idnits 2.17.1 draft-ietf-intarea-adhoc-wireless-com-01.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. 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 -- The document date (January 12, 2016) is 3020 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- No issues found here. Summary: 0 errors (**), 0 flaws (~~), 1 warning (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Internet Area E. Baccelli 3 Internet-Draft INRIA 4 Intended status: Informational C. Perkins 5 Expires: July 15, 2016 Futurewei 6 January 12, 2016 8 Multi-hop Ad Hoc Wireless Communication 9 draft-ietf-intarea-adhoc-wireless-com-01 11 Abstract 13 This document describes characteristics of communication between 14 interfaces in a multi-hop ad hoc wireless network, that protocol 15 engineers and system analysts should be aware of when designing 16 solutions for ad hoc networks at the IP layer. 18 Status of This Memo 20 This Internet-Draft is submitted in full conformance with the 21 provisions of BCP 78 and BCP 79. 23 Internet-Drafts are working documents of the Internet Engineering 24 Task Force (IETF). Note that other groups may also distribute 25 working documents as Internet-Drafts. The list of current Internet- 26 Drafts is at http://datatracker.ietf.org/drafts/current/. 28 Internet-Drafts are draft documents valid for a maximum of six months 29 and may be updated, replaced, or obsoleted by other documents at any 30 time. It is inappropriate to use Internet-Drafts as reference 31 material or to cite them other than as "work in progress." 33 This Internet-Draft will expire on July 15, 2016. 35 Copyright Notice 37 Copyright (c) 2016 IETF Trust and the persons identified as the 38 document authors. All rights reserved. 40 This document is subject to BCP 78 and the IETF Trust's Legal 41 Provisions Relating to IETF Documents 42 (http://trustee.ietf.org/license-info) in effect on the date of 43 publication of this document. Please review these documents 44 carefully, as they describe your rights and restrictions with respect 45 to this document. Code Components extracted from this document must 46 include Simplified BSD License text as described in Section 4.e of 47 the Trust Legal Provisions and are provided without warranty as 48 described in the Simplified BSD License. 50 Table of Contents 52 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 53 2. Multi-hop Ad Hoc Wireless Networks . . . . . . . . . . . . . 2 54 3. Common Packet Transmission Characteristics in 55 Multi-hop Ad Hoc Wireless Networks . . . . . . . . . . . . . 3 56 3.1. Asymmetry, Time-Variation, and Non-Transitivity . . . . . 4 57 3.2. Radio Range and Wireless Irregularities . . . . . . . . . 4 58 4. Alternative Terminology . . . . . . . . . . . . . . . . . . . 7 59 5. Security Considerations . . . . . . . . . . . . . . . . . . . 8 60 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 9 61 7. Informative References . . . . . . . . . . . . . . . . . . . 9 62 Appendix A. Acknowledgements . . . . . . . . . . . . . . . . . . 12 63 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 12 65 1. Introduction 67 Experience gathered with ad hoc routing protocol development, 68 deployment and operation, shows that wireless communication presents 69 specific challenges [RFC2501] [DoD01], which Internet protocol 70 designers should be aware of, when designing solutions for ad hoc 71 networks at the IP layer. This document does not prescribe 72 solutions, but instead briefly describes these challenges in hopes of 73 increasing that awareness. 75 As background, RFC 3819 [RFC3819] provides an excellent reference for 76 higher-level considerations when designing protocols for shared 77 media. From MTU to subnet design, from security to considerations 78 about retransmissions, RFC 3819 provides guidance and design 79 rationale to help with many aspects of higher-level protocol design. 81 The present document focuses more specifically on challenges in 82 multi-hop ad hoc wireless networking. For example, in that context, 83 even though a wireless link may experience high variability as a 84 communications channel, such variation does not mean that the link is 85 "broken"; indeed many layer-2 technologies serve to reduce error 86 rates by various means. Nevertheless, such errors as noted in this 87 document may still become visible above layer-2 and so become 88 relevant to the operation of higher layer protocols. 