idnits 2.17.1 draft-moskowitz-rfc4423-bis-00.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** You're using the IETF Trust Provisions' Section 6.b License Notice from 12 Sep 2009 rather than the newer Notice from 28 Dec 2009. (See https://trustee.ietf.org/license-info/) Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == Line 1060 has weird spacing: '...ncement to th...' == The document seems to contain a disclaimer for pre-RFC5378 work, but was first submitted on or after 10 November 2008. The disclaimer is usually necessary only for documents that revise or obsolete older RFCs, and that take significant amounts of text from those RFCs. If you can contact all authors of the source material and they are willing to grant the BCP78 rights to the IETF Trust, you can and should remove the disclaimer. Otherwise, the disclaimer is needed and you can ignore this comment. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (December 4, 2009) is 5250 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Unused Reference: 'RFC4025' is defined on line 1043, but no explicit reference was found in the text ** Obsolete normative reference: RFC 5202 (Obsoleted by RFC 7402) ** Obsolete normative reference: RFC 5204 (Obsoleted by RFC 8004) ** Obsolete normative reference: RFC 5205 (Obsoleted by RFC 8005) -- Obsolete informational reference (is this intentional?): RFC 2535 (Obsoleted by RFC 4033, RFC 4034, RFC 4035) -- Obsolete informational reference (is this intentional?): RFC 2766 (Obsoleted by RFC 4966) -- Obsolete informational reference (is this intentional?): RFC 4306 (Obsoleted by RFC 5996) Summary: 4 errors (**), 0 flaws (~~), 4 warnings (==), 4 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group R. Moskowitz 3 Internet-Draft ICSAlabs 4 Obsoletes: 4423 (if approved) P. Nikander 5 Intended status: Standards Track Ericsson Research Nomadic Lab 6 Expires: June 7, 2010 December 4, 2009 8 Host Identity Protocol Architecture 9 draft-moskowitz-rfc4423-bis-00 11 Abstract 13 This memo describes a new namespace, the Host Identity namespace, and 14 a new protocol layer, the Host Identity Protocol, between the 15 internetworking and transport layers. Herein are presented the 16 basics of the current namespaces, their strengths and weaknesses, and 17 how a new namespace will add completeness to them. The roles of this 18 new namespace in the protocols are defined. 20 This document obsoletes RFC 4423 and addresses the concerns raised by 21 the IESG, particularly that of crypto agility. It also incorporates 22 lessons learned from the implementations of RFC 5201. 24 Status of this Memo 26 This Internet-Draft is submitted to IETF in full conformance with the 27 provisions of BCP 78 and BCP 79. 29 Internet-Drafts are working documents of the Internet Engineering 30 Task Force (IETF), its areas, and its working groups. Note that 31 other groups may also distribute working documents as Internet- 32 Drafts. 34 Internet-Drafts are draft documents valid for a maximum of six months 35 and may be updated, replaced, or obsoleted by other documents at any 36 time. It is inappropriate to use Internet-Drafts as reference 37 material or to cite them other than as "work in progress." 39 The list of current Internet-Drafts can be accessed at 40 http://www.ietf.org/ietf/1id-abstracts.txt. 42 The list of Internet-Draft Shadow Directories can be accessed at 43 http://www.ietf.org/shadow.html. 45 This Internet-Draft will expire on June 7, 2010. 47 Copyright Notice 48 Copyright (c) 2009 IETF Trust and the persons identified as the 49 document authors. All rights reserved. 51 This document is subject to BCP 78 and the IETF Trust's Legal 52 Provisions Relating to IETF Documents 53 (http://trustee.ietf.org/license-info) in effect on the date of 54 publication of this document. Please review these documents 55 carefully, as they describe your rights and restrictions with respect 56 to this document. Code Components extracted from this document must 57 include Simplified BSD License text as described in Section 4.e of 58 the Trust Legal Provisions and are provided without warranty as 59 described in the BSD License. 61 This document may contain material from IETF Documents or IETF 62 Contributions published or made publicly available before November 63 10, 2008. The person(s) controlling the copyright in some of this 64 material may not have granted the IETF Trust the right to allow 65 modifications of such material outside the IETF Standards Process. 66 Without obtaining an adequate license from the person(s) controlling 67 the copyright in such materials, this document may not be modified 68 outside the IETF Standards Process, and derivative works of it may 69 not be created outside the IETF Standards Process, except to format 70 it for publication as an RFC or to translate it into languages other 71 than English. 73 Table of Contents 75 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 4 76 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . 4 77 2.1. Terms common to other documents . . . . . . . . . . . . . . 5 78 2.2. Terms specific to this and other HIP documents . . . . . . . 5 79 3. Background . . . . . . . . . . . . . . . . . . . . . . . . . 7 80 3.1. A desire for a namespace for computing platforms . . . . . . 7 81 4. Host Identity namespace . . . . . . . . . . . . . . . . . . 9 82 4.1. Host Identifiers . . . . . . . . . . . . . . . . . . . . . . 10 83 4.2. Storing Host Identifiers in DNS . . . . . . . . . . . . . . 10 84 4.3. Host Identity Tag (HIT) . . . . . . . . . . . . . . . . . . 11 85 4.4. Local Scope Identifier (LSI) . . . . . . . . . . . . . . . . 11 86 5. New stack architecture . . . . . . . . . . . . . . . . . . . 11 87 5.1. Transport associations and end-points . . . . . . . . . . . 12 88 6. End-host mobility and multi-homing . . . . . . . . . . . . . 13 89 6.1. Rendezvous mechanism . . . . . . . . . . . . . . . . . . . . 13 90 6.2. Protection against flooding attacks . . . . . . . . . . . . 14 91 7. HIP and IPsec . . . . . . . . . . . . . . . . . . . . . . . 14 92 8. HIP and NATs . . . . . . . . . . . . . . . . . . . . . . . . 15 93 8.1. HIP and TCP checksums . . . . . . . . . . . . . . . . . . . 16 94 9. Multicast . . . . . . . . . . . . . . . . . . . . . . . . . 16 95 10. HIP policies . . . . . . . . . . . . . . . . . . . . . . . . 16 96 11. Benefits of HIP . . . . . . . . . . . . . . . . . . . . . . 17 97 11.1. HIP's answers to NSRG questions . . . . . . . . . . . . . . 18 98 12. Security considerations . . . . . . . . . . . . . . . . . . 20 99 12.1. HITs used in ACLs . . . . . . . . . . . . . . . . . . . . . 21 100 12.2. Non-security considerations . . . . . . . . . . . . . . . . 22 101 13. IANA considerations . . . . . . . . . . . . . . . . . . . . 22 102 14. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . 23 103 15. References . . . . . . . . . . . . . . . . . . . . . . . . . 23 104 15.1. Normative References . . . . . . . . . . . . . . . . . . . . 