INTERNET DRAFT James Kempf, Editor Category: Informational Sun Microsystems Title: draft-seamoby-paging-problem-statement-00.txt Date: Feburary 2001 Paging Problem Statement Status of this Memo This document is a working group contribution for the Seamoby Working Group. Distribution of this memo is unlimited. This document is an Internet-Draft and is in full conformance with all provisions of Section 10 of RFC2026. 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. Copyright (C) The Internet Society 2000. All Rights Reserved. Abstract The IESG has requested that the Seamoby Working Group develop a problem statement about the need for additional protocol work to support paging for seamless IP mobility. This draft describes paging, assesses the need for paging at L3, and presents a list of recommendations for Seamoby charter items regarding work on paging. 1.0 Introduction Many existing radio link protocols and mobile systems support James Kempf, Editor expires August 2001 [Page 1] INTERNET DRAFT Feburary 2001 location of and radio link establishment with mobile nodes that are not actively listening for delivery of IP packets. This functionality allows mobile nodes to reduce power consumption and decreases signalling load on the network for tracking mobiles that are not actively participating in IP packet generation or reception. When a mobile is in low power consumption mode, special steps need to be taken to locate the mobile. These steps differ depending on the radio link, but the generic name for this process is paging. In this document, after some initial definitions and material related to more clearly explaining what paging is, we assess the need for paging in existing IP mobility protocols (namely Mobile IP [1] [2]) and briefly discuss the potential need for paging in a future micromobility protocol. We then develop a list of work items for the Seamoby working group related to this need. 2.0 Definitions The following definitions are relevent with respect to clarifying the paging functionality: Dormant Mode - A state in which a mobile can not receive IP packets, but may still be able to receive L2 signalling related to tracking the mobile's location. A mobile may transition out of dormant mode due to a lower layer interface message (possibly triggered by receipt of an IP packet) or a timer expiration. A dormant terminal may signal to the network when it changes paging area, but may delay the signal until the next time it wakes up (if, for example, the mobile is using a time-slotted dormant mode algorithm). Note that a terminal which is continuously listening to L2 for receipt of L3 packets is not in dormant mode. Paging - As a consequence of a mobile-bound packet destined for a mobile currently in dormant mode, messaging by the L2 access point directed to locating the mobile and establishing a last hop connection. This messaging is in addition to simply delivering the packet to the mobile, i.e. last hop routing of packets is NOT considered to be paging. Paging Area - Collection of last hop routers and/or L2 access points that are searched to locate a dormant mode node. A paging area does not necessarily correspond to an IP subnet. 3.0 Discussion of Paging Dormant mode is advantageous to a mobile node for the following James Kempf, Editor expires August 2001 [Page 2] INTERNET DRAFT Feburary 2001 reasons: - Power savings. By reducing the amount of time the mobile is required to listen to L2, the drain on the mobile node's battery is reduced. - Reduced signalling for location tracking. By requiring the mobile to only signal when it crosses a paging area boundary rather than when it switches between L2 access points, the amount of signalling for tracking the mobile is reduced. Most existing radio L2 protocols provide support for dormant mode. Some such protocols also support paging, while others do not. The important distinction is whether the radio L2 protocol supports periodic signalling between the mobile and the network on a reduced basis, for example only when the mobile crosses a paging area boundary, for the purpose of giving the network a rough idea of the mobile's location. Some deployments of paging reduce signalling even further by using heuristics to determine where the mobile is located, in which case, no signalling is required while the mobile is in dormant mode. An incoming packet or, for voice-based systems, voice call is directed to the paging area where the mobile last reported. The network performs an L2 page by sending out a signal on a special radio channel that awakens the mobile. The signal may be repeated until the mobile answers or a timeout occurs. In the former case, the packet is delivered or the call connected, in the latter, the mobile is assumed to be unreachable. An L3 paging protocol is sensible only if L2 supports paging. In the absence of L2 paging, L3 paging achieves no reduction in power drain because the mobile is forced to monitor L2 for L3 paging messages. Having the mobile constantly listen to L2 for L3 packets is NOT considered to be adequate L2 support, because it does not achieve enough power savings to be meaningful. Some reduction in signalling load might be achieved, however, since the mobile would only need to update its location with the network when it moves from one paging area to another. 