《计算机网络》课程教学资源（参考文献）MOBILE NETWORKING THROUGH MOBILE IP

TUT。RIAL MOBILE NETWORKING THROUGH MOBILE P CHARLES E. PERKINS Mobile IP is a proposed standard protocol that builds on the Internet Protocol by making mobility transparent to applications and higher level protocols like TCP to us as we move will giv net offers access us the tools to build new information computing environments sources worldwide, typi wherever we go. Those cally we do not expect 国变 who have little interest in nefit from that access mobility per se will still ntil we arrive at some familiar point-whether home, office, benefit from the ability to resume previous applications when or school. However, the increasing variety of wireless devices they reconnect. This is especially convenient in a wireless lan offering IP connectivity, such as PDAs, handhelds, and digi- office environment, where the boundaries between attachment tal cellular phones, is beginning to change our perceptions of points are not sharp and are often invisible the Internet The evolution of mobile networking will differ from that of To understand the contrast between the current realities telephony in some important respects. The endpoints of a tele- of IP connectivity and future possibilities, consider the tran- phone connection are typically human; computer applications sition toward mobility that has occurred in telephony over are likely to involve interactions between machines without the past 20 years. An analogous transition in the domain of human intervention. Obvious examples of this are mobile com networking, from dependence on fixed points of attachment puting devices on airplanes, ship aepen: on position-finding and automobiles. mobile to the flexibility afforded by mobility, has just begun. e Mobile computing and networking should not be con- devices, such as a satellite global positioning system, to work sed with the portable computing and networking we have in tandem with wireless access to the Internet. today. In mobile networking, computing activities are not dis- Another difference may well be rate of adoption. It took rupted when the user changes the computer's point of attach- many years for mobile phones to become cheap and light ment to the Internet. Instead, all the needed reconnection weight enough to be perceived as convenient. Because wireless o truly mobile computing offers many advantages. Confident nizers have already found user acceptance, mobile comput- cess to the Internet anytime, anywhere will help free us from ing may become popular much more quickly. the ties that bind us to our desktops. Consider how cellular However, there are still some technical obstacles that must phones have given people new freedom in carrying out their be overcome before mobile networking can become wide work. Taking along an entire computing environment has the spread. The most fundamental is the way the Internet Proto- potential not just to extend that flexibility but to fundamental- col, the protocol that connects the networks of todays Inter ly change the existing work ethic. Having the Internet available net, routes packets to their destinations according to IP 98/s100o1998IEE IEEE INTERNET COMPUTING

. TUTORIAL 58 1089-7801/98/$10.00 ©1998 IEEE IEEE INTERNET COMPUTING TUTORIAL MOBILE NETWORKING THROUGH MOBILE IP CHARLES E. PERKINS Sun Microsystems Although the Inter￾net offers access to information sources worldwide, typi￾cally we do not expect to benefit from that access until we arrive at some familiar point—whether home, office, or school. However, the increasing variety of wireless devices offering IP connectivity, such as PDAs, handhelds, and digi￾tal cellular phones, is beginning to change our perceptions of the Internet. To understand the contrast between the current realities of IP connectivity and future possibilities, consider the tran￾sition toward mobility that has occurred in telephony over the past 20 years. An analogous transition in the domain of networking, from dependence on fixed points of attachment to the flexibility afforded by mobility, has just begun. Mobile computing and networking should not be con￾fused with the portable computing and networking we have today. In mobile networking, computing activities are not dis￾rupted when the user changes the computer’s point of attach￾ment to the Internet. Instead, all the needed reconnection occurs automatically and noninteractively. Truly mobile computing offers many advantages. Confident access to the Internet anytime, anywhere will help free us from the ties that bind us to our desktops. Consider how cellular phones have given people new freedom in carrying out their work. Taking along an entire computing environment has the potential not just to extend that flexibility but to fundamental￾ly change the existing work ethic. Having the Internet available to us as we move will give us the tools to build new computing environments wherever we go. Those who have little interest in mobility per se will still benefit from the ability to resume previous applications when they reconnect. This is especially convenient in a wireless LAN office environment, where the boundaries between attachment points are not sharp and are often invisible. The evolution of mobile networking will differ from that of telephony in some important respects. The endpoints of a tele￾phone connection are typically human; computer applications are likely to involve interactions between machines without human intervention. Obvious examples of this are mobile com￾puting devices on airplanes, ships, and automobiles. Mobile networking may well also come to depend on position-finding devices, such as a satellite global positioning system, to work in tandem with wireless access to the Internet. Another difference may well be rate of adoption. It took many years for mobile phones to become cheap and light￾weight enough to be perceived as convenient. Because wireless mobile computing devices such as PDAs and pocket orga￾nizers have already found user acceptance, mobile comput￾ing may become popular much more quickly. However, there are still some technical obstacles that must be overcome before mobile networking can become wide￾spread. The most fundamental is the way the Internet Proto￾col, the protocol that connects the networks of today’s Inter￾net, routes packets to their destinations according to IP Mobile IP is a proposed standard protocol that builds on the Internet Protocol by making mobility transparent to applications and higher level protocols like TCP

M O addresses These addresses are associated with a fixed net- Ip address is connected. The network number is derived york location much as a nonmobile phone number is asso- from the IP address by masking off some of the low-order ciated with a physical jack in a wall. When the packets des- bits. Thus, the Ip address typically carries with it informa- tination is a mobile node, this means that each new point of tion that specifies the Ip nodes point of attachment attachment made by the node is associated with a new net- To maintain existing transport-layer connections(see the work number and, hence, a new IP address, making trans- sidebar"Nomadicity: How Mobility Will Affect the Protocol parent mobility impossible. Stack"on the next pages) as the mobile node moves from place Mobile IP(RFC 2002), a standard proposed by a work to place, it must keep its IP address the same InTCP(which ing group within the Internet Engineering Task Force, was accounts for the overwhelming majority of Internet connec- designed to solve this problem by allowing the mobile node tions), connections are indexed by a quadruplet that contains to use two IP addresses: a fixed home address and a care-of the IP addresses and port numbers of both connection end- address that changes at each new point of attachment. This points. Changing any of these four numbers will cause the con- article will present the Mobile IP standard in moderate nection to be disrupted and lost. On the other hand, correct technical detail and point the reader toward a wealth of fur- delivery of packets to the mobile node's current point of attach- ther information.2.In addition, readers can go to the side- ment depends on the network number contained within the bar Mobile IP Web Resources in this issue's IC Online at mobile node's IP address, which changes at new points of http://computer.org/internet/foraconvenientsetofhyperattachmentTochangetheroutingrequiresanewIpaddress linked resources associated with the new point I also describe how Mobile ip uses two iP addresses. of attachment Mobile ip has beel IP version 6, 4.5 the product a fixed home address and a care-of designed to solve this problem of a major effort within the address that changes at each new by allowing the mobile nodeto IETF to engineer an eventu- point of attachment use two IP addresses(see the al replacement for the cur- sidebar rent version of IP Although Terminology" for definitions Pv6 will support mobility to a greater degree than IPv4, it of italicized terms). In Mobile IP, the home address static and will still need Mobile IP to make mobility transparent to is used, for instance, to identify TCP connections. The care-of applications and higher level protocols such as TCP. address changes at each new point of attachment and can be There is a great deal of interest in mobile computing and thought of as the mobile node's topologically significant pparently in Mobile Ip as a way to provide for it. a quick address; it indicates the network number and thus identifies Web search for items related to Mobile IP returned over the mobile node's point of attachment with respect to the net 60,000 hits-impressive even given the notorious lack of work topology. The home address makes it appear that the selectivity for such procedures. Mobile IP forms the basis mobile node is continually able to receive data on its home net- either directly or indirectly of many current research efforts work, where Mobile IP requires the existence of a network and products. The Cellular Digital Packet Data(CDPD), node known as the home agent. Whenever the mobile node is for example, has created a widely deployed communications not attached to its home network(and is therefore attached infrastructure based on a previous draft specification of the to what is termed a foreign network), the home agent gets all protocol. In addition, most major router vendors have devel- the packets destined for the mobile node and arranges to deliv oped implementations for Mobile IP er them to the mobile node's current point of attachment The outlook for Mobile IP in the complex Internet mar- Whenever the mobile node moves, it registersits new care- ketplace is far from clear, and some technical problems remain, of address with its home agent. To get a packet to a mobile security being the most important. However, once the securi- node from its home network, the home agent delivers the ty solutions are solid, nomadic users may finally begin to enjoy packet from the home network to the care-of address. The the convenience of seamless untethered roaming and effective further delivery requires that the packet be modified so that application transparency that is the promise of Mobile IP. the care-of address appears as the destination IP address. This modification can be understood as a packet transformation HOW MOBILE IP WORKS or, more specifically, a redirection. When the packet arrives at IP routes packets from a source endpoint to a destination by the care-of address, the reverse transformation is applied so allowing routers to forward packets from incoming network that the packet once again appears to have the mobile node interfaces to outbound interfaces according to routing tables. home address as the destination IP address. When the pack The routing tables typically maintain the next-hop(out- et arrives at the mobile node, addressed to the home address bound interface)information for each destination IP it will be processed properly by TCP or whatever higher level address, according to the number of networks to which that protocol logically receives it from the mobile node's lp (that IEEE INTERNET COMPUTING http://computer.org/internet/JanUaRy.fEbruAry1998