23 105 15.2. Informative references . . . . . . . . . . . . . . . . . . . 23 106 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . 24 108 1. Introduction 110 The Internet has two important global namespaces: Internet Protocol 111 (IP) addresses and Domain Name Service (DNS) names. These two 112 namespaces have a set of features and abstractions that have powered 113 the Internet to what it is today. They also have a number of 114 weaknesses. Basically, since they are all we have, we try and do too 115 much with them. Semantic overloading and functionality extensions 116 have greatly complicated these namespaces. 118 The proposed Host Identity namespace fills an important gap between 119 the IP and DNS namespaces. The Host Identity namespace consists of 120 Host Identifiers (HI). A Host Identifier is cryptographic in its 121 nature; it is the public key of an asymmetric key-pair. Each host 122 will have at least one Host Identity, but it will typically have more 123 than one. Each Host Identity uniquely identifies a single host, 124 i.e., no two hosts have the same Host Identity. The Host Identity, 125 and the corresponding Host Identifier, can either be public (e.g. 126 published in the DNS), or unpublished. Client systems will tend to 127 have both public and unpublished Identities. 129 There is a subtle but important difference between Host Identities 130 and Host Identifiers. An Identity refers to the abstract entity that 131 is identified. An Identifier, on the other hand, refers to the 132 concrete bit pattern that is used in the identification process. 134 Although the Host Identifiers could be used in many authentication 135 systems, such as IKEv2 [RFC4306], the presented architecture 136 introduces a new protocol, called the Host Identity Protocol (HIP), 137 and a cryptographic exchange, called the HIP base exchange; see also 138 Section 7. The HIP protocols under development provide for limited 139 forms of trust between systems, enhance mobility, multi-homing and 140 dynamic IP renumbering, aid in protocol translation / transition, and 141 reduce certain types of denial-of-service (DoS) attacks. 143 When HIP is used, the actual payload traffic between two HIP hosts is 144 typically, but not necessarily, protected with IPsec. The Host 145 Identities are used to create the needed IPsec Security Associations 146 (SAs) and to authenticate the hosts. When IPsec is used, the actual 147 payload IP packets do not differ in any way from standard IPsec 148 protected IP packets. 150 2. Terminology 151 2.1. Terms common to other documents 153 +---------------+---------------------------------------------------+ 154 | Term | Explanation | 155 +---------------+---------------------------------------------------+ 156 | Public key | The public key of an asymmetric cryptographic key | 157 | | pair. Used as a publicly known identifier for | 158 | | cryptographic identity authentication. | 159 | | | 160 | Private key | The private or secret key of an asymmetric | 161 | | cryptographic key pair. Assumed to be known only | 162 | | to the party identified by the corresponding | 163 | | public key. Used by the identified party to | 164 | | authenticate its identity to other parties. | 165 | | | 166 | Public key | An asymmetric cryptographic key pair consisting | 167 | pair | of public and private keys. For example, | 168 | | Rivest-Shamir-Adelman (RSA) and Digital Signature | 169 | | Algorithm (DSA) key pairs are such key pairs. | 170 | | | 171 | End-point | A communicating entity. For historical reasons, | 172 | | the term 'computing platform' is used in this | 173 | | document as a (rough) synonym for end-point. | 174 +---------------+---------------------------------------------------+ 176 2.2. Terms specific to this and other HIP documents 178 It should be noted that many of the terms defined herein are 179 tautologous, self-referential or defined through circular reference 180 to other terms. This is due to the succinct nature of the 181 definitions. See the text elsewhere in this document for more 182 elaborate explanations. 184 +---------------+---------------------------------------------------+ 185 | Term | Explanation | 186 +---------------+---------------------------------------------------+ 187 | Computing | An entity capable of communicating and computing, | 188 | platform | for example, a computer. See the definition of | 189 | | 'End-point', above. | 190 | | | 191 | HIP base | A cryptographic protocol; see also Section 7. | 192 | exchange | | 193 | | | 194 | HIP packet | An IP packet that carries a 'Host Identity | 195 | | Protocol' message. | 196 | | | 197 | Host Identity | An abstract concept assigned to a 'computing | 198 | | platform'. See 'Host Identifier', below. | 199 | | | 200 | Host Identity | A name space formed by all possible Host | 201 | namespace | Identifiers. | 202 | | | 203 | Host Identity | A protocol used to carry and authenticate Host | 204 | Protocol | Identifiers and other information. | 205 | | | 206 | Host Identity | A 128-bit datum created by taking a cryptographic | 207 | Tag | hash over a Host Identifier. | 208 | | | 209 | Host | A public key used as a name for a Host Identity. | 210 | Identifier | | 211 | | | 212 | Local Scope | A 32-bit datum denoting a Host Identity. | 213 | Identifier | | 214 | | | 215 | Public Host | A published or publicly known Host Identfier used | 216 | Identifier | as a public name for a Host Identity, and the | 217 | and Identity | corresponding Identity. | 218 | | | 219 | Unpublished | A Host Identifier that is not placed in any | 220 | Host | public directory, and the corresponding Host | 221 | Identifier | Identity. Unpublished Host Identities are | 222 | and Identity | typically short lived in nature, being often | 223 | | replaced and possibly used just once. | 224 | | | 225 | Rendezvous | A mechanism used to locate mobile hosts based on | 226 | Mechanism | their HIT. | 227 +---------------+---------------------------------------------------+ 229 3. Background 231 The Internet is built from three principal components: computing 232 platforms (end-points), packet transport (i.e., internetworking) 233 infrastructure, and services (applications). The Internet exists to 234 service two principal components: people and robotic services 235 (silicon based people, if you will). All these components need to be 236 named in order to interact in a scalable manner. Here we concentrate 237 on naming computing platforms and packet transport elements. 239 There are two principal namespaces in use in the Internet for these 240 components: IP numbers, and Domain Names. Domain Names provide 241 hierarchically assigned names for some computing platforms and some 242 services. Each hierarchy is delegated from the level above; there is 243 no anonymity in Domain Names. Email, HTTP, and SIP addresses all 244 reference Domain Names. 