4.0 Is L3 Paging Necessary? In this section, we consider whether L3 paging support is necessary. As discussed above, L2 support for paging is assumed. There are two cases to consider: paging for networks consisting of a homogeneous radio L2 technology and networks in which handoffs are occuring between heterogeneous radio L2 technologies. We first consider the two cases given the existing IETF mobility protocol, namely Mobile IP. We also briefly consider the relationship between paging and a potential future micromobility protocol. James Kempf, Editor expires August 2001 [Page 3] INTERNET DRAFT Feburary 2001 4.1 Homogeneous Technology Networks Subnets constitute the unit of signalling for presence at L3. When a mobile node moves from one subnet to another, Mobile IP signalling is required to change the mobile's care-of address. This signalling establishes the mobile's presence in the new subnet. Paging areas constitute the unit of signalling for presence at L2. There are many ways to configure how paging areas are utilized to obtain a rough idea of where the mobile is. One way is to use regional registrations. When the mobile moves from one paging area to another, it signals the network indicating that it has arrived in the new paging area. This signalling establishes the mobile presence in the paging area. Alternate possibilities are for the network to page in the home area, then the area of last origination, then in the area of last power-on registration, etc. In this case, the network is not tracking the mobile, so the L2 paging system has only a very rough idea of where the mobile is located. The exact algorithm depends on how a particular operator configures their network, as well as the available support at L2. Given these two units of presence at L2 and L3, the most important point with regard to whether or not paging is needed in homogeneous networks is how subnets are mapped into paging areas. To determine whether additional L3 paging protocol work is required, we first need to determine whether we can support location of and link establishment with a mobile in dormant mode given some mapping between subnets and paging areas using Mobile IP. Standard Mobile IP requires the mobile node to actively listen for or solicit foreign agent or router advertisements and participate in registering, so it is of little use if the mobile is in dormant mode. However, network assisted handoff techniques used for fast handoff [3] [4] allow the network to track the mobile without requiring any signalling from the mobile. The L2 signalling that occurs when a dormant mobile crosses paging area boundaries (provided the operator has configured L2 regional registrations) can be used as the "L2 trigger" for network assisted handoff techniques to allow the network to track the mobile in dormant mode, with perhaps some enhancements to optimize the ability of the network to update the home agent and/or any hierarchical agents about the mobile's location. We need to examine paging area to subnet mappings in order to determine when we can utilize the L2 paging area update trigger to trigger network assisted handoff in dormant mode. In general, the mapping between paging areas and subnets can be arbitrary, but we consider here a smooth subset relationship, in which paging areas are subsets of subnets or vice versa. Network topologies in which one James Kempf, Editor expires August 2001 [Page 4] INTERNET DRAFT Feburary 2001 subnet is split between two or more paging areas are therefore eliminated. The restriction is arbitrary, but by starting here, we can discover whether additional work is needed. If it turns out that work is needed, then more complex topologies can be investigated in the course of actually doing the work. There are three cases: 1) The topological boundaries of the paging area and subnet are identical. 2) Multiple paging areas are part of the same subnet. 