the access router, respectively. Step 2 (pre-configuration phase): The mobile node realizes that its point of attachment is likely to change from oPoA to a new one, say nPoA (new point of attachment). It then performs pre-configuration, with the configuration agent using the configuration protocol to obtain an IP address, say nCoA (new care-of address), and other configuration parameters from the candidate target network, and with the access router using the tunnel management protocol to establish a proactive handover tunnel. In the tunnel management protocol, the mobile node registers oCoA and nCoA as the tunnel outer address and the tunnel inner address, respectively. The signaling messages of the pre-configuration protocol are protected using the MN-CA key and the MN-AR key. When the configuration and the access router are co-located in the same device, the two protocols may be integrated into a single protocol like IKEv2. After completion of the tunnel establishment, the mobile node is able to communicate using both oCoA and nCoA by the end of Step 4. Step 3 (secure proactive handover main phase): The mobile node determines to switch to the new point of attachment by some means. Before the mobile node switches to the new point of attachment, it starts secure proactive handover by executing binding update of a mobility management protocol and transmitting subsequent data traffic over the tunnel (main phase). Step 4 (secure proactive handover pre-switching phase): The mobile node completes binding update and becomes ready to switch to the new point of attachment point. The mobile may execute the tunnel management protocol to delete or disable the proactive handover Dutta, et al. Expires August 14, 2005 [Page 13] Internet-Draft MPA Framework February 2005 tunnel and cache nCoA after deletion or disabling of the tunnel. The decision as to when the mobile node is ready to switch to the new point of attachment depends on handover policy. Step 5 (switching): It is expected that a link-layer handover occurs in this step. Step 6 (secure proactive handover post-switching phase): The mobile node executes the switching procedure. Upon successful completion of the switching procedure, the mobile node immediately restores the cached nCoA and assigns it to the physical interface attached to the new point of attachment. If the proactive handover tunnel was not deleted or disabled in Step 4, the tunnel is deleted or disabled as well. After this, direct transmission of data packets using nCoA is possible without using a proactive handover tunnel. Dutta, et al. Expires August 14, 2005 [Page 14] Internet-Draft MPA Framework February 2005 +-----------------------------------+ | Candidate Target Network | | (Future Target Network) | IP address(es) MN oPoA | nPoA AA CA AR | Available for | | | | | | | | Use by MN | | +-----------------------------------+ | | | | | | . +---------------+ | | | | | . |(1) Found a CTN| | | | | | . +---------------+ | | | | | | | Pre-authentication | | | | | [authentication protocol] | | | |<--------+------------->|MN-CA key| | | | | | |-------->|MN-AR key| | +--------------------+ | | |------------------>| | |(2) Increased chance| | | | | | [oCoA] |to switch to the CTN| | | | | | | +--------------------+ | | | | | | | | | | | | | | Pre-configuration | | | | | [configuration protocol to get nCoA] | | |<--------+----------------------->| | | | Pre-configuration | | | | | [tunnel management protocol to establish PHT] V |<--------+--------------------------------->| | | | | | | ^ +-----------------+ | | | | | | |(3) Determined to| | | | | | | |switch to the CTN| | | | | | | +-----------------+ | | | | | | | | | | | | | | Secure proactive handover main phase | | | [execution of binding update of MMP and | | | transmission of data packets through AR | [oCoA, nCoA] | based on nCoA over the PHT] | | | |<<=======+=================================>+--->... | . . . . . . . . . . . . . . . . . . . . . Figure 1: Basic Communication Flow (1/2) Dutta, et al. Expires August 14, 2005 [Page 15] Internet-Draft MPA Framework February 2005 | | | | | | | +-----------------+ | | | | | | |(4) Completion | | | | | | | |of MMP BU and | | | | | | | |ready to switch | | | | | | | +-----------------+ | | | | | | | Secure proactive handover pre-switching phase | | [tunnel management protocol to delete PHT] V |<--------+--------------------------------->| +---------------+ | | | | |(5)Switching | | | | | +---------------+ | | | | | | | | | +---------------+ | | | | |(6) Completion | | | | | |of switching | | | | | +---------------+ | | | | o<- Secure proactive handover post-switching phase ^ | [Re-assignment of TIA to the physical I/F] | | | | | | | | Transmission of data packets through AR | [nCoA] | based on nCoA| | | | | |<---------------+---------------------------+-->... | | | | | | . Figure 2: Basic Communication Flow (2/2) Dutta, et al. Expires August 14, 2005 [Page 16] Internet-Draft MPA Framework February 2005 5. Detailed Issues In order to provide an optimized handover for a mobile experiencing rapid subnet and domain handover, one needs to look into several issues. These issues include discovery of neighboring networking elements, choosing the right network to connect to based on certain policy, changing the layer 2 point of attachment, obtaining an IP address from a DHCP or PPP server, confirming the uniqueness of the IP address, pre-authenticating with the authentication agent such as AAA server in a specific domain, sending the binding update to the correspondent host and obtaining the redirected streaming traffic to the new point of attachment. We describe these issues in details in the following paragraphs and describe how we have optimized these in case of MPA-based secure proactive handover. 