246 IP numbers are a confounding of two namespaces, the names of a host's 247 networking interfaces and the names of the locations ('confounding' 248 is a term used in statistics to discuss metrics that are merged into 249 one with a gain in indexing, but a loss in informational value). The 250 names of locations should be understood as denoting routing direction 251 vectors, i.e., information that is used to deliver packets to their 252 destinations. 254 IP numbers name networking interfaces, and typically only when the 255 interface is connected to the network. Originally, IP numbers had 256 long-term significance. Today, the vast number of interfaces use 257 ephemeral and/or non-unique IP numbers. That is, every time an 258 interface is connected to the network, it is assigned an IP number. 260 In the current Internet, the transport layers are coupled to the IP 261 addresses. Neither can evolve separately from the other. IPng 262 deliberations were strongly shaped by the decision that a 263 corresponding TCPng would not be created. 265 There are three critical deficiencies with the current namespaces. 266 Firstly, dynamic readdressing cannot be directly managed. Secondly, 267 anonymity is not provided in a consistent, trustable manner. 268 Finally, authentication for systems and datagrams is not provided. 269 All of these deficiencies arise because computing platforms are not 270 well named with the current namespaces. 272 3.1. A desire for a namespace for computing platforms 274 An independent namespace for computing platforms could be used in 275 end-to-end operations independent of the evolution of the 276 internetworking layer and across the many internetworking layers. 278 This could support rapid readdressing of the internetworking layer 279 because of mobility, rehoming, or renumbering. 281 If the namespace for computing platforms is based on public-key 282 cryptography, it can also provide authentication services. If this 283 namespace is locally created without requiring registration, it can 284 provide anonymity. 286 Such a namespace (for computing platforms) and the names in it should 287 have the following characteristics: 289 o The namespace should be applied to the IP 'kernel'. The IP kernel 290 is the 'component' between applications and the packet transport 291 infrastructure. 293 o The namespace should fully decouple the internetworking layer from 294 the higher layers. The names should replace all occurrences of IP 295 addresses within applications (like in the Transport Control 296 Block, TCB). This may require changes to the current APIs. In 297 the long run, it is probable that some new APIs are needed. 299 o The introduction of the namespace should not mandate any 300 administrative infrastructure. Deployment must come from the 301 bottom up, in a pairwise deployment. 303 o The names should have a fixed length representation, for easy 304 inclusion in datagram headers and existing programming interfaces 305 (e.g the TCB). 307 o Using the namespace should be affordable when used in protocols. 308 This is primarily a packet size issue. There is also a 309 computational concern in affordability. 311 o Name collisions should be avoided as much as possible. The 312 mathematics of the birthday paradox can be used to estimate the 313 chance of a collision in a given population and hash space. In 314 general, for a random hash space of size n bits, we would expect 315 to obtain a collision after approximately 1.2*sqrt(2**n) hashes 316 were obtained. For 64 bits, this number is roughly 4 billion. A 317 hash size of 64 bits may be too small to avoid collisions in a 318 large population; for example, there is a 1% chance of collision 319 in a population of 640M. For 100 bits (or more), we would not 320 expect a collision until approximately 2**50 (1 quadrillion) 321 hashes were generated. 323 o The names should have a localized abstraction so that it can be 324 used in existing protocols and APIs. 326 o It must be possible to create names locally. This can provide 327 anonymity at the cost of making resolvability very difficult. 329 * Sometimes the names may contain a delegation component. This 330 is the cost of resolvability. 332 o The namespace should provide authentication services. 334 o The names should be long lived, but replaceable at any time. This 335 impacts access control lists; short lifetimes will tend to result 336 in tedious list maintenance or require a namespace infrastructure 337 for central control of access lists. 339 In this document, a new namespace approaching these ideas is called 340 the Host Identity namespace. Using Host Identities requires its own 341 protocol layer, the Host Identity Protocol, between the 342 internetworking and transport layers. The names are based on public- 343 key cryptography to supply authentication services. Properly 344 designed, it can deliver all of the above stated requirements. 346 4. Host Identity namespace 348 A name in the Host Identity namespace, a Host Identifier (HI), 349 represents a statistically globally unique name for naming any system 350 with an IP stack. This identity is normally associated with, but not 351 limited to, an IP stack. A system can have multiple identities, some 352 'well known', some unpublished or 'anonymous'. A system may self- 353 assert its own identity, or may use a third-party authenticator like 354 DNSSEC [RFC2535], PGP, or X.509 to 'notarize' the identity assertion. 355 It is expected that the Host Identifiers will initially be 356 authenticated with DNSSEC and that all implementations will support 357 DNSSEC as a minimal baseline. 359 In theory, any name that can claim to be 'statistically globally 360 unique' may serve as a Host Identifier. However, in the authors' 361 opinion, a public key of a 'public key pair' makes the best Host 362 Identifier. As will be specified in the Host Identity Protocol 363 specification, a public-key-based HI can authenticate the HIP packets 364 and protect them for man-in-the-middle attacks. Since authenticated 365 datagrams are mandatory to provide much of HIP's denial-of-service 366 protection, the Diffie-Hellman exchange in HIP has to be 367 authenticated. Thus, only public-key HI and authenticated HIP 368 messages are supported in practice. In this document, the non- 369 cryptographic forms of HI and HIP are presented to complete the 370 theory of HI, but they should not be implemented as they could 371 produce worse denial-of-service attacks than the Internet has without 372 Host Identity. 374 4.1. Host Identifiers 376 Host Identity adds two main features to Internet protocols. The 377 first is a decoupling of the internetworking and transport layers; 378 see Section 5. This decoupling will allow for independent evolution 379 of the two layers. Additionally, it can provide end-to-end services 380 over multiple internetworking realms. The second feature is host 381 authentication. Because the Host Identifier is a public key, this 382 key can be used for authentication in security protocols like IPsec. 