addresses. These addresses are associated with a fixed net￾work location much as a nonmobile phone number is asso￾ciated with a physical jack in a wall. When the packet’s des￾tination is a mobile node, this means that each new point of attachment made by the node is associated with a new net￾work number and, hence, a new IP address, making trans￾parent mobility impossible. Mobile IP (RFC 2002),1 a standard proposed by a work￾ing group within the Internet Engineering Task Force, was designed to solve this problem by allowing the mobile node to use two IP addresses: a fixed home address and a care-of address that changes at each new point of attachment. This article will present the Mobile IP standard in moderate technical detail and point the reader toward a wealth of fur￾ther information.2,3 In addition, readers can go to the side￾bar Mobile IP Web Resources in this issue’s IC Online at http://computer.org/internet/ for a convenient set of hyper￾linked resources. I also describe how Mobile IP will change with IP version 6,4,5 the product of a major effort within the IETF to engineer an eventu￾al replacement for the cur￾rent version of IP.6 Although IPv6 will support mobility to a greater degree than IPv4, it will still need Mobile IP to make mobility transparent to applications and higher level protocols such as TCP. There is a great deal of interest in mobile computing and apparently in Mobile IP as a way to provide for it. A quick Web search for items related to Mobile IP returned over 60,000 hits—impressive even given the notorious lack of selectivity for such procedures. Mobile IP forms the basis either directly or indirectly of many current research efforts and products. The Cellular Digital Packet Data (CDPD),7 for example, has created a widely deployed communications infrastructure based on a previous draft specification of the protocol. In addition, most major router vendors have devel￾oped implementations for Mobile IP. The outlook for Mobile IP in the complex Internet mar￾ketplace is far from clear, and some technical problems remain, security being the most important. However, once the securi￾ty solutions are solid, nomadic users may finally begin to enjoy the convenience of seamless untethered roaming and effective application transparency that is the promise of Mobile IP. HOW MOBILE IP WORKS IP routes packets from a source endpoint to a destination by allowing routers to forward packets from incoming network interfaces to outbound interfaces according to routing tables. The routing tables typically maintain the next-hop (out￾bound interface) information for each destination IP address, according to the number of networks to which that IP address is connected. The network number is derived from the IP address by masking off some of the low-order bits. Thus, the IP address typically carries with it informa￾tion that specifies the IP node’s point of attachment. To maintain existing transport-layer connections (see the sidebar “Nomadicity: How Mobility Will Affect the Protocol Stack” on the next pages) as the mobile node moves from place to place, it must keep its IP address the same. In TCP (which accounts for the overwhelming majority of Internet connec￾tions), connections are indexed by a quadruplet that contains the IP addresses and port numbers of both connection end￾points. Changing any of these four numbers will cause the con￾nection to be disrupted and lost. On the other hand, correct delivery of packets to the mobile node’s current point of attach￾ment depends on the network number contained within the mobile node’s IP address, which changes at new points of attachment. To change the routing requires a new IP address associated with the new point of attachment. Mobile IP has been designed to solve this problem by allowing the mobile node to use two IP addresses (see the sidebar “Mobile Networking Terminology” for definitions of italicized terms). In Mobile IP, the home address is static and is used, for instance, to identify TCP connections. The care-of address changes at each new point of attachment and can be thought of as the mobile node’s topologically significant address; it indicates the network number and thus identifies the mobile node’s point of attachment with respect to the net￾work topology. The home address makes it appear that the mobile node is continually able to receive data on its home net￾work, where Mobile IP requires the existence of a network node known as the home agent. Whenever the mobile node is not attached to its home network (and is therefore attached to what is termed a foreign network), the home agent gets all the packets destined for the mobile node and arranges to deliv￾er them to the mobile node’s current point of attachment. Whenever the mobile node moves, it registers its new care￾of address with its home agent. To get a packet to a mobile node from its home network, the home agent delivers the packet from the home network to the care-of address. The further delivery requires that the packet be modified so that the care-of address appears as the destination IP address. This modification can be understood as a packet transformation or, more specifically, a redirection. When the packet arrives at the care-of address, the reverse transformation is applied so that the packet once again appears to have the mobile node’s home address as the destination IP address. When the pack￾et arrives at the mobile node, addressed to the home address, it will be processed properly by TCP or whatever higher level protocol logically receives it from the mobile node’s IP (that MOBILE IP 59 IEEE INTERNET COMPUTING http://computer.org/internet/ JANUARY • FEBRUARY 1998 . Mobile IP uses two IP addresses: a fixed home address and a care-of address that changes at each new point of attachment

MO B LEc。 M PUTIN C NOMADICITY: HOW MOBILITY WILL AFFECT THE PROTOCOL STACK Mobile IP is a large part, but by no means the only part, of the media typically introduce many other design requirements at story of mobile computing and networking. To see Mobile IP in layer two. In particular, the desire for confidentiality leads to its true place requires an understanding of the relationships the incorporation of encryption techniques, especially for wire- between the various layers of network protocols. Each layer less links. Often, lower bandwidth(compared with wired should present a clear model of operation to the architect. Once media) suggests the use of compression techniques. And, typ- the model is identified, the effects of mobility can be studied in ically, transmitting a signal causes the local receiver to lose relation to it. Nomadicity is the name used by the Cross-industry detection of any other signal because of the great difference Working Team (XIWT) at the Corporation for National in effective power levels between local and remote transmit Research Initiatives(CNRi) to denote an architecture for the ters. Thus, collision-detection techniques, such as those used with entire mobile computing environment Ethernet, must be replaced by less reliable collision-avoidance Figure A is a simplified view of the International Standards measures and careful etiquette Organizations protocol stack as it applies to Internet network- Other distinguishing characteristics of wireless communica- ing. The major goal of Mobile IP protocol design was to handle tions media include the difficulty of establishing a precise range mobility at the network layer and to leave transport and other (cell size)for connectivity to the medium, and the ability for sep- higher layers unaffected, so that the existing routing infra- arate stations to use the media without interference. This latter structure, nonmobile hosts, and current applications would not property of reuse depends upon avoiding interference between be required to change neighboring transmitters, and a great engineering discipline Protocol layer two, the data link layer, is responsible for link has been built up to understand optimal placement of such wire- establishment and maintenance. Thus, physical effects from less equipment as base stations. To reuse the physical wireless mobility are likely to require changes in the layer-two proto- medium to the fullest extent, the cell size should be as small as ols. Changes in position affect the signal-to-interference ratio possible. This means that as demand for wireless communica (SIR). Link layers that adapt forward error correction to SIR can tions increases, cell sizes will decrease, and the frequency with xhibit variable bandwidth but far fewer lost packets. Wireless which mobile computers will switch cells(change their point of attachment to the Internet) will grow correspondingly The Internet Protocol is at layer three, the network layer. Networking Layers Standard Protocols elects routes(determines paths) through a loosely confeder- Applications Http Nfs, Snmp Dns. Telnet ated association of independent network links. IP offers rout- ing from one network to another, in addition to some minor Window Mgr services such as fragmentation and reassembly, and check Sockets summing. Moving from one place to another can be modeled TCP UDP RTP as changing the network nodes point of attachment to the Inter IP ICMP IGMP IPSec. Mobile IP net Supporting mobility at this layer is therefore naturally mod- (IPX, Appletalk eled as changing the routing of datagrams destined for the Data Link IEEE 802* PPP mobile node so that they arrive at the new point of attachment. Network adapter This turns out to be a very convenient choice, and was the option chosen by the Mobile IP working group At the fransport layer, TCP(RFC 793)and other transport Figure A. The Internet networking stack showing protocols attempt to offer a more convenient abstraction for common protocols associated with each layer. data services than the characteristically chaotic stream of data rer 3)processing layer. More information on the abstract until it arrives at the care-of address. Such encapsulation is also ling as a way to perform layer 3 redirection on packets called tunneling which suggests that the packet burrows can be found in Bhagwat, Perkins, and Tripathi through the Internet, bypassing the usual effects of IP routing In Mobile IP the home agent redirects packets from the Mobile IP, then, is best understood as the cooperation of home network to the care-of address by constructing a new Ip three separable mechanisms header that contains the mobile node's care-of address as the destination IP address. This new header then shields or encap- Discovering the care-of address: sulates the original packet, causing the mobile node's home Registering the care-of address address to have no effect on the encapsulated packets routing Tunneling to the care-of address JanUarY.feBruAry1998http://computer.org/internet/ IEEE INTERNET COMPUTING