5.1 Discovery Discovery of neighboring networking elements such as access points, access routers, authentication servers help expedite the handover process during a mobile's rapid movement between networks. By discovering the network neighborhood with a desired set of coordinates, capabilities and parameters the mobile can perform many of the operation such as pre-authentication, proactive IP address acquisition, proactive address resolution, and binding update while in the previous network. There are several ways a mobile can discover the neighboring networks. The Candidate Access Router Discovery protocol [I-D.ietf-seamoby-card-protocol] helps discover the candidate access routers in the neighboring networks. Given a certain network domain SLP and DNS help provide address of the networking components for a given set of services in the specific domain. In some cases many of the network layer and upper layer parameters may be sent over link-layer management frames such as beacons when the mobile approaches the vicinity of the neighboring networks. IEEE 802.11u is considering issues such as discovering neighborhood using information contained in link-layer. However, if the link-layer management frames are encrypted by some link-layer security mechanism, then the mobile node may not able to obtain the requisite information before establishing link-layer connectivity to the access point. In addition this may add burden to the bandwidth constrained wireless medium. In such cases a higher layer protocol is preferred to obtain the information regarding the neighboring elements. There is some proposal such as [NETDISC] that helps obtain these information about the neighboring networks from a mobility server. When the mobile's movement is imminent, it starts the discovery process by querying a specific server and obtains the required parameters such as the IP address of the access point, its characteristics, routers, SIP Dutta, et al. Expires August 14, 2005 [Page 17] Internet-Draft MPA Framework February 2005 servers or authentication servers of the neighboring networks. In the event of multiple networks, it may obtain the required parameters from more than one neighboring networks and keep these in cache. At some point the mobile finds out several candidate target networks out of many probable networks and starts the pre-authentication process by communicating with the required entities in the candidate target networks. 5.2 Proactive IP address acquisition In general a mobility management protocol works in conjunction with Foreign Agent or in co-located address mode. Our MPA approach can use both co-located address mode and foreign agent address mode. We discuss here the address assignment component that is used in co-located address mode. There are several ways a mobile node can obtain an IP address and configure itself. Most commonly a mobile can configure itself statically in the absence of any configuring element such as a server or router in the network. The IETF Zeroconf working group defines auto-IP mechanism where a mobile is configured in an ad-hoc manner and picks a unique address from a specified range such as 169.254.x.x. In a LAN environment the mobile can obtain IP address from DHCP servers. In case of IPv6 networks, a mobile has the option of obtaining the IP address using stateless auto-configuration as well. In a wide area networking environment, mobile uses PPP to obtain the IP address by communicating with a NAS. Each of these processes takes of the order of few hundred milliseconds to few seconds depending upon the type of IP address acquisition process and operating system of the clients and servers. Since IP address acquisition is part of the handover process, it adds to the handover delay and thus it is desirable to reduce this timing as much as possible. There are few optimized techniques such as DHCP Rapid Commit [I-D.ietf-dhc-rapid-commit-opt], GPS-coordinate based IP address [GPSIP] available that attempt to reduce the handover time due to IP address acquisition time. However in all these cases the mobile also obtains the IP address after it moves to the new subnet and incurs some delay because of the signaling handshake between the mobile node and the DHCP server. In the following paragraph we describe few ways a mobile node can obtain the IP address proactively from the candidate target network and the associated tunnel setup procedure. These can broadly be defined into three categories such as PANA-assisted proactive IP address acquisition, IKE-assisted proactive IP address acquisition and proactive IP address acquisition using DHCP only. Dutta, et al. Expires August 14, 2005 [Page 18] Internet-Draft MPA Framework February 2005 5.2.1 PANA-assisted proactive IP address acquisition In case of PANA-assisted proactive IP address acquisition, the mobile node obtains an IP address proactively from a candidate target network. The mobile node makes use of PANA messages to trigger the address acquisition process on the DHCP relay agent that co-locates with PANA authentication agent in the access router in the candidate target network. Upon receiving a PANA message from the mobile node, the DHCP relay agent performs normal DHCP message exchanges to obtain the IP address from the DHCP server in the candidate target network. This address is piggy-backed in a PANA message and delivered to the client. 5.2.2 IKEv2-assisted proactive IP address acquisition IKEv2-assisted proactive IP address acquisition works when an IPsec gateway and a DHCP relay agent are resident within each access router in the candidate target networks. In this case, the IPsec gateway and DHCP relay agent in a candidate target network help the mobile node acquire the IP address from the DHCP server in the candidate target network. The MN-AR key established during the pre-authentication phase is used as the IKEv2 pre-shared secret needed to run IKEv2 between the mobile node and the access router. The IP address from the candidate target network is obtained as part of standard IKEv2 procedure, with using the co-located DHCP relay agent for obtaining the IP address from the DHCP server in the target network using standard DHCP. The obtained IP address is sent back to the client in the IKEv2 Configuration Payload exchange. In this case, IKEv2 is also used as the tunnel management protocol for a proactive handover tunnel (see Section 5.4). 