384 The only completely defined structure of the Host Identity is that of 385 a public/private key pair. In this case, the Host Identity is 386 referred to by its public component, the public key. Thus, the name 387 representing a Host Identity in the Host Identity namespace, i.e., 388 the Host Identifier, is the public key. In a way, the possession of 389 the private key defines the Identity itself. If the private key is 390 possessed by more than one node, the Identity can be considered to be 391 a distributed one. 393 Architecturally, any other Internet naming convention might form a 394 usable base for Host Identifiers. However, non-cryptographic names 395 should only be used in situations of high trust - low risk. That is 396 any place where host authentication is not needed (no risk of host 397 spoofing) and no use of IPsec. However, at least for interconnected 398 networks spanning several operational domains, the set of 399 environments where the risk of host spoofing allowed by non- 400 cryptographic Host Identifiers is acceptable is the null set. Hence, 401 the current HIP documents do not specify how to use any other types 402 of Host Identifiers but public keys. 404 The actual Host Identities are never directly used in any Internet 405 protocols. The corresponding Host Identifiers (public keys) may be 406 stored in various DNS or LDAP directories as identified elsewhere in 407 this document, and they are passed in the HIP base exchange. A Host 408 Identity Tag (HIT) is used in other protocols to represent the Host 409 Identity. Another representation of the Host Identities, the Local 410 Scope Identifier (LSI), can also be used in protocols and APIs. 412 4.2. Storing Host Identifiers in DNS 414 The public Host Identifiers should be stored in DNS; the unpublished 415 Host Identifiers should not be stored anywhere (besides the 416 communicating hosts themselves). The (public) HI is stored in a new 417 RR type. This RR type is defined in HIP DNS Extension [RFC5205]. 419 Alternatively, or in addition to storing Host Identifiers in the DNS, 420 they may be stored in various kinds of Public Key Infrastructure 421 (PKI). Such a practice may allow them to be used for purposes other 422 than pure host identification. 424 4.3. Host Identity Tag (HIT) 426 A Host Identity Tag is a 128-bit representation for a Host Identity. 427 It is created by taking a cryptographic hash over the corresponding 428 Host Identifier. There are two advantages of using a hash over using 429 the Host Identifier in protocols. Firstly, its fixed length makes 430 for easier protocol coding and also better manages the packet size 431 cost of this technology. Secondly, it presents the identity in a 432 consistent format to the protocol independent of the cryptographic 433 algorithms used. 435 In the HIP packets, the HITs identify the sender and recipient of a 436 packet. Consequently, a HIT should be unique in the whole IP 437 universe as long as it is being used. In the extremely rare case of 438 a single HIT mapping to more than one Host Identity, the Host 439 Identifiers (public keys) will make the final difference. If there 440 is more than one public key for a given node, the HIT acts as a hint 441 for the correct public key to use. 443 4.4. Local Scope Identifier (LSI) 445 An LSI is a 32-bit localized representation for a Host Identity. The 446 purpose of an LSI is to facilitate using Host Identities in existing 447 protocols and APIs. LSI's advantage over HIT is its size; its 448 disadvantage is its local scope. 450 Examples of how LSIs can be used include: as the address in an FTP 451 command and as the address in a socket call. Thus, LSIs act as a 452 bridge for Host Identities into IPv4-based protocols and APIs. LSIs 453 also make it possible for some IPv4 applications to run over an IPv6 454 network. 456 5. New stack architecture 458 One way to characterize Host Identity is to compare the proposed new 459 architecture with the current one. As discussed above, the IP 460 addresses can be seen to be a confounding of routing direction 461 vectors and interface names. Using the terminology from the IRTF 462 Name Space Research Group Report [I-D.irtf-nsrg-report] and, e.g., 463 the unpublished Internet-Draft Endpoints and Endpoint Names 464 [chiappa-endpoints], the IP addresses currently embody the dual role 465 of locators and end-point identifiers. That is, each IP address 466 names a topological location in the Internet, thereby acting as a 467 routing direction vector, or locator. At the same time, the IP 468 address names the physical network interface currently located at the 469 point-of-attachment, thereby acting as a end-point name. 471 In the HIP architecture, the end-point names and locators are 472 separated from each other. IP addresses continue to act as locators. 473 The Host Identifiers take the role of end-point identifiers. It is 474 important to understand that the end-point names based on Host 475 Identities are slightly different from interface names; a Host 476 Identity can be simultaneously reachable through several interfaces. 478 The difference between the bindings of the logical entities are 479 illustrated in Figure 1. 481 Service ------ Socket Service ------ Socket 482 | | 483 | | 484 | | 485 | | 486 End-point | End-point --- Host Identity 487 \ | | 488 \ | | 489 \ | | 490 \ | | 491 Location --- IP address Location --- IP address 493 Figure 1 495 5.1. Transport associations and end-points 497 Architecturally, HIP provides for a different binding of transport- 498 layer protocols. That is, the transport-layer associations, i.e., 499 TCP connections and UDP associations, are no longer bound to IP 500 addresses but to Host Identities. 502 It is possible that a single physical computer hosts several logical 503 end-points. With HIP, each of these end-points would have a distinct 504 Host Identity. Furthermore, since the transport associations are 505 bound to Host Identities, HIP provides for process migration and 506 clustered servers. That is, if a Host Identity is moved from one 507 physical computer to another, it is also possible to simultaneously 508 move all the transport associations without breaking them. 509 Similarly, if it is possible to distribute the processing of a single 510 Host Identity over several physical computers, HIP provides for 511 cluster based services without any changes at the client end-point. 513 6. End-host mobility and multi-homing 515 HIP decouples the transport from the internetworking layer, and binds 516 the transport associations to the Host Identities (through actually 517 either the HIT or LSI). Consequently, HIP can provide for a degree 518 of internetworking mobility and multi-homing at a low infrastructure 519 cost. HIP mobility includes IP address changes (via any method) to 520 either party. Thus, a system is considered mobile if its IP address 521 can change dynamically for any reason like PPP, DHCP, IPv6 prefix 522 reassignments, or a NAT device remapping its translation. Likewise, 523 a system is considered multi-homed if it has more than one globally 524 routable IP address at the same time. HIP links IP addresses 525 together, when multiple IP addresses correspond to the same Host 526 Identity, and if one address becomes unusable, or a more preferred 527 address becomes available, existing transport associations can easily 528 be moved to another address. 