90 2. Multi-hop Ad Hoc Wireless Networks 92 For the purposes of this document, a multi-hop ad hoc wireless 93 network will be considered to be a collection of devices that each 94 have a radio transceiver (i.e., wireless network interface), and that 95 are moreover configured to self-organize and provide store-and- 96 forward functionality as needed to enable communications. This 97 document focuses on the characteristics of communications through 98 such a network interface. 100 Although the characteristics of packet transmission over multi-hop ad 101 hoc wireless networks, described below, are not the typical 102 characteristics expected by IP [RFC6250], it is desirable and 103 possible to run IP over such networks, as demonstrated in certain 104 deployments currently in operation, such as Freifunk [FREIFUNK], and 105 Funkfeuer [FUNKFEUER]. These deployments use routers running IP 106 protocols e.g., OLSR (Optimized Link State Routing [RFC3626]) on top 107 of IEEE 802.11 in ad hoc mode with the same ESSID (Extended Service 108 Set Identification) at the link layer. Multi-hop ad hoc wireless 109 networks may also run on link layers other than IEEE 802.11, and may 110 use routing protocols other than OLSR (for instance, AODV [RFC3561], 111 TBRPF [RFC3684], DSR [RFC4728], or OSPF-MPR [RFC5449]). 113 Note that in contrast, devices communicating via an IEEE 802.11 114 access point in infrastructure mode do not form a multi-hop ad hoc 115 wireless network, since the central role of the access point is 116 predetermined, and devices other than the access point do not 117 generally provide store-and-forward functionality. 119 3. Common Packet Transmission Characteristics in Multi-hop Ad Hoc 120 Wireless Networks 122 In the following, we will consider several devices in a multi-hop ad 123 hoc wireless network N. Each device will be considered only through 124 its own wireless interface to network N. For conciseness and 125 readability, this document uses the expressions "device A" (or simply 126 "A") as a synonym for "the wireless interface of device A to network 127 N". 129 Let A and B be two devices in network N. Suppose that, when device A 130 transmits an IP packet through its interface on network N, that 131 packet is correctly and directly received by device B without 132 requiring storage and/or forwarding by any other device. We will 133 then say that B can "detect" A. Note that therefore, when B detects 134 A, an IP packet transmitted by A will be rigorously identical to the 135 corresponding IP packet received by B. 137 Let S be the set of devices that detect device A through its wireless 138 interface on network N. The following section gathers common 139 characteristics concerning packet transmission over such networks, 140 which were observed through experience with MANET routing protocol 141 development (for instance, OLSR[RFC3626], AODV[RFC3561], 142 TBRPF[RFC3684], DSR[RFC4728], and OSPF-MPR[RFC5449]), as well as 143 deployment and operation (Freifunk[FREIFUNK], Funkfeuer[FUNKFEUER]). 145 3.1. Asymmetry, Time-Variation, and Non-Transitivity 147 First, even though a device C in set S can (by definition) detect 148 device A, there is no guarantee that C can, conversely, send IP 149 packets directly to A. In other words, even though C can detect A 150 (since it is a member of set S), there is no guarantee that A can 151 detect C. Thus, multi-hop ad hoc wireless communications may be 152 "asymmetric". Such cases are common. 154 Second, there is no guarantee that, as a set, S is at all stable, 155 i.e. the membership of set S may in fact change at any rate, at any 156 time. Thus, multi-hop ad hoc wireless communications may be "time- 157 variant". Time variation is often observed in multi-hop ad hoc 158 wireless networks due to variability of the wireless medium, and to 159 device mobility. 161 Now, conversely, let V be the set of devices which A detects. 162 Suppose that A is communicating at time t0 through its interface on 163 network N. As a consequence of time variation and asymmetry, we 164 observe that A: 166 1. cannot assume that S = V, 168 2. cannot assume that S and/or V are unchanged at time t1 later than 169 t0. 