5.2.3 Proactive IP address acquisition using DHCP only As another alternative, DHCP may be used for proactively obtaining an IP address from a candidate target network without relying on PANA or IKEv2-based approaches by allowing direct DHCP communication between the mobile node and the DHCP relay or DHCP server in the candidate target network. In this case, the mobile node sends a unicast DHCP message to the DHCP relay agent or DHCP server in the candidate target network requesting an address, with using the address associated with the current physical interface as the source address of the request. When the message is sent to the DHCP relay agent, the DHCP relay agent relays the DHCP messages back and forth between the mobile node and the DHCP server. In the absence of a DHCP relay agent the mobile can also directly communicate with the DHCP server in the target network. The broadcast option in client's unicast DISCOVER message Dutta, et al. Expires August 14, 2005 [Page 19] Internet-Draft MPA Framework February 2005 should be set to 0 so that the relay agent or the DHCP server can send back the reply directly to the mobile using the mobile node's source address. In order to prevent malicious nodes from obtaining an IP address from the DHCP server, DHCP authentication should be used or the access router should install a filter to block unicast DHCP message sent to the remote DHCP server from mobile nodes that are not pre-authenticated. When DHCP authentication is used, the DHCP authentication key may be derived from the MPA-SA established between the mobile node and the authentication agent in the candidate target network. The proactively obtained IP address is not assigned to the mobile node's physical interface until the mobile has not moved to the new network. The IP address thus obtained proactively from the target network should not be assigned to the physical interface but rather to a virtual interface of the client. Thus such a proactively acquired IP address via direct DHCP communication between the mobile node and the DHCP relay or the DHCP server in the candidate target network may be carried with additional information that is used to distinguish it from other address assigned to the physical interface. Upon the mobile's entry to the new network, the mobile node can perform DHCP over the physical interface to the new network to get other configuration parameters such as SIP server, DNS server, etc., by using e.g., DHCP INFORM. This should not affect the ongoing communication between the mobile and correspondent host. Also, the mobile node can perform DHCP over the physical interface to the new network to extend the lease of the address that was proactively obtained before entering the new network. In order to maintain the DHCP binding for the mobile node and keep track of the dispensed IP address before and after the secure proactive handover, the same DHCP client identifier needs to be used for the mobile node for both DHCP for proactive IP address acquisition and DHCP performed after the mobile node enters the target network. The DHCP client identifier may be the MAC address of the mobile node or some other identifier. 5.3 Address resolution issues 5.3.1 Proactive duplicate address detection When the DHCP server dispenses an IP address, it updates its lease table, so that this same address is not given to another client for that specific period of time. At the same time the client also keeps a lease table locally so that it can renew when needed. In some Dutta, et al. Expires August 14, 2005 [Page 20] Internet-Draft MPA Framework February 2005 cases where a network consists of both DHCP and non-DHCP enabled clients, there is a probability that another client with the LAN may have been configured with an IP address from the DHCP address pool. In such scenario the server does a duplicate address detection based on ARP (Address Resolution Protocol) or IPv6 Neighbor Discovery before assigning the IP address. This detection procedure may take up to 4 sec to 15 sec [MAGUIRE] and will thus contribute to a larger handover delay. In case of proactive IP address acquisition process, this detection is performed ahead of time and thus does not affect the handover delay at all. By performing the duplicate address detection ahead of time, we reduce the handover delay factor. 5.3.2 Proactive address resolution update During the process of pre-configuration, the address resolution mappings needed by the mobile node to communicate with nodes in the target network after attaching to the target network can also be known, where the nodes may be the access router, authentication agent, configuration agent and correspondent node. There are several possible ways of performing such proactive address resolution. o Use an information service mechanism [NETDISC] to resolve the MAC addresses of the nodes. This might require each node in the target network to involve in the information service so that the server of the information service can construct the database of proactive address resolution. o Extend the authentication protocol used for pre-authentication or the configuration protocol used for pre-configuration to support proactive address resolution. For example, if PANA is used as the authentication protocol for pre-authentication, PANA messages may carry AVPs used for proactive address resolution. In this case, the PANA authentication agent in the target network may perform address resolution for on behalf of the mobile node. o One can also make use of DNS to map the MAC address of the specific interface associated with a specific IP address of the network element in the target network. One may define a new DNS resource record (RR) to proactively resolve the MAC addresses of the nodes in the target network. But this approach may have its own limitations since a MAC address is a resource that is bound to an IP address, not directly to a domain name. When the mobile node attaches to the target network, it installs the proactively obtained address resolution mappings without necessarily performing address resolution query for the nodes in the target network. Dutta, et al. Expires August 14, 2005 [Page 21] Internet-Draft MPA Framework February 2005 On the other hand, the nodes that reside in the target network and are communicating