530 When a node moves while communication is already on-going, address 531 changes are rather straightforward. The peer of the mobile node can 532 just accept a HIP or an integrity protected IPsec packet from any 533 address and ignore the source address. However, as discussed in 534 Section 6.2 below, a mobile node must send a HIP readdress packet to 535 inform the peer of the new address(es), and the peer must verify that 536 the mobile node is reachable through these addresses. This is 537 especially helpful for those situations where the peer node is 538 sending data periodically to the mobile node (that is re-starting a 539 connection after the initial connection). 541 6.1. Rendezvous mechanism 543 Making a contact to a mobile node is slightly more involved. In 544 order to start the HIP exchange, the initiator node has to know how 545 to reach the mobile node. Although infrequently moving HIP nodes 546 could use Dynamic DNS [RFC2136] to update their reachability 547 information in the DNS, an alternative to using DNS in this fashion 548 is to use a piece of new static infrastructure to facilitate 549 rendezvous between HIP nodes. 551 The mobile node keeps the rendezvous infrastructure continuously 552 updated with its current IP address(es). The mobile nodes must trust 553 the rendezvous mechanism to properly maintain their HIT and IP 554 address mappings. 556 The rendezvous mechanism is also needed if both of the nodes happen 557 to change their address at the same time, either because they are 558 mobile and happen to move at the same time, because one of them is 559 off-line for a while, or because of some other reason. In such a 560 case, the HIP readdress packets will cross each other in the network 561 and never reach the peer node. 563 The HIP rendezvous mechanism is defined in HIP Rendezvous [RFC5204]. 565 6.2. Protection against flooding attacks 567 Although the idea of informing about address changes by simply 568 sending packets with a new source address appears appealing, it is 569 not secure enough. That is, even if HIP does not rely on the source 570 address for anything (once the base exchange has been completed), it 571 appears to be necessary to check a mobile node's reachability at the 572 new address before actually sending any larger amounts of traffic to 573 the new address. 575 Blindly accepting new addresses would potentially lead to flooding 576 Denial-of-Service attacks against third parties [RFC4225]. In a 577 distributed flooding attack an attacker opens high volume HIP 578 connections with a large number of hosts (using unpublished HIs), and 579 then claims to all of these hosts that it has moved to a target 580 node's IP address. If the peer hosts were to simply accept the move, 581 the result would be a packet flood to the target node's address. To 582 close this attack, HIP includes an address check mechanism where the 583 reachability of a node is separately checked at each address before 584 using the address for larger amounts of traffic. 586 Whenever HIP is used between two hosts that fully trust each other, 587 the hosts may optionally decide to skip the address tests. However, 588 such performance optimization must be restricted to peers that are 589 known to be trustworthy and capable of protecting themselves from 590 malicious software. 592 7. HIP and IPsec 594 The preferred way of implementing HIP is to use IPsec to carry the 595 actual data traffic. As of today, the only completely defined method 596 is to use IPsec Encapsulated Security Payload (ESP) to carry the data 597 packets [RFC5202]. In the future, other ways of transporting payload 598 data may be developed, including ones that do not use cryptographic 599 protection. 601 In practice, the HIP base exchange uses the cryptographic Host 602 Identifiers to set up a pair of ESP Security Associations (SAs) to 603 enable ESP in an end-to-end manner. This is implemented in a way 604 that can span addressing realms. 606 While it would be possible, at least in theory, to use some existing 607 cryptographic protocol, such as IKEv2 together with Host Identifiers, 608 to establish the needed SAs, HIP defines a new protocol. There are a 609 number of historical reasons for this, and there are also a few 610 architectural reasons. First, IKE (and IKEv2) were not designed with 611 middle boxes in mind. As adding a new naming layer allows one to 612 potentially add a new forwarding layer (see Section 8, below), it is 613 very important that the HIP protocols are friendly towards any middle 614 boxes. 616 Second, from a conceptual point of view, the IPsec Security Parameter 617 Index (SPI) in ESP provides a simple compression of the HITs. This 618 does require per-HIT-pair SAs (and SPIs), and a decrease of policy 619 granularity over other Key Management Protocols, such as IKE and 620 IKEv2. In particular, the current thinking is limited to a situation 621 where, conceptually, there is only one pair of SAs between any given 622 pair of HITs. In other words, from an architectural point of view, 623 HIP only supports host-to-host (or endpoint-to-endpoint) Security 624 Associations. If two hosts need more pairs of parallel SAs, they 625 should use separate HITs for that. However, future HIP extensions 626 may provide for more granularity and creation of several ESP SAs 627 between a pair of HITs. 629 Since HIP is designed for host usage, not for gateways or so called 630 Bump-in-the-Wire (BITW) implementations, only ESP transport mode is 631 supported. An ESP SA pair is indexed by the SPIs and the two HITs 632 (both HITs since a system can have more than one HIT). The SAs need 633 not to be bound to IP addresses; all internal control of the SA is by 634 the HITs. Thus, a host can easily change its address using Mobile 635 IP, DHCP, PPP, or IPv6 readdressing and still maintain the SAs. 636 Since the transports are bound to the SA (via an LSI or a HIT), any 637 active transport is also maintained. Thus, real-world conditions 638 like loss of a PPP connection and its re-establishment or a mobile 639 handover will not require a HIP negotiation or disruption of 640 transport services [Bel1998]. 642 Since HIP does not negotiate any SA lifetimes, all lifetimes are 643 local policy. The only lifetimes a HIP implementation must support 644 are sequence number rollover (for replay protection), and SA timeout. 645 An SA times out if no packets are received using that SA. 646 Implementations may support lifetimes for the various ESP transforms. 