is, layer 3) processing layer. More information on the abstract modeling as a way to perform layer 3 redirection on packets can be found in Bhagwat, Perkins, and Tripathi.8 In Mobile IP the home agent redirects packets from the home network to the care-of address by constructing a new IP header that contains the mobile node’s care-of address as the destination IP address. This new header then shields or encap￾sulates the original packet, causing the mobile node’s home address to have no effect on the encapsulated packet’s routing until it arrives at the care-of address. Such encapsulation is also called tunneling, which suggests that the packet burrows through the Internet, bypassing the usual effects of IP routing. Mobile IP, then, is best understood as the cooperation of three separable mechanisms: ■ Discovering the care-of address; ■ Registering the care-of address; ■ Tunneling to the care-of address. MOBILE COMPUTING 60 JANUARY • FEBRUARY 1998 http://computer.org/internet/ IEEE INTERNET COMPUTING . Mobile IP is a large part, but by no means the only part, of the story of mobile computing and networking. To see Mobile IP in its true place requires an understanding of the relationships between the various layers of network protocols. Each layer should present a clear model of operation to the architect. Once the model is identified, the effects of mobility can be studied in relation to it. Nomadicity is the name used by the Cross-Industry Working Team (XIWT) at the Corporation for National Research Initiatives (CNRI) to denote an architecture for the entire mobile computing environment.1 Figure A is a simplified view of the International Standards Organization’s protocol stack as it applies to Internet network￾ing. The major goal of Mobile IP protocol design was to handle mobility at the network layer and to leave transport and other higher layers unaffected, so that the existing routing infra￾structure, nonmobile hosts, and current applications would not be required to change. Protocol layer two, the data link layer, is responsible for link establishment and maintenance. Thus, physical effects from mobility are likely to require changes in the layer-two proto￾cols. Changes in position affect the signal-to-interference ratio (SIR). Link layers that adapt forward error correction to SIR can exhibit variable bandwidth but far fewer lost packets. Wireless media typically introduce many other design requirements at layer two. In particular, the desire for confidentiality leads to the incorporation of encryption techniques, especially for wire￾less links. Often, lower bandwidth (compared with wired media) suggests the use of compression techniques. And, typ￾ically, transmitting a signal causes the local receiver to lose detection of any other signal because of the great difference in effective power levels between local and remote transmit￾ters. Thus, collision-detection techniques, such as those used with Ethernet, must be replaced by less reliable collision-avoidance measures and careful etiquette. Other distinguishing characteristics of wireless communica￾tions media include the difficulty of establishing a precise range (cell size) for connectivity to the medium, and the ability for sep￾arate stations to use the media without interference. This latter property of reuse depends upon avoiding interference between neighboring transmitters, and a great engineering discipline has been built up to understand optimal placement of such wire￾less equipment as base stations. To reuse the physical wireless medium to the fullest extent, the cell size should be as small as possible. This means that as demand for wireless communica￾tions increases, cell sizes will decrease, and the frequency with which mobile computers will switch cells (change their point of attachment to the Internet) will grow correspondingly. The Internet Protocol is at layer three, the network layer. IP selects routes (determines paths) through a loosely confeder￾ated association of independent network links. IP offers rout￾ing from one network to another, in addition to some minor services such as fragmentation and reassembly, and check￾summing. Moving from one place to another can be modeled as changing the network node’s point of attachment to the Inter￾net. Supporting mobility at this layer is therefore naturally mod￾eled as changing the routing of datagrams destined for the mobile node so that they arrive at the new point of attachment. This turns out to be a very convenient choice, and was the option chosen by the Mobile IP working group. At the transport layer, TCP (RFC 793)2 and other transport protocols attempt to offer a more convenient abstraction for data services than the characteristically chaotic stream of data NOMADICITY: HOW MOBILITY WILL AFFECT THE PROTOCOL STACK Networking Layers Standard Protocols Applications HTTP, NFS, SNMP, DNS, Telnet, FTP, ... Window Mgr Sockets Transport TCP, UDP, RTP Network IP, ICMP, IGMP, IPSec, Mobile IP ... (IPX, Appletalk) Data Link IEEE 802.*, PPP Physical Network adapter Figure A. The Internet networking stack showing common protocols associated with each layer

M O emanating from IP. The vagaries and time dependencies of can easily affect the applications desired operation. The simplest routers and Internet congestion often cause datagrams to be example is the need for Web applications to adjust their presen- delivered out of order, duplicated, or even dropped entirely tation of graphical data depending on the available end-to-end before reaching their destination. TCP attempts to solve those bandwidth. Today that bandwidth is largely constrained by the problems, but offers little help in supplying a steady (constant link conditions at the endpoints and the congestion status of infra- bandwidth)stream of data, or in delivering data within spec- structure connectivity. Mobile computers introduce more variabl- fied time bounds. Over time, TCP has been modified to treat ity into this mix and thus exacerbate the growing need for multi- dropped packets as an indication of network congestion, and media applications to detect and act on dynamic connection therefore to throttle transmissions as soon as a lost packet is parameters, such as link bandwidth, error rate, and round-trip detected (by managing sequence numbers). This is the wrong times. Other logical parameters, such as cost and security, may strategy when packets are corrupted by transmission over a eventually exhibit similar dynamic behavior and further compli- noisy wireless channel, because for such packets immediate cate application response to connection status information retransmission is much better than delayed retransmission. Lastly, a word should be said about the granularity of pro Ways to change this behavior are still under debate tocol response to node movement. Today's typical user must be At the top layer are the application protocols Depending content with portable computing, which requires reinitializing on the transport model employed, application protocols are and reestablishing connections at each new point of attach- largely freed from much of the drudgery of error correction, ment to the Internet. However, acceptance of this mode of oper transmission, flow control, and the like. However, mobility ation may well evaporate if the reinitialization process has to be creates new needs at the application layer, which require performed a lot more frequently Left unchecked, the expected additional protocol support: automatic configuration, service decreases in cell sizes will require exactly that in the future. The discovery, link awareness, and environment awareness existing methods typify portable network computing, which These protocol support mechanisms form a set of middle- means establishing the availability of network computing when are services. For example, a mobile computer might need to one arrives at a new point of attachment but being unable to be reconfigured differently at each different point of attach- continue previous computing activities. The point of Mobile IP. ment. Among other things, a new DNS server, IP address, link DHCP and similar protocols is to provide completely automat MTU, and default router may be required. These configuration ic, noninteractive reconnection to those activities items are usually thought of as being worked out at setup time for desktop systems, but for mobile computers no single answer REFERENCES can be sufficient. Recent deployment of the Dynamic Host Con- 1. Corporation for National Research Initiatives. XIWT: Cross-Industry figuration Protocol(RFCs 2131, 2132)", goes some way WorkingTeam,1994,http://www.cnri.restonva.us3000/xiwt/ toward resolving configuration difficulties, but is not the whole public. html. answer. Discovering services can be modeled as a requirement 2. "Transmission Control Protocol, "J.B. Postel, ed, RFC 793, Sept. 1981 for automatic configuration, but is more naturally useful when 3. W. Stevens, "TCP Sow Start, Congestion Avoidance, Fast Retransmit, services are located upon demand and according to the needs and Fast Recovery Algorithms, "RFC 2001, Jan 1997 of application protocols. This need is just now being met by the 4. R. Droms, ' Dynamic Host Configuration Protocol, "RFC 2131, Mar Service Location Protocol(RFC 2165).6 1997, ftp: //ds internic. net/rfc/rfc2131. txt. One of the more challenging aspects of architecting such 5. S. Alexander and R Droms, "DHCP Options and BOOTP Vendor Exten- iddleware lies in offering applications the opportunity to detect sions, " RFC 2132. Mar 1997 he state of the physical link, which changes dynamically and 6. J. Veizades, et al., "Service Location Protocol, "RFC 2165, July 1997 Discovering the Care-of Address tisements are extended to also contain the needed care-of The Mobile IP discoveryprocess has been built on top of an address, they are known as agent advertisements Home existing standard protocol, Router Advertisement, specified agents and foreign agents typically broadcast agent adver- in RFC 1256. Mobile ip discovery does not modify the tisements at regular intervals(for example, once a second or original fields of existing router advertisements but simply once every few seconds ). If a mobile node needs to get extends them to associate mobility functions. Thus, a router care-of address and does not wish to wait for the periodic advertisement can carry information about default routers, advertisement, the mobile node can broadcast or multicast just as before, and in addition carry further information solicitation that will be answered by any foreign agent about one or more care-of addresses. When the router adver- home agent that receives it IEEE INTERNET COMPUTING http://computer.org/internet/JanUaRy.fEbruAry1998

Discovering the Care-of Address The Mobile IP discovery process has been built on top of an existing standard protocol, Router Advertisement, specified in RFC 1256.9 Mobile IP discovery does not modify the original fields of existing router advertisements but simply extends them to associate mobility functions. Thus, a router advertisement can carry information about default routers, just as before, and in addition carry further information about one or more care-of addresses. When the router adver￾tisements are extended to also contain the needed care-of address, they are known as agent advertisements. Home agents and foreign agents typically broadcast agent adver￾tisements at regular intervals (for example, once a second or once every few seconds). If a mobile node needs to get a care-of address and does not wish to wait for the periodic advertisement, the mobile node can broadcast or multicast a solicitation that will be answered by any foreign agent or home agent that receives it. MOBILE IP 61 IEEE INTERNET COMPUTING http://computer.org/internet/ JANUARY • FEBRUARY 1998 . emanating from IP. The vagaries and time dependencies of routers and Internet congestion often cause datagrams to be delivered out of order, duplicated, or even dropped entirely before reaching their destination. TCP attempts to solve those problems, but offers little help in supplying a steady (constant bandwidth) stream of data, or in delivering data within spec￾ified time bounds. Over time, TCP has been modified to treat dropped packets as an indication of network congestion, and therefore to throttle transmissions as soon as a lost packet is detected (by managing sequence numbers).3 This is the wrong strategy when packets are corrupted by transmission over a noisy wireless channel, because for such packets immediate retransmission is much better than delayed retransmission. Ways to change this behavior are still under debate. At the top layer are the application protocols. Depending on the transport model employed, application protocols are largely freed from much of the drudgery of error correction, retransmission, flow control, and the like. However, mobility creates new needs at the application layer, which require additional protocol support: automatic configuration, service discovery, link awareness, and environment awareness. These protocol support mechanisms form a set of middle￾ware services. For example, a mobile computer might need to be reconfigured differently at each different point of attach￾ment. Among other things, a new DNS server, IP address, link MTU, and default router may be required. These configuration items are usually thought of as being worked out at setup time for desktop systems, but for mobile computers no single answer can be sufficient. Recent deployment of the Dynamic Host Con￾figuration Protocol (RFCs 2131, 2132)4,5 goes some way toward resolving configuration difficulties, but is not the whole answer. Discovering services can be modeled as a requirement for automatic configuration, but is more naturally useful when services are located upon demand and according to the needs of application protocols. This need is just now being met by the Service Location Protocol (RFC 2165).6 One of the more challenging aspects of architecting such middleware lies in offering applications the opportunity to detect the state of the physical link, which changes dynamically and can easily affect the application’s desired operation. The simplest example is the need for Web applications to adjust their presen￾tation of graphical data depending on the available end-to-end bandwidth. Today that bandwidth is largely constrained by the link conditions at the endpoints and the congestion status of infra￾structure connectivity. Mobile computers introduce more variabil￾ity into this mix and thus exacerbate the growing need for multi￾media applications to detect and act on dynamic connection parameters, such as link bandwidth, error rate, and round-trip times. Other logical parameters, such as cost and security, may eventually exhibit similar dynamic behavior and further compli￾cate application response to connection status information. Lastly, a word should be said about the granularity of pro￾tocol response to node movement. Today’s typical user must be content with portable computing, which requires reinitializing and reestablishing connections at each new point of attach￾ment to the Internet. However, acceptance of this mode of oper￾ation may well evaporate if the reinitialization process has to be performed a lot more frequently. Left unchecked, the expected decreases in cell sizes will require exactly that in the future. The existing methods typify portable network computing, which means establishing the availability of network computing when one arrives at a new point of attachment but being unable to continue previous computing activities. The point of Mobile IP, DHCP and similar protocols is to provide completely automat￾ic, noninteractive reconnection to those activities. REFERENCES 1. Corporation for National Research Initiatives. XIWT: Cross-Industry Working Team, 1994, http://www.cnri.reston.va.us:3000/XIWT/ public.html. 2. “Transmission Control Protocol,” J. B. Postel, ed., RFC 793, Sept. 1981. 3. W. Stevens, “TCP Sow Start, Congestion Avoidance, Fast Retransmit, and Fast Recovery Algorithms,” RFC 2001, Jan. 1997. 4. R. Droms, “Dynamic Host Configuration Protocol,” RFC 2131, Mar. 1997, ftp://ds.internic.net/rfc/rfc2131.txt. 5. S. Alexander and R. Droms, “DHCP Options and BOOTP Vendor Exten￾sions,” RFC 2132, Mar. 1997. 6. J. Veizades, et al., “Service Location Protocol,” RFC 2165, July 1997