with the mobile node should also update their address resolution mappings for the mobile node as soon as the mobile node attaches to the target network. The above proactive address resolution methods could also be used for those nodes to proactively resolve the MAC address of the mobile node before the mobile node attaches to the target network. However, this is not useful since the those nodes need to detect the attachment of the mobile node to the target network before adopting the proactively resolved address resolution mapping. A better approach would be integration of attachment detection and address resolution mapping update. This is based on gratuitously performing address resolution [RFC3344], [RFC3775] in which the mobile node sends an ARP Request or an ARP Reply in the case of IPv4 or a Neighbor Advertisement in the case of IPv6 immediately after the mobile node attaches to the new network so that the nodes in the target network can quickly update the address resolution mapping for the mobile node. 5.4 Tunnel management After an IP address is proactively acquired from the DHCP server in a candidate target network, a proactive handover tunnel is established between the mobile node and the access router in the candidate target network. The mobile node uses the acquired IP address as the tunnel inner address and most likely it assigns the address to a virtual interface. The proactive handover tunnel is established using a tunnel management protocol. When IKEv2 is used for proactive IP address acquisition, IKEv2 is also used as the tunnel management protocol. Alternatively, when PANA is used for proactive IP address acquisition, PANA may be used as the secure tunnel management protocol. Once the proactive handover tunnel is established between the mobile node and the access router in the candidate target network, the access router also needs to perform proxy address resolution on behalf of the mobile node so that it can capture any packets destined to the mobile node's new address. Since mobile needs to be able to communicate with the correspondent node while in the previous network some or all part of binding update and data from the correspondent node to mobile node need to be sent back to the mobile node over a proactive handover tunnel. When SIP Mobility is used for the mobility management protocol, the new address as the contact address is reported to the correspondent node using SIP Re-INVITE. Once the correspondent node's SIP user agent obtains the new contact address it sends the OK to the new contact Dutta, et al. Expires August 14, 2005 [Page 22] Internet-Draft MPA Framework February 2005 address which actually belongs to the target network. The access router in the target network picks up the OK signal as it was directed to the new contact address and tunnels it to the mobile in its previous network. Final ACK message is received from the mobile to the correspondent node. Data from the mobile to the correspondent node may not need to be tunneled in the absence of ingress filtering. After completion of the SIP Re-INVITE signaling handshake, the data from the correspondent node is sent to mobile via the proactive handover tunnel. In order for the traffic to be directed to the mobile node after the mobile node attaches to the target network, the proactive handover tunnel needs to be deleted or disabled. The tunnel management protocol used for establishing the tunnel is used for this purpose. Alternatively, when PANA is used as the authentication protocol the tunnel deletion or disabling at the access router can be triggered by means of PANA update mechanism as soon as the mobile moves to the target network. A link-layer trigger ensures that the mobile node is indeed connected to the target network and can also be used as the trigger to delete or disable the tunnel. 5.5 Binding Update There are several kinds of binding update mechanisms for different mobility management schemes. In some cases such as Mobile IPv4 without RO binding update is sent to home agent only, binding update is sent both to the home agent and corresponding host in case of Mobile IPv6. In case of SIP-based terminal mobility the mobile sends binding update using Re-INVITE to the correspondent node and REGISTER message to the Registrar. Based on the distance between the mobile and the correspondent node the binding update may contribute to the handover delay. SIP-fast handover [SIPFAST] provides several ways of reducing the handover delay due to binding update. In case of secure proactive handover using SIP-based mobility management we rule out the delay due to binding update completely, as it takes place in the previous network. Thus this scheme looks more attractive when the correspondent node is too far from the communicating mobile node. 5.6 Preventing packet loss In MPA case we did not observe any packet loss due to IP address acquisition, secured authentication and binding update. However, there may be some transient packets during link-layer handover and until the traffic to be directed to the mobile node after attaching to the target network. Those transient packets can be lost. Bicasting or buffering the transient packets at the access router can be used to minimize or eliminate packet loss. However, bicasting does not eliminate packet loss if link-layer handover is not Dutta, et al. Expires August 14, 2005 [Page 23] Internet-Draft MPA Framework February 2005 seamlessly performed. On the other hand, buffering does not reduce packet delay. While packet delay can be compensated by playout buffer at the receiver side for streaming application, playout buffer does not help much for interactive VoIP application which cannot tolerate for large delay jitters. Thus it is still important to optimize the link-layer handover anyway. 