3) Multiple subnets are part of the same paging area. In the case where L2 paging areas map one to one onto IP subnets (and hence Mobile IPv4 Foreign Agents or IPv6 Access Routers), it is possible to use L2 paging together with Mobile IP network assisted handoff techniques for the network to track the mobile's location. Enhancements to network-assisted handoff techniques can allow the network to track the mobile as it moves from paging area (== subnet) to paging area. New Mobile IP signalling for the mobile node to inform the network when it is going into dormant mode could possibly optimize this. With network tracking, if a packet for the dormant mobile comes in, it is routed by the Regional Attendant, Regional Foreign Agent, Mobile Access Point, or Home Agent to the last hop Foreign Agent or Access Router at the mobile's last known position, where the L2 page is performed. Once the L2/L3 address mapping is known, the packet is delivered. The case where multiple L2 paging areas map to a single IP subnet is the same as above, with the exception that the last hop Mobile IPv4 Foreign Agent or IPv6 Access Router for the subnet performs paging in multiple paging areas to locate the mobile. In the case where a single L2 paging area maps onto multiple IP subnets, it is not possible to directly use enhanced network assisted handoff techniques between last hop Access Routers or Foreign Agents to track the mobile's location as it moves, because the mobile does not signal its location when it changes subnets. Within the set of subnets that span the paging area, the mobile's movement is invisible to the L2 paging system, so a packet delivered to the mobile's last known location may result in a page that is answered in a different subnet. In this case, some support is required at L3 for locating a dormant mode mobile that has moved into a different subnet. James Kempf, Editor expires August 2001 [Page 5] INTERNET DRAFT Feburary 2001 4.2 Heterogeneous Technology Networks In a network composed of links with multiple technologies, there may be commonalities in the corresponding L2 paging protocols that would allow a mapping to be established between the L2 protocols and an abstract L3 protocol. For example, assume we have a common paging area identifier defined at L3 that is mapped to each L2 paging protocol by the access points. An L3 paging message containing the identifier is sent to multiple access points, where the appropriate L2 paging message is sent based on the particular technology implemented by the access points. The results are then returned by the L2 paging responses, mapped back into L3 by the access points, and delivered back to the origin of the page. Note, however, that the same analysis as applied to homogeneous networks regarding the mapping between paging areas and subnets applies here as well. For example, if there is a one-to-one mapping between L2 paging areas and subnets, then there can be only a single technology per subnet, and movement across paging area boundries by a dual technology moble in dormant mode on both interfaces can be tracked through mobile IP network assisted handoff. Similarly, if more than one paging area exists per subnet, even if the areas are from multiple technologies, the Foreign Agent or Access Router can page in the multiple areas at L2 and a dual mode mobile can be tracked as it crosses paging area boundaries. But if multiple subnets are contained within a paging area or for more complex topologies, the lack of clear identification between subnet boundaries at L3 and paging area boundaries at L2 requires some means of locating the mobile at L3. An additional case to consider is when a subnet consists of multiple access tchnologies. A wireless access point usually provides L2 bridge behavior to the wired link with which it is connected. If two access point with incompatible technologies and non-overlapping cells are connected to the same subnet, a mobile node with interfaces to both technologies would need paging from both technologies. If reachability can be established simply by ARP or neighbor discovery, no L3 paging is needed. However, note that ARP or neighbor discovery requires that a functional L2 link be available to the mobile, since these techniques reach all the way to the mobile and assume a functional L2. If the mobile is currently in the sleep phase of a time-slotted dormant mode, it will fail to respond to these requests. In this case, some means of triggering an L2 page from L3 is necessary to find the mobile. 4.3 Paging and Micromobility If the Seamoby Working Group decides that an L3 micromobility James Kempf, Editor expires August 2001 [Page 6] INTERNET DRAFT Feburary 2001 protocol is necessary for intra-subnet mobility, then the above analysis is no longer complete. A micromobility protocol may require some type of paging support. The design team does not want to include any further discussion of paging and micromobility at this point, because it is not clear whether micromobility will be pursued by Seamoby and hence such discussion would be premature. 