MO B LEc。 M PUTIN C mobile nodes network interface. In this sit uation, the mobile node should begin to FA advertises service hunt for a new care-of address, or possibly se a care-of address known from advertise ments it is still receiving. The mobile node MH requests service may choose to wait for another advertise- FA relays request to HA ment if it has not received any recently adver tised care-of addresses, or it may send an nt Registering the Care-of Address Once a mobile node has a care-of address FA relays status to MH HA accepts or denies its home agent must find out about it. Fig- ure 1 shows the registration process defined by Mobile IP for this purpose. The process begins when the mobile node, possibly with the assistance of a foreign agent, sends a reg Figure 1. Registration operations in Mobile IP. FA is foreign agent, HA is istration request with the care-of address home address and MH is mobile host information. When the home agent receives this request, it(typically) adds the necessary information to its routing table, approves Home agents use agent advertisements to make them- the request, and sends a registration reply back to the mobile selves known, even if they do not offer any care-of address- node. Although the home agent is not required by the es. However, it is not possible to associate preferences to the Mobile IP protocol to handle registration requests by updat various care-of addresses in the router advertisement, as is ing entries in its routing table, doing so offers a natural the case with default routers. The IETF working group was implementation strategy, and all implementations I know concerned that dynamic preference values might destabilize of take this approach the operation of Mobile IP. Because no one could defend static preference assignments except for backup mobility Authentication. Registration requests contain parameters agents, which do not help distribute the routing load, the and flags that characterize the tunnel through which the group eventually decided not to use the preference assign- home agent will deliver packets to the care-of address. Tun ments with the care-of address list nels can be constructed in various ways, described briefly in Thus, an agent advertisement performs the following the next section. 0. I When a home agent accepts the functions. request, it begins to associate the home address of the mobile node with the care-of address and maintains this associa allows for the detection of mobility agents tion until the registration lifetime expires. The triplet that a lists one or more available care-of addresses contains the home address, care-of address, and registration forms the mobile node about special features provided lifetime is called a binding for the mobile node. a registra by foreign agents, for example, alternative encapsulation tion request can be considered a binding update sent by the a lets mobile nodes determine the network number and mobile node techniques a binding update is an example of a remote redirect, status of their link to the internet and because it is sent remotely to the home agent to affect the lets the mobile node know whether the agent is a home home agents routing table. This view of registration makes agent, a foreign agent, or both, and therefore whether it the need for authentication very clear. 2 The home agent is on its home network or a foreign network must be certain registration was originated by the mobile node and not by some other malicious node pretending to Mobile nodes use router solicitations as defined in rfc 1256 be the mobile node. a malicious node could cause the home to detect any change in the set of mobility agents available at agent to alter its routing table with erroneous care-of address the current point of attachment. (In Mobile IP this is then information, and the mobile node would be unreachable to termed agent solicitation. If advertisements are no longer all incoming communications from the Internet. detectable from a foreign agent that previously had offered a The need to authenticate registration information has care-of address to the mobile node, the mobile node should played a major role in determining the acceptable design presume that foreign agent is no longer within range of the parameters for Mobile IP. Each mobile node and home agent JanUarY.feBruAry1998http://computer.org/internet/ IEEE INTERNET COMPUTING

Home agents use agent advertisements to make them￾selves known, even if they do not offer any care-of address￾es. However, it is not possible to associate preferences to the various care-of addresses in the router advertisement, as is the case with default routers. The IETF working group was concerned that dynamic preference values might destabilize the operation of Mobile IP. Because no one could defend static preference assignments except for backup mobility agents, which do not help distribute the routing load, the group eventually decided not to use the preference assign￾ments with the care-of address list. Thus, an agent advertisement performs the following functions: ■ allows for the detection of mobility agents; ■ lists one or more available care-of addresses; ■ informs the mobile node about special features provided by foreign agents, for example, alternative encapsulation techniques; ■ lets mobile nodes determine the network number and status of their link to the Internet; and ■ lets the mobile node know whether the agent is a home agent, a foreign agent, or both, and therefore whether it is on its home network or a foreign network. Mobile nodes use router solicitations as defined in RFC 1256 to detect any change in the set of mobility agents available at the current point of attachment. (In Mobile IP this is then termed agent solicitation.) If advertisements are no longer detectable from a foreign agent that previously had offered a care-of address to the mobile node, the mobile node should presume that foreign agent is no longer within range of the mobile node’s network interface. In this sit￾uation, the mobile node should begin to hunt for a new care-of address, or possibly use a care-of address known from advertise￾ments it is still receiving. The mobile node may choose to wait for another advertise￾ment if it has not received any recently adver￾tised care-of addresses, or it may send an agent solicitation. Registering the Care-of Address Once a mobile node has a care-of address, its home agent must find out about it. Fig￾ure 1 shows the registration process defined by Mobile IP for this purpose. The process begins when the mobile node, possibly with the assistance of a foreign agent, sends a reg￾istration request with the care-of address information. When the home agent receives this request, it (typically) adds the necessary information to its routing table, approves the request, and sends a registration reply back to the mobile node. Although the home agent is not required by the Mobile IP protocol to handle registration requests by updat￾ing entries in its routing table, doing so offers a natural implementation strategy, and all implementations I know of take this approach. Authentication. Registration requests contain parameters and flags that characterize the tunnel through which the home agent will deliver packets to the care-of address. Tun￾nels can be constructed in various ways, described briefly in the next section.10,11 When a home agent accepts the request, it begins to associate the home address of the mobile node with the care-of address, and maintains this associa￾tion until the registration lifetime expires. The triplet that contains the home address, care-of address, and registration lifetime is called a binding for the mobile node. A registra￾tion request can be considered a binding update sent by the mobile node. A binding update is an example of a remote redirect, because it is sent remotely to the home agent to affect the home agent’s routing table. This view of registration makes the need for authentication very clear.12 The home agent must be certain registration was originated by the mobile node and not by some other malicious node pretending to be the mobile node. A malicious node could cause the home agent to alter its routing table with erroneous care-of address information, and the mobile node would be unreachable to all incoming communications from the Internet. The need to authenticate registration information has played a major role in determining the acceptable design parameters for Mobile IP. Each mobile node and home agent MOBILE COMPUTING 62 JANUARY • FEBRUARY 1998 http://computer.org/internet/ IEEE INTERNET COMPUTING . FA FA HA FA FA advertises service MH requests service FA relays request to HA FA relays status to MH HA accepts or denies Figure 1. Registration operations in Mobile IP. FA is foreign agent, HA is home address, and MH is mobile host

M O must share a security association and be able to use Message Src dest proto Digest 5(RFC 1321)with 128-bit keys to create unforge- X MH I? Pa avload Encapsulatated diagram able digital signatures for registration requests. The signa- ture is computed by performing MD5s one-way hash algo Src Dest Proto Src Dest Proto rithm over all the data within the registration message header Home HA CoM 4 or 55 X MH? Payload and the extensions that precede the signature To secure the registration request, each request must con- tain unique data so that two different registrations will in practical terms never have the same MD5 hash. Otherwise the protocol would be susceptible to replay attacks in which a malicious node could record valid registrations for later Foreign replay, effectively disrupting the ability of the home agent agent to tunnel to the current care-of address of the mobile node at that later time. To ensure this does not happen, Mobile Src Dest proto IP includes within the registration message a special identi- LX? Payload fication field that changes with every new registration. The obile mode exact semantics of the identification field depend on sever- Figure 2. Tunneling operations in Mobile IP al details, which are described at greater length in the pro tocol specification. Briefly, there are two main ways to make the identification field unique. As Figure 1 shows, in Mobile IP foreign agents are mostly One is to use a timestamp; then each new registration will passive, relaying registration requests and replies back and have a later timestamp and thus differ from previous regis- forth between the home agent and the mobile node, doing trations. The other is to cause the identification to be a mostly what they are told. The foreign agent also decapsulates pseudorandom number, with enough bits of randomness, it traffic from the home agent and forwards it to the mobile is highly unlikely that two independently chosen values for node. Note that foreign agents do not have to authen the identification field will be the same. When randomness is ticate themselves to the mobile node or home agent. a bogus sed, Mobile IP defines a method that protects both the reg. foreign agent could impersonate a real foreign agent simply istration request and reply from replay, and calls for 32 bits by following protocol and offering agent advertisements to of randomness in the identification field. If the mobile node the mobile node. The bogus agent could, for instance, then and the home agent get too far out of synchronization for the refuse to forward decapsulated packets to the mobile node use of timestamps, or if they lose track of the expected ran- when they were received. However, the result is no worse than dom numbers, the home agent will reject the registration if any node were tricked into using the wrong default router request and include information to allow resynchronization which is possible using unauthenticated router advertisements within the reply. Using random numbers instead of time- as specified in RFC 1256 stamps avoids problems stemming from attacks on the nTP protocol that might cause the mobile node to lose time syn- Automatic home agent discovery. When the mobile chronization with the home agent or to issue authenticated node cannot contact its home agent, Mobile Ip has a mech- registration requests for some future time that could be used anism that lets the mobile node try to register with another by a malicious node to subvert a future registration. unknown home agent on its home network. This method The identification field is also used by the foreign agent to of automatic home agent discovery works by using a broad match pending registration requests to registration replies cast IP address instead of the home agents IP address as the when they arrive at the home agent and to subsequently be target for the registration request. When the broadcast pack- able to relay the reply to the mobile node. The foreign agent et gets to the home network, other home agents on the net- also stores other information for pending registrations, work will send a rejection to the mobile node; however, their including the mobile node's home address, the mobile nodes rejection notice will contain their address for the mobile Media Access Layer(MAC)address, the source port num- node to use in a freshly attempted registration message. Note ber for the registration request from the mobile node, the that the broadcast is not an Internet-wide broadcast, but a registration lifetime proposed by the mobile node, and the directed broadcast that reaches only IP nodes on the home home agent's address. The foreign agent can limit registra- network tion lifetimes to a configurable value that it puts into its agent advertisements. The home agent can reduce the reg. Tunneling to the Care-of Address istration lifetime, which it includes as part of the registra- Figure 2 shows the tunneling operations in Mobile IP. The tion reply, but it can never increase it. default encapsulation mechanism that must be supported IEEE INTERNET COMPUTING http://computer.org/internet/JanUaRy.fEbruAry1998