5.7 Link-layer security and mobility Using the MPA-SA established between the mobile node and the authentication agent in a candidate target network, during the pre-authentication phase, it is possible to bootstrap link-layer security in the candidate target network while the mobile node is in the current network in the following way. (1) The authentication agent in the candidate target network and the mobile node derives a PMK (Pair-wise Master Key) [I-D.ietf-eap-keying] using the MPA-SA that is established as a result of successful pre-authentication. Executions of EAP and an AAA protocol may be involved during pre-authentication to establish the MPA-SA. From the PMK, distinct TSKs (Transient Session Keys) [I-D.ietf-eap-keying] for the mobile node are directly or indirectly derived for each point of attachment of the candidate target network. (2) The authentication agent may install the keys derived from the PMK and used for secure association to points of attachment. The derived keys may be TSKs or intermediary keys from which TSKs are derived. (3) After the mobile node chooses the candidate target network as the target network and switches to a point of attachment in the target network (which now becomes the new network for the mobile node), it executes a secure association protocol such as IEEE 802.11i 4-way handshake [802.11i] using the PMK in order to establish PTKs (Pair-wise Transient Keys) and GTKs (Group Transient Keys) [I-D.ietf-eap-keying] used for protecting link-layer packets between the mobile node and the point of attachment. No additional execution of EAP authentication is needed here. (4) While the mobile node is roaming in the new network, the mobile node only needs to perform a secure association protocol with its point of attachment point and no additional execution of EAP authentication is needed either. Integration of MPA with link-layer handover optimization mechanisms such as 802.11r can be archived this way. The mobile node may need to know the link-layer identities of the point of attachments in the candidate target network to derive TSKs. Dutta, et al. Expires August 14, 2005 [Page 24] Internet-Draft MPA Framework February 2005 If PANA is used as the authentication protocol for pre-authentication, this is possible by carrying Device-Id AVPs in the PANA-Bind-Request message sent from the PAA [I-D.ietf-pana-pana], with each AVP containing the BSSID of a distinct access point. _________________ ____________________________ | Current Network | | CTN | | ____ | | ____ | | | | (1) pre-authentication | | | | | MN |<------------------------------->| AA | | | |____| | | |____| | | . | | | | | . | | | | |____.____________| | | | .movement | |(2) Keys | ____.___________________| | | | _v__ _____ | | | | |(3) secure assoc. | | | | | | MN |<------------------>| AP1 |<-------+ | | |____| |_____| | | | . | | | .movement | | | . | | | . | | | _v__ _____ | | | | |(4) secure assoc. | | | | | | MN |<------------------>| AP2 |<-------+ | | |____| |_____| | |_____________________________________________________| Figure 3: Bootstrapping Link-layer Security 5.8 Authentication in initial network attachment When the mobile node initially attaches to a network, network access authentication would occur regardless of the use of MPA. The protocol used for network access authentication when MPA is used for handover optimization can be a link-layer network access authentication protocol such as IEEE 802.1X or a higher-layer network access authentication protocol such as PANA. Dutta, et al. Expires August 14, 2005 [Page 25] Internet-Draft MPA Framework February 2005 6. Initial Implementation and Results We describe a specific scenario where we evaluate both MPA and non-MPA based approaches. This section describes details of one of the specific implementation for MPA and non-MPA. In addition to implementation details, this section also provides the evaluation results of optimized hand-off with MPA and compares it with non-MPA-based handover. 6.1 Network structure The experiment network structure is shown in Figure 4. Network 1 Network 2 Network 3 (oPoA) (nPoA) +--------+ +------------+ |Router 1|---------|Router 2(RA)|---------+ +---+----+ |PAA(AA) | | | |DHCP Relay | | | +--------+ |Agent (CA) | | +------------+ |-|DHCP | +------------+ | |CN | | |Server 1| | +------------+ |-|SIP-M Client| | +--------+ |-|DHCP | | +------------+ | | |Server 2 | | | +------------+ | | | | | +-----+ | +-----+ | |-|AP 1 | |-|AP 2 | | +-----+ +-----+ : : : : +------------+ +------------+ |MN |---->|MN | |SIP-M Client| |SIP-M Client| |PaC | |PaC | +------------+ +------------+ Figure 4: Network Structure There are three networks defined in the implementation environment. Network 1 is old point of attachment (oPoA), Network 2 is new point of attachment (nPoA), and network 3 is where the correspondent node (CN) resides. The mobile is initially in Network 1 and starts communicating with the correspondent node. Network 1, network 2, and network 3 do not need to be adjacent. In the implementation scenario however, network 1, network 2 and network 3 are one hop away. In the event of mobile's movement, a specific Mobility Management Protocol (MMP) can take care of continuity of streaming traffic set up by the Dutta, et al. Expires August 14, 2005 [Page 26] Internet-Draft MPA Framework February 2005 peer-to-peer application. Network 1 consists of DHCP Server 1, access point (AP) 1 and Access Router 1. Network 2 consists a DHCP Server 2, AP 2 and Access Router 2. AP 1 and AP 2 are 802.11 wireless LAN access points. Router 2 also works as a PANA Authentication Agent (PAA) [I-D.ietf-pana-pana] and a DHCP Relay Agent [RFC3046] for Network 2, but they can be separated. DHCP relay-agent also acts like a Configuration Agent (CA) that helps obtain the IP address for the mobile proactively from the neighboring target network. Network 3 consists of a Correspondent Node (CN) that communicates with the mobile node in Network 1. Both the correspondent node and mobile node are equipped with mobility enabled SIP client. Mobile SIP client is also equipped with PANA Client (PaC). In this specific case SIP proxies are not involved to set up the initial communication between the correspondent node and mobile node. Mobile Node (MN) uses 802.11 wireless LAN as the access method and can communicate via AP 1 before it moves to Network 2 where it communicates via AP 2. In this specific case, the Mobility Management Protocol (MMP) is SIP Mobility (SIP-M), configuration protocol is DHCP, authentication agent (AA) is PAA, configuration agent (CA) is DHCP Relay Agent and Access Router (AR) is Router 2 that can provide IP-in-IP tunneling [RFC1853] management functions. The MN is also equipped with IP-in-IP tunneling management function. Thus the mobile has the ability to set up a tunnel interface and detunnel the packets sent over the tunnel between the router 2 and the mobile. In this specific case, we have used IPv4, although one can as well use mobility management for IPv6 such as MIPv6 or SIP mobility over IPv6. 