648 8. HIP and NATs 650 Passing packets between different IP addressing realms requires 651 changing IP addresses in the packet header. This may happen, for 652 example, when a packet is passed between the public Internet and a 653 private address space, or between IPv4 and IPv6 networks. The 654 address translation is usually implemented as Network Address 655 Translation (NAT) [RFC3022] or NAT Protocol translation (NAT-PT) 656 [RFC2766]. 658 In a network environment where identification is based on the IP 659 addresses, identifying the communicating nodes is difficult when NAT 660 is used. With HIP, the transport-layer end-points are bound to the 661 Host Identities. Thus, a connection between two hosts can traverse 662 many addressing realm boundaries. The IP addresses are used only for 663 routing purposes; they may be changed freely during packet traversal. 665 For a HIP-based flow, a HIP-aware NAT or NAT-PT system tracks the 666 mapping of HITs, and the corresponding IPsec SPIs, to an IP address. 667 The NAT system has to learn mappings both from HITs and from SPIs to 668 IP addresses. Many HITs (and SPIs) can map to a single IP address on 669 a NAT, simplifying connections on address poor NAT interfaces. The 670 NAT can gain much of its knowledge from the HIP packets themselves; 671 however, some NAT configuration may be necessary. 673 NAT systems cannot touch the datagrams within the IPsec envelope, 674 thus application-specific address translation must be done in the end 675 systems. HIP provides for 'Distributed NAT', and uses the HIT or the 676 LSI as a placeholder for embedded IP addresses. 678 8.1. HIP and TCP checksums 680 There is no way for a host to know if any of the IP addresses in an 681 IP header are the addresses used to calculate the TCP checksum. That 682 is, it is not feasible to calculate the TCP checksum using the actual 683 IP addresses in the pseudo header; the addresses received in the 684 incoming packet are not necessarily the same as they were on the 685 sending host. Furthermore, it is not possible to recompute the 686 upper-layer checksums in the NAT/NAT-PT system, since the traffic is 687 IPsec protected. Consequently, the TCP and UDP checksums are 688 calculated using the HITs in the place of the IP addresses in the 689 pseudo header. Furthermore, only the IPv6 pseudo header format is 690 used. This provides for IPv4 / IPv6 protocol translation. 692 9. Multicast 694 There was little if any concrete thoughts about how HIP might affect 695 IP-layer or application-layer multicast. 697 10. HIP policies 699 There are a number of variables that will influence the HIP exchanges 700 that each host must support. All HIP implementations should support 701 at least 2 HIs, one to publish in DNS and an unpublished one for 702 anonymous usage. Although unpublished HIs will be rarely used as 703 responder HIs, they are likely be common for initiators. Support for 704 multiple HIs is recommended. 706 Many initiators would want to use a different HI for different 707 responders. The implementations should provide for a policy of 708 initiator HIT to responder HIT. This policy should also include 709 preferred transforms and local lifetimes. 711 Responders would need a similar policy, describing the hosts allowed 712 to participate in HIP exchanges, and the preferred transforms and 713 local lifetimes. 715 11. Benefits of HIP 717 In the beginning, the network layer protocol (i.e., IP) had the 718 following four "classic" invariants: 720 o Non-mutable: The address sent is the address received. 722 o Non-mobile: The address doesn't change during the course of an 723 "association". 725 o Reversible: A return header can always be formed by reversing the 726 source and destination addresses. 728 o Omniscient: Each host knows what address a partner host can use to 729 send packets to it. 731 Actually, the fourth can be inferred from 1 and 3, but it is worth 732 mentioning for reasons that will be obvious soon if not already. 734 In the current "post-classic" world, we are intentionally trying to 735 get rid of the second invariant (both for mobility and for multi- 736 homing), and we have been forced to give up the first and the fourth. 737 Realm Specific IP [RFC3102] is an attempt to reinstate the fourth 738 invariant without the first invariant. IPv6 is an attempt to 739 reinstate the first invariant. 741 Few systems on the Internet have DNS names that are meaningful. That 742 is, if they have a Fully Qualified Domain Name (FQDN), that name 743 typically belongs to a NAT device or a dial-up server, and does not 744 really identify the system itself but its current connectivity. 745 FQDNs (and their extensions as email names) are application-layer 746 names; more frequently naming services than a particular system. 747 This is why many systems on the Internet are not registered in the 748 DNS; they do not have services of interest to other Internet hosts. 750 DNS names are references to IP addresses. This only demonstrates the 751 interrelationship of the networking and application layers. DNS, as 752 the Internet's only deployed, distributed database is also the 753 repository of other namespaces, due in part to DNSSEC and application 754 specific key records. Although each namespace can be stretched (IP 755 with v6, DNS with KEY records), neither can adequately provide for 756 host authentication or act as a separation between internetworking 757 and transport layers. 759 The Host Identity (HI) namespace fills an important gap between the 760 IP and DNS namespaces. An interesting thing about the HI is that it 761 actually allows one to give up all but the 3rd network-layer 762 invariant. That is to say, as long as the source and destination 763 addresses in the network-layer protocol are reversible, then things 764 work ok because HIP takes care of host identification, and 765 reversibility allows one to get a packet back to one's partner host. 766 You do not care if the network-layer address changes in transit 767 (mutable) and you don't care what network-layer address the partner 768 is using (non-omniscient). 770 11.1. HIP's answers to NSRG questions 772 The IRTF Name Space Research Group has posed a number of evaluating 773 questions in their report [I-D.irtf-nsrg-report]. In this section, 774 we provide answers to these questions. 776 1. How would a stack name improve the overall functionality of the 777 Internet? 779 HIP decouples the internetworking layer from the transport 780 layer, allowing each to evolve separately. The decoupling 781 makes end-host mobility and multi-homing easier, also across 782 IPv4 and IPv6 networks. HIs make network renumbering easier, 783 and they also make process migration and clustered servers 784 easier to implement. Furthermore, being cryptographic in 785 nature, they provide the basis for solving the security 786 problems related to end-host mobility and multi-homing. 788 2. What does a stack name look like? 790 A HI is a cryptographic public key. However, instead of using 791 the keys directly, most protocols use a fixed size hash of the 792 public key. 794 3. What is its lifetime? 