must share a security association and be able to use Message Digest 5 (RFC 1321) with 128-bit keys to create unforge￾able digital signatures for registration requests.13 The signa￾ture is computed by performing MD5’s one-way hash algo￾rithm over all the data within the registration message header and the extensions that precede the signature. To secure the registration request, each request must con￾tain unique data so that two different registrations will in practical terms never have the same MD5 hash. Otherwise, the protocol would be susceptible to replay attacks, in which a malicious node could record valid registrations for later replay, effectively disrupting the ability of the home agent to tunnel to the current care-of address of the mobile node at that later time. To ensure this does not happen, Mobile IP includes within the registration message a special identi￾fication field that changes with every new registration. The exact semantics of the identification field depend on sever￾al details, which are described at greater length in the pro￾tocol specification.1 Briefly, there are two main ways to make the identification field unique. One is to use a timestamp; then each new registration will have a later timestamp and thus differ from previous regis￾trations. The other is to cause the identification to be a pseudorandom number; with enough bits of randomness, it is highly unlikely that two independently chosen values for the identification field will be the same. When randomness is used, Mobile IP defines a method that protects both the reg￾istration request and reply from replay, and calls for 32 bits of randomness in the identification field. If the mobile node and the home agent get too far out of synchronization for the use of timestamps, or if they lose track of the expected ran￾dom numbers, the home agent will reject the registration request and include information to allow resynchronization within the reply. Using random numbers instead of time￾stamps avoids problems stemming from attacks on the NTP protocol that might cause the mobile node to lose time syn￾chronization with the home agent or to issue authenticated registration requests for some future time that could be used by a malicious node to subvert a future registration. The identification field is also used by the foreign agent to match pending registration requests to registration replies when they arrive at the home agent and to subsequently be able to relay the reply to the mobile node. The foreign agent also stores other information for pending registrations, including the mobile node’s home address, the mobile node’s Media Access Layer (MAC) address, the source port num￾ber for the registration request from the mobile node, the registration lifetime proposed by the mobile node, and the home agent’s address. The foreign agent can limit registra￾tion lifetimes to a configurable value that it puts into its agent advertisements. The home agent can reduce the reg￾istration lifetime, which it includes as part of the registra￾tion reply, but it can never increase it. As Figure 1 shows, in Mobile IP foreign agents are mostly passive, relaying registration requests and replies back and forth between the home agent and the mobile node, doing mostly what they are told. The foreign agent also decapsulates traffic from the home agent and forwards it to the mobile node. Note that foreign agents do not have to authen￾ticate themselves to the mobile node or home agent. A bogus foreign agent could impersonate a real foreign agent simply by following protocol and offering agent advertisements to the mobile node. The bogus agent could, for instance, then refuse to forward decapsulated packets to the mobile node when they were received. However, the result is no worse than if any node were tricked into using the wrong default router, which is possible using unauthenticated router advertisements as specified in RFC 1256.9 Automatic home agent discovery. When the mobile node cannot contact its home agent, Mobile IP has a mech￾anism that lets the mobile node try to register with another unknown home agent on its home network. This method of automatic home agent discovery works by using a broad￾cast IP address instead of the home agent’s IP address as the target for the registration request. When the broadcast pack￾et gets to the home network, other home agents on the net￾work will send a rejection to the mobile node; however, their rejection notice will contain their address for the mobile node to use in a freshly attempted registration message. Note that the broadcast is not an Internet-wide broadcast, but a directed broadcast that reaches only IP nodes on the home network. Tunneling to the Care-of Address Figure 2 shows the tunneling operations in Mobile IP. The default encapsulation mechanism that must be supported MOBILE IP 63 IEEE INTERNET COMPUTING http://computer.org/internet/ JANUARY • FEBRUARY 1998 . Src Dest Proto X MH ? Payload Src Dest Proto X MH ? Payload Src Dest Proto Src Dest Proto HA COM 4 or 55 X MH ? Payload Home agent Foreign agent Encapsulatated diagram Mobile mode Figure 2. Tunneling operations in Mobile IP

MO B LEc。 M PUTIN C by all mobility agents using Mobile IP is IP-within-IPi0 Route Optimization Using IP-within-IP the home agent, the tunnel source inserts IPv6 mobility borrows heavily from the route optimization a new IP header, or tunnel header, in front of the IP header ideas specified for IPv4, 20 particularly the idea of delivering of any datagram addressed to the mobile node's home binding updates directly to correspondent nodes. When it address The new tunnel header uses the mobile node's care- knows the mobile node s current care-of address a corre- of address as the destination IP address, or tunnel destina- spondent node can deliver packets directly to the mobile on. The tunnel source IP address is the home agent, and nodes home address without any assistance from the home the tunnel header uses 4 as the higher level protocol num- agent. Route optimization is likely to dramatically improve ating that the next protocol header is again an IP performance for IPv6 mobile nodes. It is realistic to require header. In IP-within-IP the entire original IP header is pre- this extra functionality of all IPv6 nodes for two reasons served as the first part of the payload of the tunnel header. First, on a practical level, IPv6 standards documents are still Therefore, to recover the original packet, the foreign agent at an early stage of standardization, so it is possible to place merely has to eliminate the tunnel header and deliver the additional requirements on IPv6 nodes. Second, processing rest to the mobile node binding updates can be implemented as a fairly simple mod Figure 2 shows that sometimes the tunnel header uses ification to IPv6's use of the destination cache protocol number 55 as the inner header. This happens when the home agent uses minimal encapsulation instead of IP- Security within-IP Processing for the minimal encapsulation header One of the biggest differences between IPv6 and IPv4 is is slightly more complicated than that for IP-within-IP that all IPv6 nodes are expected to implement strong because some of the information from the tunnel header is authentication and encryption features21, 22 to improve combined with the information in the inner minimal encap- Intern t security. This affords a major simplification for sulation header to reconstitute the original IP header On IPv6 mobility support, since all authentication procedures the other hand header overhead is reduced can be assumed to exist when needed and do not have to be specified in the Mobile IPv6 protocol. Even with the CHANGES WITH IP VERSION 6 security features in IPv6, however, the current working How will Mobile IP change when IP version 6 is adopted? group draft for IPv6 mobility support specifies the use of IPv6 inchudes many features for streamlining mobility sup- authentication procedures as infrequently as possible. The port that are missing in IP version 4 (current version), reasons for this are twofold. First, good authentication including Stateless Address Autoconfiguration and Neigh- comes at the cost of performance and so should be required bor Discovery. 15 IPv6 also only occasionally. Second, attempts to drastically sim- questions about the avail plity the process of renu- As proposed by the Mobile IP working ability of Internet-wide key cal to the future routability of group, mobility support in IPv6 follows the resolved at this time the Internet. 16 Because the design for Mobile IPv4, using encapsulation number of mobile computers to deliver packets from the home network to Source Routing accessing the Internet will the mobile point of attachment In contrast to the way in likely increase, efficient sup- which route optimization is port for mobility will make a pecified in IPv4, in IPv6 decisive difference in the Internets future performance. This, correspondent nodes do not tunnel packets to mobile nodes along with the growing importance of the Internet and the Instead, they use IPv6 routing headers, which implement a Web, indicates the need to pay attention to supporting variation of IPv4's source routing option. a number of early mobility proposals for supporting mobility in IPv4 specified a similar Mobility Support in IPv6, 8 as proposed by the Mobile use of source routing opti 23. 2 but two main probler IP working group, follows the design for Mobile IPv4. It precluded their use retains the ideas of a home network, home agent, and the use of encapsulation to deliver packets from the home net- IPv4 source routing options require the receiver of work to the mobile node's current point of attachment. source-routed packets to follow the reversed path to the While discovery of a care-of address is still required, a mobile sender back along the indicated intermediate nodes. This node can configure its a care-of address by using Stateless means that malicious nodes using source routes from ddress Autoconfiguration and Neighbor Discovery. Thus, remote locations within the Internet could impersonate IPv6-within-IPv6 tunneling is also already specified 196. other nodes, a problem exacerbated by the lack of foreign agents are not required to support mobility in IP authentication protocols JanUarY.feBruAry1998http://computer.org/internet/ IEEE INTERNET COMPUTING