6.2 MPA Scenario The communication flow for MPA in our implementation environment is described below and in Figure 5 Step 0: As the MN bootstraps it associates with AP 1 and obtains the IP address old Care of Address (oCoA) from the DHCP Server 1 in network 1. The MN's SIP user agent communicates with CN's SIP user agent. After a successful connection setup between the mobile and correspondent node, a voice traffic flows between the MN and the CN. This voice traffic is carried over RTP/UDP. We have used RAT (Robust Audio Tool) as the media agent. In Step 1 (pre-authentication phase), there are some triggers to Step 1 such as AP 1's link level going down because of MN's movement. MN prepares to start the handover process and obtains the information about the required elements of the target network from an information server. Then the MN performs pre-authentication with PAA and derives the MN-CA key and MN-AR key from the MPA-SA if the pre-authentication Dutta, et al. Expires August 14, 2005 [Page 27] Internet-Draft MPA Framework February 2005 is successful. In Step 2 (pre-configuration phase), the MN performs pre-configuration by communicating with DHCP Proxy to obtain IP address and so forth. DHCP proxy and Authentication Agent (AA) are co-located in this case. This IP address is the new Care of Address (nCoA) the mobile would have obtained after moving to the new network. DHCP Proxy gets the IP address from DHCP Server 2. The new IP address of the mobile is relayed back to the mobile as part of its pre-authentication process. After the MN gets the new IP address (nCoA), an IP-in-IP tunnel is created between Router 2 and the mobile. At this point the behavior of the MN and Router 2 is basically followed by [RFC1853] and the signals are cryptographically protected by using the MN-CA key. In Step 3 (secure proactive handover main phase), once the mobile is configured with the new IP address (nCoA) on its virtual interface and a tunnel is set up between the mobile and R2, the MN sends SIP Re-invite with nCoA as its contact address to the CN. All the SIP Re-invite signaling are carried over the tunnel and so as the new RTP stream. Thus mobile receives the traffic in the old network even if the CN sends traffic to nCoA. Steps 4, 5 and 6 (secure proactive handover pre-switching phase, switching and secure proactive handover post-switching phase): As the mobile detects the new point of attachment and makes a decision to switch over to the new network it associates with AP 2. At this point the mobile configures itself by assigning the nCoA to its physical interface and updates the default router from the local cache that is stored during the pre-configuration phase in network 1. The MN sends a PANA-Update-Request message to the access router R2. This update message deletes the tunnel on the router R2 and deletes the tunnel locally on the mobile. Mobile's ARP entry with nCoA is also updated in the router R2 during the secure proactive handover thus reducing the delay due to ARP process that usually happens when a new node comes to a network. Dutta, et al. Expires August 14, 2005 [Page 28] Internet-Draft MPA Framework February 2005 Router 2 (RA) PAA (AA) DHCP DHCP Relay DHCP MN AP1 Server 1 AP2 Agent Server 2 CN |L2 Association| | | | | | |<- - - - - - >| | | | | | | oCoA assignment | | | | | |<------------------->| | | | | | SIP and voice communication start | | | |<----------------------------------------------------------->| | Step 1 Pre authentication with PAA | | | |<-------------------------------------->| | | | Step 2 Pre configuration with DHCP RA | | | |<-------------------------------------->| | | | | | | |DHCP Relay | | | | | | |<--------->| | | nCoA assignment | | | | | |<-------------------------------------->| | | |IP in IP tunnel is established with Router 2 | | |<-------------------------------------->| | | | Step 3 SIP Re-invite | | | | |<======================================>|<------------------>| |Voice traffic goes through IP in IP tunnel | | |<======================================>|<------------------>| | Step 4 Deletion of the tunnel | | | |<-------------------------------------->| | | X Step 5 Association with AP 2| | | | X<- - - - - - - - - - - - - - >| | | | X Voice traffic goes to nCoA | | | | |<----------------------------------------------------------->| <- - - - ->802.11 frame <--------->IP packet <=========>IP in IP tunneling packet X Voice Packet loss is happened. Figure 5: MPA Communication Flow in the implementation environment 6.3 Non-MPA Scenario For the comparison purposes, non-MPA scenario is also experimented and is described here. Non-MPA scenario does not provide any proactive handover mechanism as such but follows standard handover procedure. Following are the steps of non-MPA scenario in a likely similar Dutta, et al. Expires August 14, 2005 [Page 29] Internet-Draft MPA Framework February 2005 situation. There is no proactive handover involved in this case. Steps involved as part of initial communication setup while the mobile is in network 1 remain same as that of MPA part. Based on some policy decision such as signal-to-noise ratio, the mobile decides to switch to the new network. In first step, the MN associates with AP 2 and obtains new IP address from DHCP Server 2, then assigns the IP address to the network interface. In second step, the MN authenticates to the PAA. No data can flow through router R2, until the mobile successfully authenticates to the PAA. This adds the delay for post-authentication. In third step, the MN sends SIP Re-invite with the new IP address obtained from the DHCP server in the new network, then