796 HIP provides both stable and temporary Host Identifiers. 797 Stable HIs are typically long lived, with a lifetime of years 798 or more. The lifetime of temporary HIs depends on how long 799 the upper-layer connections and applications need them, and 800 can range from a few seconds to years. 802 4. Where does it live in the stack? 804 The HIs live between the transport and internetworking layers. 806 5. How is it used on the end points 808 The Host Identifiers may be used directly or indirectly (in 809 the form of HITs or LSIs) by applications when they access 810 network services. Additionally, the Host Identifiers, as 811 public keys, are used in the built in key agreement protocol, 812 called the HIP base exchange, to authenticate the hosts to 813 each other. 815 6. What administrative infrastructure is needed to support it? 817 In some environments, it is possible to use HIP 818 opportunistically, without any infrastructure. However, to 819 gain full benefit from HIP, the HIs must be stored in the DNS 820 or a PKI, and a new rendezvous mechanism is needed[RFC5205]. 822 7. If we add an additional layer would it make the address list in 823 SCTP unnecessary? 825 Yes 827 8. What additional security benefits would a new naming scheme 828 offer? 830 HIP reduces dependency on IP addresses, making the so called 831 address ownership [Nik2001] problems easier to solve. In 832 practice, HIP provides security for end-host mobility and 833 multi-homing. Furthermore, since HIP Host Identifiers are 834 public keys, standard public key certificate infrastructures 835 can be applied on the top of HIP. 837 9. What would the resolution mechanisms be, or what characteristics 838 of a resolution mechanisms would be required? 840 For most purposes, an approach where DNS names are resolved 841 simultaneously to HIs and IP addresses is sufficient. 843 However, if it becomes necessary to resolve HIs into IP 844 addresses or back to DNS names, a flat resolution 845 infrastructure is needed. Such an infrastructure could be 846 based on the ideas of Distributed Hash Tables, but would 847 require significant new development and deployment. 849 12. Security considerations 851 HIP takes advantage of the new Host Identity paradigm to provide 852 secure authentication of hosts and to provide a fast key exchange for 853 IPsec. HIP also attempts to limit the exposure of the host to 854 various denial-of-service (DoS) and man-in-the-middle (MitM) attacks. 855 In so doing, HIP itself is subject to its own DoS and MitM attacks 856 that potentially could be more damaging to a host's ability to 857 conduct business as usual. 859 Resource exhausting denial-of-service attacks take advantage of the 860 cost of setting up a state for a protocol on the responder compared 861 to the 'cheapness' on the initiator. HIP allows a responder to 862 increase the cost of the start of state on the initiator and makes an 863 effort to reduce the cost to the responder. This is done by having 864 the responder start the authenticated Diffie-Hellman exchange instead 865 of the initiator, making the HIP base exchange 4 packets long. There 866 are more details on this process in the Host Identity Protocol under 867 development. 869 HIP optionally supports opportunistic negotiation. That is, if a 870 host receives a start of transport without a HIP negotiation, it can 871 attempt to force a HIP exchange before accepting the connection. 872 This has the potential for DoS attacks against both hosts. If the 873 method to force the start of HIP is expensive on either host, the 874 attacker need only spoof a TCP SYN. This would put both systems into 875 the expensive operations. HIP avoids this attack by having the 876 responder send a simple HIP packet that it can pre-build. Since this 877 packet is fixed and easily replayed, the initiator only reacts to it 878 if it has just started a connection to the responder. 880 Man-in-the-middle attacks are difficult to defend against, without 881 third-party authentication. A skillful MitM could easily handle all 882 parts of the HIP base exchange, but HIP indirectly provides the 883 following protection from a MitM attack. If the responder's HI is 884 retrieved from a signed DNS zone or secured by some other means, the 885 initiator can use this to authenticate the signed HIP packets. 886 Likewise, if the initiator's HI is in a secure DNS zone, the 887 responder can retrieve it and validate the signed HIP packets. 888 However, since an initiator may choose to use an unpublished HI, it 889 knowingly risks a MitM attack. The responder may choose not to 890 accept a HIP exchange with an initiator using an unknown HI. 892 In HIP, the Security Association for IPsec is indexed by the SPI; the 893 source address is always ignored, and the destination address may be 894 ignored as well. Therefore, HIP-enabled IPsec Encapsulated Security 895 Payload (ESP) is IP address independent. This might seem to make it 896 easier for an attacker, but ESP with replay protection is already as 897 well protected as possible, and the removal of the IP address as a 898 check should not increase the exposure of IPsec ESP to DoS attacks. 900 Since not all hosts will ever support HIP, ICMPv4 'Destination 901 Unreachable, Protocol Unreachable' and ICMPv6 'Parameter Problem, 902 Unrecognized Next Header' messages are to be expected and present a 903 DoS attack. Against an initiator, the attack would look like the 904 responder does not support HIP, but shortly after receiving the ICMP 905 message, the initiator would receive a valid HIP packet. Thus, to 906 protect against this attack, an initiator should not react to an ICMP 907 message until a reasonable time has passed, allowing it to get the 908 real responder's HIP packet. A similar attack against the responder 909 is more involved. 911 Another MitM attack is simulating a responder's administrative 912 rejection of a HIP initiation. This is a simple ICMP 'Destination 913 Unreachable, Administratively Prohibited' message. A HIP packet is 914 not used because it would either have to have unique content, and 915 thus difficult to generate, resulting in yet another DoS attack, or 916 just as spoofable as the ICMP message. Like in the previous case, 917 the defense against this attack is for the initiator to wait a 918 reasonable time period to get a valid HIP packet. If one does not 919 come, then the initiator has to assume that the ICMP message is 920 valid. Since this is the only point in the HIP base exchange where 921 this ICMP message is appropriate, it can be ignored at any other 922 point in the exchange. 