by all mobility agents using Mobile IP is IP-within-IP.10 Using IP-within-IP, the home agent, the tunnel source, inserts a new IP header, or tunnel header, in front of the IP header of any datagram addressed to the mobile node’s home address. The new tunnel header uses the mobile node’s care￾of address as the destination IP address, or tunnel destina￾tion. The tunnel source IP address is the home agent, and the tunnel header uses 4 as the higher level protocol num￾ber, indicating that the next protocol header is again an IP header. In IP-within-IP the entire original IP header is pre￾served as the first part of the payload of the tunnel header. Therefore, to recover the original packet, the foreign agent merely has to eliminate the tunnel header and deliver the rest to the mobile node. Figure 2 shows that sometimes the tunnel header uses protocol number 55 as the inner header. This happens when the home agent uses minimal encapsulation11 instead of IP￾within-IP. Processing for the minimal encapsulation header is slightly more complicated than that for IP-within-IP, because some of the information from the tunnel header is combined with the information in the inner minimal encap￾sulation header to reconstitute the original IP header. On the other hand, header overhead is reduced. CHANGES WITH IP VERSION 6 How will Mobile IP change when IP version 6 is adopted? IPv6 includes many features for streamlining mobility sup￾port that are missing in IP version 4 (current version), including Stateless Address Autoconfiguration14 and Neigh￾bor Discovery.15 IPv6 also attempts to drastically sim￾plify the process of renum￾bering, which could be criti￾cal to the future routability of the Internet.16 Because the number of mobile computers accessing the Internet will likely increase, efficient sup￾port for mobility will make a decisive difference in the Internet’s future performance. This, along with the growing importance of the Internet and the Web, indicates the need to pay attention to supporting mobility.17 Mobility Support in IPv6,18 as proposed by the Mobile IP working group, follows the design for Mobile IPv4. It retains the ideas of a home network, home agent, and the use of encapsulation to deliver packets from the home net￾work to the mobile node’s current point of attachment. While discovery of a care-of address is still required, a mobile node can configure its a care-of address by using Stateless Address Autoconfiguration and Neighbor Discovery. Thus, foreign agents are not required to support mobility in IPv6. IPv6-within-IPv6 tunneling is also already specified.19 Route Optimization IPv6 mobility borrows heavily from the route optimization ideas specified for IPv4,20 particularly the idea of delivering binding updates directly to correspondent nodes. When it knows the mobile node’s current care-of address, a corre￾spondent node can deliver packets directly to the mobile node’s home address without any assistance from the home agent. Route optimization is likely to dramatically improve performance for IPv6 mobile nodes. It is realistic to require this extra functionality of all IPv6 nodes for two reasons. First, on a practical level, IPv6 standards documents are still at an early stage of standardization, so it is possible to place additional requirements on IPv6 nodes. Second, processing binding updates can be implemented as a fairly simple mod￾ification to IPv6’s use of the destination cache.15 Security One of the biggest differences between IPv6 and IPv4 is that all IPv6 nodes are expected to implement strong authentication and encryption features21,22 to improve Internet security. This affords a major simplification for IPv6 mobility support, since all authentication procedures can be assumed to exist when needed and do not have to be specified in the Mobile IPv6 protocol. Even with the security features in IPv6, however, the current working group draft for IPv6 mobility support specifies the use of authentication procedures as infrequently as possible. The reasons for this are twofold. First, good authentication comes at the cost of performance and so should be required only occasionally. Second, questions about the avail￾ability of Internet-wide key management are far from resolved at this time. Source Routing In contrast to the way in which route optimization is specified in IPv4, in IPv6 correspondent nodes do not tunnel packets to mobile nodes. Instead, they use IPv6 routing headers, which implement a variation of IPv4’s source routing option. A number of early proposals for supporting mobility in IPv4 specified a similar use of source routing options,23,24 but two main problems precluded their use: ■ IPv4 source routing options require the receiver of source-routed packets to follow the reversed path to the sender back along the indicated intermediate nodes. This means that malicious nodes using source routes from remote locations within the Internet could impersonate other nodes, a problem exacerbated by the lack of authentication protocols. MOBILE COMPUTING 64 JANUARY • FEBRUARY 1998 http://computer.org/internet/ IEEE INTERNET COMPUTING . As proposed by the Mobile IP working group, mobility support in IPv6 follows the design for Mobile IPv4, using encapsulation to deliver packets from the home network to the mobile point of attachment

M O MOBILE NETWORKING TERMINOLOGY Agent advertisement. The procedure by which a mobility agent Home address. The IP address assigned to the mobile node. becomes known to the mobile node making it logically appear attached to its home network Agent discovery. The process by which a mobile node can Home agent. A node on the home network that effectively obtain the IP address of a home agent or foreign agent, causes the mobile node to be reachable at its home address depending upon whether the mobile node is home or away even when the mobile node is not attached to its home net from home. Agent discovery occurs when a mobile node work receives an agent advertisement, either as a result of periodic Home network. The network at which the mobile node seems broadcast or in response to a solicitation reachable, to the rest of the Internet, by virtue of its assigned Automatic home agent discovery. The process by which a IP address. mobile node can obtain the IP address of a home agent on its Minimal encapsulation. A variant encapsulation technique home network, involving the transmission of a registration specified in RFC 2003 that temporarily alters the structure of request to the subnet broadcast address of its home network. the original IP header, but uses fewer bytes for tunneling pack Binding. The triplet of numbers that contains the mobile nodes ets to the care-of address than the default method(IP-within home address, its care-of address, and the registration life- IP)uses time-how long the mobility agents may use the binding. Mobile node A node that, as part of normal use, changes its Binding update. The message that supplies a new binding to an point of attachment to the Internet entity that needs to know the new care-of address for a mobile Mobility agent. A node(typically, a router) that offers support node. The binding update contains the mobile node's home services to mobile nodes. A mobility agent can be either a address, new care-of address, and a new registration lifetime. home agent or a foreign agent. Care-of address. An IP address at the mobile nodes current Nomadicity. The full range of network technology being point of attachment to the Internet, when the mobile node is not designed to come to the assistance of the mobile(or nomadic attached to the home network. A collocated care-of address computer user, not limited to network-layer protocols is a care-of address assigned to one of the mobile node's net- Redirection. A message that is intended to cause a change in work interfaces, instead of one being offered by a foreign the routing behavior of the node receiving it. Registration. The process by which the mobile node informs Correspondent node. A node that sends or receives a packet to the home agent about its current care-of address a mobile node; the correspondent node may be another mobile Remote redirection. A redirect sent from a source not present node or a nonmobile Internet node on the local network. The source can be located anywhere in Discovery. In this article, short for agent discovery the global Internet and may have malicious intent and be Encapsulation. The process of incorporating an original IP untraceable packet (less any preceding fields such as a MAC header) inside Replay attacks. A security violation whereby a malicious enti- another IP packet, making the fields within the original IP head- ty attempts to imitate a transaction recorded during a previous er temporarily lose their effect and valid transaction between two protocol entities. Both pro Foreign agent. A mobility agent on the foreign network that tocol entities have to be aware that the subsequent identical can assist the mobile node in receiving datagrams delivered to traffic streams may no longer be valid. Since the previous trans- the care-of address action was valid, the algorithms for detecting replay attacks Foreign network. The network to which the mobile node is need to incorporate data that can never be reproduced in any attached when it is not attached to its home network, and on correct subsequent transaction which the care-of address is reachable from the rest of the Route optimization. A process that enables the delivery of Internet packets directly to the care-of address from a corresponder Fully qualified domain name(FQDN). An Internet node's node without having to detour through the home network FQDN is its complete domain name as defined by the Domain Source routing. A routing technique that causes some or all Name System(DNS). A node can be known locally by a rela- intermediate routing points to be represented directly in the tive domain name that is a substring of its FQDN, but such a data packet to be forwarded. This is in contrast to the typical relative name cannot be resolved correctly by Internet nodes situation in which intermediate routers rely on acquired rout- outside of the part of the domain name hierarchy indicated by ing state information to forward incoming packets the relative name. The fully qualified name can be resolved Tunneling. The same as encapsulation, but with additional con from anywhere in the Internet, subject to access control and notations about changing the effects of Internet routing on the routability of the resolution request original IP packet IEEE INTERNET COMPUTING http://computer.org/internet/JanUaRy.fEbruAry1998