the voice traffic is destined to the new IP address. This binding update can taken potentially a lot of time if the mobile's target network and the correspondent node are far apart. Dutta, et al. Expires August 14, 2005 [Page 30] Internet-Draft MPA Framework February 2005 Router 2(RA) PAA(AA) DHCP DHCP Relay DHCP MN AP1 Server 1 AP2 Agent Server 2 CN |L2 Association| | | | | | |<- - - - - - >| | | | | | | IP address assignment | | | | |<------------------->| | | | | | SIP and voice communication start | | | |<----------------------------------------------------------->| | Association with AP 2 | | | | X<- - - - - - - - - - - - - - >| | | | X new IP address assignment | | | | X<-------------------------------------------------->| | X Authentication with PAA | | | | X<-------------------------------------->| | | X SIP Re-invite | | | | X<----------------------------------------------------------->| X Voice traffic goes to new IP address | | | |<----------------------------------------------------------->| <- - - - ->802.11 frame <--------->IP packet X Voice Packet loss is happened. Figure 6: Communication Flow for Non-MPA in the implementation environment 6.4 The evaluation and the results In case of MPA scenario, there is no packet loss during pre-authentication, and secure proactive handover before link-layer handover takes place when the mobile moves to the new network. Delay and associated packet loss are observed due to link-layer handover delay and tunnel deletion mechanism during the handover. This handover delay is limited to 170 ms including the link-layer delay. This amounts to 6 RTP packets being lost because of these processes. Note that in the measurement the RTP packets were not spaced equally with 20 ms intervals even at the sending side. Optimizing link-layer delay using a scheme such as [MACD] will reduce the total packets lost further. It is important to note that the scheme described in [MACD] has been experimented with HOSTAP driver only. We are also implementing other methods as described in Section 5 to optimize the handoff procedure further. It is also noted that the end-host processing contributes to the handoff delay as well for things such as tunnel deletion. Thus any system optimization techniques can also help reduce the handoff delay. Dutta, et al. Expires August 14, 2005 [Page 31] Internet-Draft MPA Framework February 2005 In case of non-MPA scenario, handover delay and attributed packet loss take place because of L2 handover during the movement, IP address assignment, post-authentication, and mobility binding update. Especially DHCP takes long time to complete the detection of duplicate of IP address in the network and binding update can take a long time if the correspondent node is too far from the mobile node. In our testbed non-MPA-based handover took up to 4 seconds delay due to all the above factors. Based on type of streaming traffic that was sent once in every 20 ms using RAT approximately 200 packets were lost. 6.5 Notes In this example network, a portion of function is omitted such as pre-authorization process, but it can be implemented to the network and it's not critical section for the handover. In this example network, candidate protocols can always be replaced by the other protocols, for example, Mobility management protocol can be replaced by Mobile IPv4 or Mobile IPv6. In that case, Home Agent could be in Network 3, similarly the tunnel management protocol can always be replaced by IKEv2 and IPsec tunnel mode. It is normal to assume that the performance values will be different based on the type of candidate protocols used. We also found that L2 delay can vary based on the drivers and operating system used. We can provide some examples of L2 delays here. For example Cisco Aironet 350 takes almost 200-300 ms under linux operating system. Orinoco and Dlink (Hostap) drivers take around 100-160 ms and 400-600 ms respectively under Linux operating system but Orinoco with Windows takes about 250 ms. Where as Hostap driver with managed mode takes about 30 ms handoff time which is comparable to that obtained by [MACD]. Dutta, et al. Expires August 14, 2005 [Page 32] Internet-Draft MPA Framework February 2005 7. Security Considerations This document describes a framework of a secure handover optimization mechanism based on performing handover-related signaling between a mobile node and one or more candidate target networks to which the mobile node may move in the future. This framework involves acquisition of the resources from the candidate target networks as well as data packet redirection from the candidate target networks to the mobile node in the current network before the mobile node physically connects to one of those candidate target networks. Acquisition of the resources from the candidate target networks must accompany with appropriate authentication and authorization procedures in order to prevent unauthorized mobile node from obtaining the resources. For this reason, it is important for the MPA framework to perform pre-authentication between the mobile node and the candidate target networks. The MN-CA key and the MN-AR key generated as a result of successful pre-authentication can protect subsequent handover signaling packets and data packets exchanged between the mobile node and the MPA functional elements in the candidate target networks. The MPA framework also addresses security issues when the handover is performed across multiple administrative domains. With MPA, it is possible for handover signaling to be performed based on direct communication between the mobile node and routers or mobility agents in the candidate target networks. This eliminates the need for a context transfer protocol for which known limitations exist in terms of security [I-D.ietf-eap-keying]. For this reason, the MPA framework does not require trust relationship among administrative domains or access routers, which makes thke framework more deployable in the Internet without compromising the security in mobile environments. Dutta, et al. Expires August 14, 2005 [Page 33] Internet-Draft MPA Framework February 2005 8. Acknowledgments We would like to