924 12.1. HITs used in ACLs 926 It is expected that HITs will be used in ACLs. Future firewalls can 927 use HITs to control egress and ingress to networks, with an assurance 928 level difficult to achieve today. As discussed above in Section 7, 929 once a HIP session has been established, the SPI value in an IPsec 930 packet may be used as an index, indicating the HITs. In practice, 931 firewalls can inspect HIP packets to learn of the bindings between 932 HITs, SPI values, and IP addresses. They can even explicitly control 933 IPsec usage, dynamically opening IPsec ESP only for specific SPI 934 values and IP addresses. The signatures in HIP packets allow a 935 capable firewall to ensure that the HIP exchange is indeed happening 936 between two known hosts. This may increase firewall security. 938 There has been considerable bad experience with distributed ACLs that 939 contain public key related material, for example, with SSH. If the 940 owner of a key needs to revoke it for any reason, the task of finding 941 all locations where the key is held in an ACL may be impossible. If 942 the reason for the revocation is due to private key theft, this could 943 be a serious issue. 945 A host can keep track of all of its partners that might use its HIT 946 in an ACL by logging all remote HITs. It should only be necessary to 947 log responder hosts. With this information, the host can notify the 948 various hosts about the change to the HIT. There has been no attempt 949 to develop a secure method to issue the HIT revocation notice. 951 HIP-aware NATs, however, are transparent to the HIP aware systems by 952 design. Thus, the host may find it difficult to notify any NAT that 953 is using a HIT in an ACL. Since most systems will know of the NATs 954 for their network, there should be a process by which they can notify 955 these NATs of the change of the HIT. This is mandatory for systems 956 that function as responders behind a NAT. In a similar vein, if a 957 host is notified of a change in a HIT of an initiator, it should 958 notify its NAT of the change. In this manner, NATs will get updated 959 with the HIT change. 961 12.2. Non-security considerations 963 The definition of the Host Identifier states that the HI need not be 964 a public key. It implies that the HI could be any value; for example 965 a FQDN. This document does not describe how to support such a non- 966 cryptographic HI. A non-cryptographic HI would still offer the 967 services of the HIT or LSI for NAT traversal. It would be possible 968 to carry HITs in HIP packets that had neither privacy nor 969 authentication. Since such a mode would offer so little additional 970 functionality for so much addition to the IP kernel, it has not been 971 defined. Given how little public key cryptography HIP requires, HIP 972 should only be implemented using public key Host Identities. 974 If it is desirable to use HIP in a low security situation where 975 public key computations are considered expensive, HIP can be used 976 with very short Diffie-Hellman and Host Identity keys. Such use 977 makes the participating hosts vulnerable to MitM and connection 978 hijacking attacks. However, it does not cause flooding dangers, 979 since the address check mechanism relies on the routing system and 980 not on cryptographic strength. 982 13. IANA considerations 984 This document has no actions for IANA. 986 14. Acknowledgments 988 For the people historically involved in the early stages of HIP, see 989 the Acknowledgements section in the Host Identity Protocol 990 specification. 992 During the later stages of this document, when the editing baton was 993 transfered to Pekka Nikander, the comments from the early 994 implementors and others, including Jari Arkko, Tom Henderson, Petri 995 Jokela, Miika Komu, Mika Kousa, Andrew McGregor, Jan Melen, Tim 996 Shepard, Jukka Ylitalo, and Jorma Wall, were invaluable. Finally, 997 Lars Eggert, Spencer Dawkins and Dave Crocker provided valuable input 998 during the final stages of publication, most of which was 999 incorporated but some of which the authors decided to ignore in order 1000 to get this document published in the first place. 1002 The authors want to express their special thanks to Tom Henderson, 1003 who took the burden of editing the document in response to IESG 1004 comments at the time when both of the authors were busy doing other 1005 things. Without his perseverance this document might have never made 1006 it into an RFC form. 1008 15. References 1010 15.1. Normative References 1012 [RFC5202] Jokela, P., Moskowitz, R., and P. Nikander, "Using the 1013 Encapsulating Security Payload (ESP) Transport Format with 1014 the Host Identity Protocol (HIP)", RFC 5202, April 2008. 1016 [RFC5204] Laganier, J. and L. Eggert, "Host Identity Protocol (HIP) 1017 Rendezvous Extension", RFC 5204, April 2008. 1019 [RFC5205] Nikander, P. and J. Laganier, "Host Identity Protocol 1020 (HIP) Domain Name System (DNS) Extensions", RFC 5205, 1021 April 2008. 1023 15.2. Informative references 1025 [RFC2136] Vixie, P., Thomson, S., Rekhter, Y., and J. Bound, 1026 "Dynamic Updates in the Domain Name System (DNS UPDATE)", 1027 RFC 2136, April 1997. 1029 [RFC2535] Eastlake, D., "Domain Name System Security Extensions", 1030 RFC 2535, March 1999. 1032 [RFC2766] Tsirtsis, G. and P. Srisuresh, "Network Address 1033 Translation - Protocol Translation (NAT-PT)", RFC 2766, 1034 February 2000. 1036 [RFC3022] Srisuresh, P. and K. Egevang, "Traditional IP Network 1037 Address Translator (Traditional NAT)", RFC 3022, 1038 January 2001. 1040 [RFC3102] Borella, M., Lo, J., Grabelsky, D., and G. Montenegro, 1041 "Realm Specific IP: Framework", RFC 3102, October 2001. 1043 [RFC4025] Richardson, M., "A Method for Storing IPsec Keying 1044 Material in DNS", RFC 4025, March 2005. 1046 [RFC4225] Nikander, P., Arkko, J., Aura, T., Montenegro, G., and E. 1047 Nordmark, "Mobile IP Version 6 Route Optimization Security 1048 Design Background", RFC 4225, December 2005. 1050 [RFC4306] Kaufman, C., "Internet Key Exchange (IKEv2) Protocol", 1051 RFC 4306, December 2005. 1053 [I-D.irtf-nsrg-report] 1054 Lear, E. and R. Droms, "What's In A Name:Thoughts from the 1055 NSRG", draft-irtf-nsrg-report-10 (work in progress), 1056 September 2003. 1058 [chiappa-endpoints] 1059 Chiappa, J., "Endpoints and Endpoint Names: A Proposed 1060 Enhancement to the Internet Architecture", 1061 URL http://www.chiappa.net/~jnc/tech/endpoints.txt, 1999. 1063 [Nik2001] Nikander, P., "Denial-of-Service, Address Ownership, and 1064 Early Authentication in the IPv6 World", in Proceesings 1065 of Security Protocols, 9th International Workshop, 1066 Cambridge, UK, April 25-27 2001, LNCS 2467, pp. 12-26, 1067 Springer, 2002. 1069 [Bel1998] Bellovin, S., "EIDs, IPsec, and HostNAT", in Proceedings 1070 of 41th IETF, Los Angeles, CA, 1071 URL http://www1.cs.columbia.edu/~smb/talks/hostnat.pdf, 1072 March 1998. 1074 Authors' Addresses 1076 Robert Moskowitz 1077 ICSAlabs, An Independent Division of Verizon Business Systems 1078 1000 Bent Creek Blvd, Suite 200 1079 Mechanicsburg, PA 1080 USA 1082 Email: robert.moskowitz@icsalabs.com 1084 Pekka Nikander 1085 Ericsson Research Nomadic Lab 1086 JORVAS FIN-02420 1087 FINLAND 1089 Phone: +358 9 299 1 1090 Email: pekka.nikander@nomadiclab.com