MOBILE IP 65 IEEE INTERNET COMPUTING http://computer.org/internet/ JANUARY • FEBRUARY 1998 . Agent advertisement. The procedure by which a mobility agent becomes known to the mobile node. Agent discovery. The process by which a mobile node can obtain the IP address of a home agent or foreign agent, depending upon whether the mobile node is home or away from home. Agent discovery occurs when a mobile node receives an agent advertisement, either as a result of periodic broadcast or in response to a solicitation. Automatic home agent discovery. The process by which a mobile node can obtain the IP address of a home agent on its home network, involving the transmission of a registration request to the subnet broadcast address of its home network. Binding. The triplet of numbers that contains the mobile node’s home address, its care-of address, and the registration life￾time—how long the mobility agents may use the binding. Binding update. The message that supplies a new binding to an entity that needs to know the new care-of address for a mobile node. The binding update contains the mobile node’s home address, new care-of address, and a new registration lifetime. Care-of address. An IP address at the mobile node’s current point of attachment to the Internet, when the mobile node is not attached to the home network. A collocated care-of address is a care-of address assigned to one of the mobile node’s net￾work interfaces, instead of one being offered by a foreign agent. Correspondent node. A node that sends or receives a packet to a mobile node; the correspondent node may be another mobile node or a nonmobile Internet node. Discovery. In this article, short for agent discovery. Encapsulation. The process of incorporating an original IP packet (less any preceding fields such as a MAC header) inside another IP packet, making the fields within the original IP head￾er temporarily lose their effect. Foreign agent. A mobility agent on the foreign network that can assist the mobile node in receiving datagrams delivered to the care-of address. Foreign network. The network to which the mobile node is attached when it is not attached to its home network, and on which the care-of address is reachable from the rest of the Internet. Fully qualified domain name (FQDN). An Internet node’s FQDN is its complete domain name as defined by the Domain Name System (DNS). A node can be known locally by a rela￾tive domain name that is a substring of its FQDN, but such a relative name cannot be resolved correctly by Internet nodes outside of the part of the domain name hierarchy indicated by the relative name. The fully qualified name can be resolved from anywhere in the Internet, subject to access control and routability of the resolution request. Home address. The IP address assigned to the mobile node, making it logically appear attached to its home network. Home agent. A node on the home network that effectively causes the mobile node to be reachable at its home address even when the mobile node is not attached to its home net￾work. Home network. The network at which the mobile node seems reachable, to the rest of the Internet, by virtue of its assigned IP address. Minimal encapsulation. A variant encapsulation technique specified in RFC 2003 that temporarily alters the structure of the original IP header, but uses fewer bytes for tunneling pack￾ets to the care-of address than the default method (IP-within￾IP) uses. Mobile node. A node that, as part of normal use, changes its point of attachment to the Internet. Mobility agent. A node (typically, a router) that offers support services to mobile nodes. A mobility agent can be either a home agent or a foreign agent. Nomadicity. The full range of network technology being designed to come to the assistance of the mobile (or nomadic) computer user, not limited to network-layer protocols. Redirection. A message that is intended to cause a change in the routing behavior of the node receiving it. Registration. The process by which the mobile node informs the home agent about its current care-of address. Remote redirection. A redirect sent from a source not present on the local network. The source can be located anywhere in the global Internet and may have malicious intent and be untraceable. Replay attacks. A security violation whereby a malicious enti￾ty attempts to imitate a transaction recorded during a previous and valid transaction between two protocol entities. Both pro￾tocol entities have to be aware that the subsequent identical traffic streams may no longer be valid. Since the previous trans￾action was valid, the algorithms for detecting replay attacks need to incorporate data that can never be reproduced in any correct subsequent transaction. Route optimization. A process that enables the delivery of packets directly to the care-of address from a correspondent node without having to detour through the home network. Source routing. A routing technique that causes some or all intermediate routing points to be represented directly in the data packet to be forwarded. This is in contrast to the typical situation in which intermediate routers rely on acquired rout￾ing state information to forward incoming packets. Tunneling. The same as encapsulation, but with additional con￾notations about changing the effects of Internet routing on the original IP packet. MOBILE NETWORKING TERMINOLOGY

MO B LEc。 M PUTIN C is a further point of contrast to route optimization in IPv4 that, MOBILE IP WEB RESOURCES in IPv6 mobility support, the mobile node delivers binding You can view the Mobile IP working groups charter and all updates to correspondent nodes instead of to the home agent. InternetdraftsandRfCdocumentsathttp://www.ietf.org/InIpV6,keymanagementbetweenthemobilenodeandcor- html. charters/mobileip-charter html You can also join the gen- respondent node is more likely to be available eral mail is for the Mobile IP working group by sending mail to Other features supported by IPv6 mobility include majordomo@smallworks com and induding the line"subscribe mobile-ip"in the body of the message. Archives of the mail list coexistence with Internet ingress filtering: are available at ftp: //fp. smallworks com/mobile-ip archive. smooth handoffs, which in Mobile IPv4 is specified for New members of the general discussion list should read the foreign agents as part of route optimization materialsfoundathttp://www.ietforg/overview.htmlandrenumberingofhomenetworks;and http://www.ietf.org/tao.html I automatic home agent discovery. Further major Web resources for Mobile IP, including var- ious freeware implementations, can be found at the following ONGOING WORK AND OPEN QUESTIONS Problems Facing Mobile IP The most pressing outstanding problem facing Mobile IP is The CMU Monarch Project that of security, but other technical as well as practical obst Protocols for Adaptive Mobile and Wireless Networking cles to deployment exist. 26 Work is also continuing to refine http://www.monarch.cs.cmu.edu/ and extend the protocol within the academic and commer cial communities and within the IETE. This section surveys Portland State Secure Mobile Networking Project the state of implementation of Mobile IP and speculates on http://www.cs.pdx.edu/research/smn/ a possible timetable for deployment Mobile IP at the National University of Singal Routing inefficiencies. The base Mobile IP specification http://mip.ee.nus.sg/ has the effect of introducing a tunnel into the routing path followed by packets sent by the correspondent node to the State University of New York, Binghamton mobile node Packets from the mobile node, on the other Linux-Mobile IP hand, can go directly to the correspondent node with no http://anchor.cs.binghamton.edu/-mobileip/ tunneling required. This asymmetry is captured by the term triangle routing, where a single leg of the triangle goes from Stanford's Operating Systems and Networking Group the mobile node to the correspondent node, and the hom MosquitoNet Mobile IP agent forms the third vertex controlling the path taken by http://mosquitonet.stanfordedu/software/mip.html data from the correspondent node to the mobile node. Tri- angle routing is alleviated by use of techniques in the route BBN Technologies Mobile IP Security page optimization draft, 20 but doing so requires changes in the http://www.net-tech.bbncom/moips/moips-index.html correspondent nodes that will take a long time to deploy for IPv4. It is hoped that triangle routing will not be a factor for lpv6 mobili Existing routers exhibit terrible performance when han- dling source routes. Consequently, the results of deploy. Security issues. a great deal of attention is being focused ing other protocols that use source routes have not been on making Mobile IP coexist with the security features com- ing into use within the Internet. Firewalls, in particular cause difficulty for Mobile IP because they block all classes However, the objections to the use of source routes do not of incoming packets that do not meet specified criteria apply to IPv6, because IPv6s more careful specification elim- Enterprise firewalls are typically configured to block pack inates the need for source-route reversal and lets routers ignore ets from entering via the Internet that appear to emanate options that do not need their attention. Consequently, cor- from internal computers. Although this permits manage respondent nodes can use routing headers without penalty. ment of internal Internet nodes without great attention to This allows the mobile node to easily determine when a cor- security, it presents difficulties for mobile nodes wishing to respondent node does not have the right care-of address. Pack- communicate with other nodes within their home enterprise ets delivered by encapsulation instead of by source routes in a networks. Such communications, originating from the routing header must have been sent by correspondent nodes mobile node, carry the mobile node's home address, and that need to receive binding updates from the mobile node. It would thus be blocked by the firewall. JanUarY.feBruAry1998http://computer.org/internet/ IEEE INTERNET COMPUTING

■ Existing routers exhibit terrible performance when han￾dling source routes. Consequently, the results of deploy￾ing other protocols that use source routes have not been favorable. However, the objections to the use of source routes do not apply to IPv6, because IPv6’s more careful specification elim￾inates the need for source-route reversal and lets routers ignore options that do not need their attention. Consequently, cor￾respondent nodes can use routing headers without penalty. This allows the mobile node to easily determine when a cor￾respondent node does not have the right care-of address. Pack￾ets delivered by encapsulation instead of by source routes in a routing header must have been sent by correspondent nodes that need to receive binding updates from the mobile node. It is a further point of contrast to route optimization in IPv4 that, in IPv6 mobility support, the mobile node delivers binding updates to correspondent nodes instead of to the home agent. In IPv6, key management between the mobile node and cor￾respondent node is more likely to be available. Other features supported by IPv6 mobility include ■ coexistence with Internet ingress filtering;25 ■ smooth handoffs, which in Mobile IPv4 is specified for foreign agents as part of route optimization; ■ renumbering of home networks; and ■ automatic home agent discovery. ONGOING WORK AND OPEN QUESTIONS Problems Facing Mobile IP The most pressing outstanding problem facing Mobile IP is that of security, but other technical as well as practical obsta￾cles to deployment exist.26 Work is also continuing to refine and extend the protocol within the academic and commer￾cial communities and within the IETF. This section surveys the state of implementation of Mobile IP and speculates on a possible timetable for deployment. Routing inefficiencies. The base Mobile IP specification has the effect of introducing a tunnel into the routing path followed by packets sent by the correspondent node to the mobile node. Packets from the mobile node, on the other hand, can go directly to the correspondent node with no tunneling required. This asymmetry is captured by the term triangle routing, where a single leg of the triangle goes from the mobile node to the correspondent node, and the home agent forms the third vertex controlling the path taken by data from the correspondent node to the mobile node. Tri￾angle routing is alleviated by use of techniques in the route optimization draft,20 but doing so requires changes in the correspondent nodes that will take a long time to deploy for IPv4. It is hoped that triangle routing will not be a factor for IPv6 mobility. Security issues. A great deal of attention is being focused on making Mobile IP coexist with the security features com￾ing into use within the Internet. Firewalls,27 in particular, cause difficulty for Mobile IP because they block all classes of incoming packets that do not meet specified criteria. Enterprise firewalls are typically configured to block pack￾ets from entering via the Internet that appear to emanate from internal computers. Although this permits manage￾ment of internal Internet nodes without great attention to security, it presents difficulties for mobile nodes wishing to communicate with other nodes within their home enterprise networks. Such communications, originating from the mobile node, carry the mobile node’s home address, and would thus be blocked by the firewall. MOBILE COMPUTING 66 JANUARY • FEBRUARY 1998 http://computer.org/internet/ IEEE INTERNET COMPUTING . MOBILE IP WEB RESOURCES You can view the Mobile IP working group’s charter and all Internet drafts and RFC documents at http://www.ietf.org/ html.charters/mobileip-charter.html. You can also join the gen￾eral mail list for the Mobile IP working group by sending mail to majordomo@smallworks.com and including the line “subscribe mobile-ip” in the body of the message. Archives of the mail list are available at ftp://ftp.smallworks.com/mobile-ip.archive. New members of the general discussion list should read the materials found at http://www.ietf.org/overview.html and http://www.ietf.org/tao.html. Further major Web resources for Mobile IP, including var￾ious freeware implementations, can be found at the following sites: The CMU Monarch Project Protocols for Adaptive Mobile and Wireless Networking • http://www.monarch.cs.cmu.edu/ Portland State Secure Mobile Networking Project • http://www.cs.pdx.edu/research/SMN/ Mobile IP at the National University of Singapore • http://mip.ee.nus.sg/ State University of New York, Binghamton Linux-Mobile IP • http://anchor.cs.binghamton.edu/~mobileip/ Stanford’s Operating Systems and Networking Group MosquitoNet Mobile IP • http://mosquitonet.stanford.edu/software/mip.html BBN Technologies Mobile IP Security page • http://www.net-tech.bbn.com/moips/moips-index.html