5.0 What Exactly is the Problem? While the above analysis has identified situations in which location of a mobile in dormant mode may require some action at L3, it is important not to misinterpret this finding as an endorsement for a complex L3 paging algorithm. The problem to be solved is the location of the mobile because it has moved while in dormant mode. L3 paging is just one solution to the problem, there are alternatives. For example, paging involves action by the network at the time a packet or, in voice systems, the call comes in. An alternative solution would be for the network to distribute information about a dormant mode mobile when it goes dormant. In the case of multiple subnets per paging area, the Foreign Agent or Access Router in the subnet where the mobile goes dormant could distribute the mobile's L2 address and IP address, and the last known Foreign Agent or Access Router address to other Foreign Agents or Access Routers in the paging area. When a packet comes in, the subnet where the mobile was last registered does the page, but the mobile answers in a different subnet. The Foreign Agent or Access Router has the IP address, contacts the previous Foreign Agent or Access Router, and negotiates a handoff and assignment of an new CoA for the mobile. This solution leverages off of the existing Mobile IP infrastructure considerably more than would a paging algorithm designed to be orthogonal to Mobile IP. If further study determines that, in fact, an L3 paging algorithm is the right solution to the problem, then the algorithm for doing L3 paging cannot involve any sort of L3 signalling to the mobile, because the mobile is, by definition, probably not listening to L2 for L3 traffic. It may be listening periodically, if the L2 dormant mode technology involves time slotting, but most of the time it will not be. So an L3 page could only extend as far as the L2 access points, at which point it would have to be translated into L2 paging. As a consequence, L2 technologies that do not support paging would still derive no benefit from the L3 paging algorithm. 6.0 Recommendations The design group recommends the following charter items for Seamboy: James Kempf, Editor expires August 2001 [Page 7] INTERNET DRAFT Feburary 2001 1) Since the design group has identified at least one network deployment scenerio where existing Mobile IP technology cannot find a mobile in dormant mode, protocol work is necessary to define a way for the network to find a mobile that is currently in dormant mode. 2) The work defined above should be pursued in a way that is maximally consistent with Mobile IP and other existing IETF protocols. In particular, it is expected, based on the analysis undertaken by the design team, that network assisted handoff can be heavily leveraged in any new protocol work. 3) If it is determined that extensions to existing protocols are completely inadequate to solve location of a dormant mode mobile, or if the Seamoby working group decides to pursue a micromobility protocol that requires paging, the Seamoby group should undertake the design of a new paging protocol. 4) There is some evidence that cellular operators' deployments of paging are highly variable, and may, in fact, be suboptimal in many cases with respect to supporting IP. The Seamoby working group should write a BCP which explains how to perform IP subnet to paging area mapping and which techniques to use when, so network designers in wireless networks have a guide when they are setting up their networks. 7.0 References [1] C. Perkins, editor. "IP Mobility Support", RFC 2002, October, 1966. [2] Johnson, D., and C. Perkins, "Mobility Support in IPv6", draft- ietf-mobileip-ipv6-13.txt, a work in progress. [3] Calhoun, P., et. al., "Foreign Agent Assisted Hand-off", draft- calhoun-mobileip-proactive-fa-03.txt, a work in progress. [4] Perkins, C., Editor, "Fast Handovers for Mobile IPv6", draft- perkins-mobileip-handover-00.txt, a work in progress. 8.0 Editor's Address Questions about this memo can be directed to: James Kempf James Kempf, Editor expires August 2001 [Page 8] INTERNET DRAFT Feburary 2001 Sun Labs California Sun Microsystems, Inc. 901 San Antonio Rd., UMPK15-214 Palo Alto, CA, 94303 USA Phone: +1 650 786 5890 Fax: +1 650 786 6445 E-Mail: james.kempf@sun.com 9.0 Full Copyright Statement Copyright (C) The Internet Society (2001). All Rights Reserved. This document and translations of it may be copied and furnished to others, and derivative works that comment on or otherwise explain it or assist in its implementation may be prepared, copied, published and distributed, in whole or in part, without restriction of any kind, provided that the above copyright notice and this paragraph are included on all such copies and derivative works. However, this docu- ment itself may not be modified in any way, such as by removing the copyright notice or references to the Internet Society or other Inter- net organizations, except as needed for the purpose of developing Internet standards in which case the procedures for copyrights defined in the Internet Standards process must be followed, or as required to translate it into languages other than English. The limited permis- sions granted above are perpetual and will not be revoked by the Internet Society or its successors or assigns. This document and the information contained herein is provided on an "AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING TASK FORCE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WAR- RANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE." James Kempf, Editor expires August 2001 [Page 9]