thank Farooq Anjum and Raziq Yakub for their review of this document, Victor Fajardo, Kensaku Fujimoto and Provin Gurung for MPA prototype implementation support, and Subir Das for standardization support in the IEEE 802.21 WG. Dutta, et al. Expires August 14, 2005 [Page 34] Internet-Draft MPA Framework February 2005 9. References 9.1 Normative References [RFC3344] Perkins, C., "IP Mobility Support for IPv4", RFC 3344, August 2002. [RFC3748] Aboba, B., Blunk, L., Vollbrecht, J., Carlson, J. and H. Levkowetz, "Extensible Authentication Protocol (EAP)", RFC 3748, June 2004. [RFC3775] Johnson, D., Perkins, C. and J. Arkko, "Mobility Support in IPv6", RFC 3775, June 2004. [I-D.ietf-mobileip-lowlatency-handoffs-v4] Malki, K., "Low latency Handoffs in Mobile IPv4", draft-ietf-mobileip-lowlatency-handoffs-v4-09 (work in progress), June 2004. [I-D.ietf-mipshop-fast-mipv6] Koodli, R., "Fast Handovers for Mobile IPv6", draft-ietf-mipshop-fast-mipv6-03 (work in progress), October 2004. [I-D.ietf-seamoby-card-protocol] Liebsch, M., "Candidate Access Router Discovery", draft-ietf-seamoby-card-protocol-08 (work in progress), September 2004. [I-D.ietf-seamoby-ctp] Loughney, J., "Context Transfer Protocol", draft-ietf-seamoby-ctp-11 (work in progress), August 2004. [I-D.ietf-eap-keying] Aboba, B., "Extensible Authentication Protocol (EAP) Key Management Framework", draft-ietf-eap-keying-04 (work in progress), November 2004. [I-D.ietf-pana-pana] Forsberg, D., Ohba, Y., Patil, B., Tschofenig, H. and A. Yegin, "Protocol for Carrying Authentication for Network Access (PANA)", draft-ietf-pana-pana-07 (work in progress), December 2004. [RG98] ITU-T, "General Characteristics of International Telephone Connections and International Telephone Circuits: One-Way Transmission Time", ITU-T Recommendation G.114 1998. Dutta, et al. Expires August 14, 2005 [Page 35] Internet-Draft MPA Framework February 2005 [ITU98] ITU-T, "The E-Model, a computational model for use in transmission planning", ITU-T Recommendation G.107 1998. [ETSI] ETSI, "Telecommunications and Internet Protocol Harmonization Over Networks (TIPHON) Release 3: End-to-end Quality of Service in TIPHON systems; Part 1: General aspects of Quality of Service.", ETSI TR 101 329-6 V2.1.1. 9.2 Informative References [I-D.ietf-mobike-design] Kivinen, T. and H. Tschofenig, "Design of the MOBIKE protocol", draft-ietf-mobike-design-01 (work in progress), January 2005. [I-D.ietf-hip-base] Moskowitz, R., "Host Identity Protocol", draft-ietf-hip-base-01 (work in progress), October 2004. [RFC2679] Almes, G., Kalidindi, S. and M. Zekauskas, "A One-way Delay Metric for IPPM", RFC 2679, September 1999. [RFC2680] Almes, G., Kalidindi, S. and M. Zekauskas, "A One-way Packet Loss Metric for IPPM", RFC 2680, September 1999. [RFC2681] Almes, G., Kalidindi, S. and M. Zekauskas, "A Round-trip Delay Metric for IPPM", RFC 2681, September 1999. [RFC1853] Simpson, W., "IP in IP Tunneling", RFC 1853, October 1995. [RFC3046] Patrick, M., "DHCP Relay Agent Information Option", RFC 3046, January 2001. [I-D.ietf-dhc-rapid-commit-opt] Kim, P., Volz, B. and S. Park, "Rapid Commit Option for DHCPv4", draft-ietf-dhc-rapid-commit-opt-05 (work in progress), June 2004. [SIPMM] Schulzrine, H., "Application Layer Mobility Using SIP", MC2R. [CELLIP] Cambell, A., Gomez, J., Kim, S., Valko, A. and C. Wan, "Design, Implementation, and Evaluation of Cellular IP", IEEE Personal communication Auguest 2000. [HAWAII] Ramjee, R., Porta, T., Thuel, S., Varadhan, K. and S. Wang, "HAWAII: A Domain-based Approach for Supporting Mobility in Wide-area Wireless networks", . Dutta, et al. Expires August 14, 2005 [Page 36] Internet-Draft MPA Framework February 2005 [IDMP] Das, S., Dutta, A., Misra, A. and S. Das, "IDMP: An Intra-Domain Mobility Management Protocol for Next Generation Wireless Networks", IEEE Wireless Communication Magazine October 2000. [I-D.ietf-mobileip-reg-tunnel] Calhoun, P., Montenegro, G., Perkins, C. and E. Gustafsson, "Mobile IPv4 Regional Registration", draft-ietf-mobileip-reg-tunnel-09 (work in progress), July 2004. [YOKOTA] Yokota, H., Idoue, A. and T. Hasegawa, "Link Layer Assisted Mobile IP Fast Handoff Method over Wireless LAN Networks", Proceedings of ACM Mobicom 2002. [MACD] Shin, S., "Reducing MAC Layer Handoff Latency in IEEE 802.11 Wireless LANs", MOBIWAC Workshop . [SUM] Dutta, A., Zhang, T., Madhani, S., Taniuchi, K., Ohba, Y. and H. Schulzrinne, "Secured Universal Mobility", WMASH 2004. [SIPFAST] Dutta, A., Madhani, S. and H. Schulzrinne, "Fast handoff Schemes for Application Layer Mobility Management", PIMRC 2004. [MITH] Gwon, Y., Fu, G. and R. Jain, "Fast Handoffs in Wireless LAN Networks using Mobile initiated Tunneling Handoff Protocol for IPv4 (MITHv4)", Wireless Communications and Networking 2003, January 2005. [NETDISC] Anjum, F., Das, S., Dutta, A., Fajardo, V., Madhani, S., Ohba, Y., Taniuchi, K., Yaqub, R. and T. Zhang, "A proposal for MIH function and Information Service", A contribution to IEEE 802.21 WG , January 2005. [GPSIP] Dutta, A., "GPS-IP based fast-handoff for Mobiles", NYMAN 2003. [MAGUIRE] Vatn, J. and G. Maguire, "The effect of using co-located care-of-address on macro handover latency", . Dutta, et al. Expires August 14, 2005 [Page 37] Internet-Draft MPA Framework February 2005 Authors' Addresses Ashutosh Dutta Telcordia Technologies 1 Telcordia Drive Piscataway, NJ 08854 USA Phone: +1 732 699 3130 EMail: adutta@research.telcordia.com Yoshihiro Ohba Toshiba America Research, Inc. 1 Telcordia Drive Piscataway, NJ 08854 USA Phone: +1 732 699 5305 EMail: yohba@tari.toshiba.com Kenichi Taniuchi Toshiba America Research, Inc. 1 Telcordia Drive Piscataway, NJ 08854 USA Phone: +1 732 699 5308 EMail: ktaniuchi@tari.toshiba.com Henning Schulzrinne Columbia University Department of Computer Science 450 Computer Science Building New York, NY 10027 USA Phone: +1 212 939 7004 EMail: hgs@cs.columbia.edu Dutta, et al. Expires August 14, 2005 [Page 38] Internet-Draft MPA Framework February 2005 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 (2005). 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. Dutta, et al. Expires August 14, 2005 [Page 39]