M O Mobile IP can be viewed as a protocol for establishing secure tunnels. Gupta and glass have proposed a firewall tra- HISTORY OF THE MOBILE IP versal solution. 8 Efforts along these lines are also being made WORKING GROUP at bbn as part of the MOIPS (Managed Objects for IP Mobil- The Mobile IP Working Group of the Internet Engineering ity Support)2 project to extend Mobile IP operation across task Force(ETF)had its origin in BOF (Birds of a Feather) firewalls, even when multiple security domains are involved. sessions held at the Atlanta(uly 1991),Santa Fe(November 1991), and San Diego(March 1992)IETF meetings. In June Ingress filtering. Complications are also presented by ingress 1992 Steve Deering, chair of the working group, submitted filtering2operations. Many border routers discard packets a proposed charter for a formal Working Group to the IETE. coming from within the enterprise if the packets do not con- and, following a revision of the charter, the Working Group tain a source IP address configured for one of the enterprises was officially formed in June 30,1992.An IPv4 Mobile Host internal networks. Because mobile nodes would otherwise use Protocol was submitted to the Internet Engineering Steering their home address as the source IP address of the packets they Group(ESG)as a proposed standard in 1996.An IPv6 pro- transmit, this presents difficulty. Solutions to this problem in tocol will be submitted to the IESG later this year Mobile IPv4 typically involve tunneling outgoing packets from the care-of address, but then the difficulty is how to find a suitable target for the tunneled packet from the mobile node. over, it is possible that such an alternative IP address would The only universally agreed on possibility is the home agent, offer a shorter routing path if, for instance, the address were but that target introduces yet another serious routing anom- apparently located on a physical link nearer to the mobile of the Internet. Montenegro has proposed the use of reverse care-of address itself. Finally, many communications are tunnels to the home agent to counter the restriction imposed short-lived and depend on neither the actual identity of the by ingress filtering. Mobile IPv6 also offers a solution in the mobile node nor its FQDN, and thus do not take advan home address destination option tage of the simplicity afforded by use of the mobile node's home address. These issues surrounding the mobile node's User perceptions of reliability. The design of Mobile IP selection of an appropriate long-term(or not-so-long-term) is founded on the premise that connections based on TCP address for use in establishing connections are complex and should survive cell changes. However, opinion is not unan- are far from being resolved. mous on the need for this feature. Many people believe that computer communications to laptop computers are suffi- Slow growth in the wireless LAN market. Mobile IP ntly bursty that there is no need to increase the reliabili has been engineered as a solution for wireless lan location ty of the connections supporting the communications. The management and communications, but the wireless LAN analogy is made to fetching Web pages by selecting the market has been slow to develop. It is difficult to make gen appropriate URLS. If a transfer fails, people are used to try- eral statements about the reasons for this slow development, ing again. This is tantamount to making the user responsible but with the recent ratification of the IEEE 802. 11 MAC for the retransmission protocol and depends for its accept- protocol, 2 wireless LaNs may become more popular. More- ability on a widespread perception that computers and the over, the bandwidth for wireless devices has been constant- Internet cannot be trusted to do things right the first time. ly improving, so that radio and infrared devices on the mar- Naturally, such assumptions are strongly distasteful to many ket today offer multimegabyte- per-second data rates Faster Internet protocol engineers, myself included. Nevertheless, wireless access over standardized Mac layers could be a he fact that products exhibiting this model are currently major catalyst for growth of this market. economically viable cannot be denied. Hopefully in the near future better engineering will counter this perception and Competition from other protocols. Mobile IP may well increase the demand for Internet reliability face competition from alternative tunneling protocols such as PPTP and L2TP. 4 These other protocols, based on PPP Issues in IP addressing. Mobile IP creates the perception offer at least portability to mobile computers. Although I that the mobile node is always attached to its home network. believe portable operation will ultimately not be a long-term This forms the basis for the reachability of the mobile node solution, it may look quite attractive in the short term in the at an IP address that can be conventionally associated with absence of full Mobile IP deployment. If these alternative its fully qualified domain name(FQDn). If the FQDN is methods are made widely available, it is unclear if the use of associated with one or more other IP addresses, perhaps Mobile IP will be displaced or instead made more immedi- dynamically, then those alternative IP addresses may deserve ately desirable as people experience the convenience of equal standing with the mobile node's home address More- mobile computing. In the future it is also possible that IEEE INTERNET COMPUTING http://computer.org/internet/JanUaRy.fEbruAry1998

Mobile IP can be viewed as a protocol for establishing secure tunnels. Gupta and Glass have proposed a firewall tra￾versal solution.28 Efforts along these lines are also being made at BBN as part of the MOIPS (Managed Objects for IP Mobil￾ity Support)29 project to extend Mobile IP operation across firewalls, even when multiple security domains are involved. Ingress filtering.Complications are also presented by ingress filtering25 operations. Many border routers discard packets coming from within the enterprise if the packets do not con￾tain a source IP address configured for one of the enterprise’s internal networks. Because mobile nodes would otherwise use their home address as the source IP address of the packets they transmit, this presents difficulty. Solutions to this problem in Mobile IPv4 typically involve tunneling outgoing packets from the care-of address, but then the difficulty is how to find a suitable target for the tunneled packet from the mobile node. The only universally agreed on possibility is the home agent, but that target introduces yet another serious routing anom￾aly for communications between the mobile node and the rest of the Internet. Montenegro has proposed the use of reverse tunnels to the home agent to counter the restriction imposed by ingress filtering.30 Mobile IPv6 also offers a solution in the home address destination option.18 User perceptions of reliability. The design of Mobile IP is founded on the premise that connections based on TCP should survive cell changes. However, opinion is not unan￾imous on the need for this feature. Many people believe that computer communications to laptop computers are suffi￾ciently bursty that there is no need to increase the reliabili￾ty of the connections supporting the communications. The analogy is made to fetching Web pages by selecting the appropriate URLs. If a transfer fails, people are used to try￾ing again. This is tantamount to making the user responsible for the retransmission protocol and depends for its accept￾ability on a widespread perception that computers and the Internet cannot be trusted to do things right the first time. Naturally, such assumptions are strongly distasteful to many Internet protocol engineers, myself included. Nevertheless, the fact that products exhibiting this model are currently economically viable cannot be denied. Hopefully in the near future better engineering will counter this perception and increase the demand for Internet reliability. Issues in IP addressing. Mobile IP creates the perception that the mobile node is always attached to its home network. This forms the basis for the reachability of the mobile node at an IP address that can be conventionally associated with its fully qualified domain name (FQDN).31 If the FQDN is associated with one or more other IP addresses, perhaps dynamically, then those alternative IP addresses may deserve equal standing with the mobile node’s home address. More￾over, it is possible that such an alternative IP address would offer a shorter routing path if, for instance, the address were apparently located on a physical link nearer to the mobile node’s care-of address, or if the alternative address were the care-of address itself. Finally, many communications are short-lived and depend on neither the actual identity of the mobile node nor its FQDN, and thus do not take advan￾tage of the simplicity afforded by use of the mobile node’s home address. These issues surrounding the mobile node’s selection of an appropriate long-term (or not-so-long-term) address for use in establishing connections are complex and are far from being resolved. Slow growth in the wireless LAN market. Mobile IP has been engineered as a solution for wireless LAN location management and communications, but the wireless LAN market has been slow to develop. It is difficult to make gen￾eral statements about the reasons for this slow development, but with the recent ratification of the IEEE 802.11 MAC protocol,32 wireless LANs may become more popular. More￾over, the bandwidth for wireless devices has been constant￾ly improving, so that radio and infrared devices on the mar￾ket today offer multimegabyte-per-second data rates. Faster wireless access over standardized MAC layers could be a major catalyst for growth of this market. Competition from other protocols. Mobile IP may well face competition from alternative tunneling protocols such as PPTP33 and L2TP.34These other protocols, based on PPP, offer at least portability to mobile computers. Although I believe portable operation will ultimately not be a long-term solution, it may look quite attractive in the short term in the absence of full Mobile IP deployment. If these alternative methods are made widely available, it is unclear if the use of Mobile IP will be displaced or instead made more immedi￾ately desirable as people experience the convenience of mobile computing. In the future, it is also possible that MOBILE IP 67 IEEE INTERNET COMPUTING http://computer.org/internet/ JANUARY • FEBRUARY 1998 . HISTORY OF THE MOBILE IP WORKING GROUP The Mobile IP Working Group of the Internet Engineering task Force (IETF) had its origin in BOF (Birds of a Feather) sessions held at the Atlanta (July 1991), Santa Fe (November 1991), and San Diego (March 1992) IETF meetings. In June 1992 Steve Deering, chair of the working group, submitted a proposed charter for a formal Working Group to the IETF, and, following a revision of the charter, the Working Group was officially formed in June 30, 1992. An IPv4 Mobile Host Protocol was submitted to the Internet Engineering Steering Group (IESG) as a proposed standard in 1996. An IPv6 pro￾tocol will be submitted to the IESG later this year