171 Furthermore, transitivity is not guaranteed over multi-hop ad hoc 172 wireless networks. Indeed, let's assume that, through their 173 respective interfaces within network N: 175 1. device B and device A can detect one another (i.e. B is a member 176 of sets S and V), and, 178 2. device A and device C can also detect one another (i.e. C is a 179 also a member of sets S and V). 181 These assumptions do not imply that B can detect C, nor that C can 182 detect B (through their interface on network N). Such "non- 183 transitivity" is common on multi-hop ad hoc wireless networks. 185 In a nutshell: multi-hop ad hoc wireless communications can be 186 asymmetric, non-transitive, and time-varying. 188 3.2. Radio Range and Wireless Irregularities 190 Section 3.1 presents an abstract description of some common 191 characteristics concerning packet transmission over multi-hop ad hoc 192 wireless networks. This section describes practical examples, which 193 illustrate the characteristics listed in Section 3.1 as well as other 194 common effects. 196 Wireless communications are subject to limitations to the distance 197 across which they may be established. The range-limitation factor 198 creates specific problems on multi-hop ad hoc wireless networks. In 199 this context, the radio ranges of several devices often partially 200 overlap. Such partial overlap causes communication to be non- 201 transitive and/or asymmetric, as described in Section 3.1. Moreover, 202 the range may vary from one device to another, depending on location 203 and environmental factors. This is in addition to the time variation 204 of range and signal strength caused by variability in the local 205 environment. 207 For example, as depicted in Figure 1, it may happen that a device B 208 detects a device A which transmits at high power, whereas B transmits 209 at lower power. In such cases, B detects A, but A cannot detect B. 210 This examplifies the asymmetry in multi-hop ad hoc wireless 211 communications as defined in Section 3.1. 213 Radio Ranges for Devices A and B 215 <~~~~~~~~~~~~~+~~~~~~~~~~~~~> 216 | <~~~~~~+~~~~~~> 217 +--|--+ +--|--+ 218 | A |======>| B | 219 +-----+ +-----+ 221 Figure 1: Asymmetric wireless communication: Device A can 222 communicate with device B, but B cannot communicate with A. 224 Another example, depicted in Figure 2, is known as the "Hidden 225 Terminal" problem. Even though the devices all have equal power for 226 their radio transmissions, they cannot all detect one another. In 227 the figure, devices A and B can detect one another, and devices A and 228 C can also detect one another. On the other hand, B and C cannot 229 detect one another. When B and C simultaneously try to communicate 230 with A, their radio signals may collide. Device A may receive 231 incoherent noise, and may even be unable to determine the source of 232 the noise. The hidden terminal problem illustrates the property of 233 non-transitivity in multi-hop ad hoc wireless communications as 234 described in Section 3.1. 236 Radio Ranges for Devices A, B, C 238 <~~~~~~~~~~~~~+~~~~~~~~~~~~~> <~~~~~~~~~~~~~+~~~~~~~~~~~~~> 239 |<~~~~~~~~~~~~~+~~~~~~~~~~~~~>| 240 +--|--+ +--|--+ +--|--+ 241 | B |=======>| A |<=======| C | 242 +-----+ +-----+ +-----+ 244 Figure 2: The hidden terminal problem. Devices C and B 245 try to communicate with device A at the same time, 246 and their radio signals collide. 248 Another situation, shown in Figure 3, is known as the "Exposed 249 Terminal" problem. In the figure, device A and device B can detect 250 each other, and A is transmitting packets to B, thus A cannot detect 251 device C -- but C can detect A. As shown in Figure 3, during the on- 252 going transmission of A, device C cannot reliably communicate with 253 device D because of interference within C's radio range due to A's 254 transmissions. Device C is then said to be "exposed", because it is 255 exposed to co-channel interference from A and is thereby prevented 256 from reliably exchanging protocol messages with D -- even though 257 these transmissions would not interfere with the reception of data 258 sent from A destined to B. 260 Radio Ranges for Devices A, B, C, D 262 <~~~~~~~~~~~~+~~~~~~~~~~~~> <~~~~~~~~~~+~~~~~~~~~~~> 263 |<~~~~~~~~~~~~+~~~~~~~~~~~~>|<~~~~~~~~~~~~+~~~~~~~~~> 264 +--|--+ +--|--+ +--|--+ +--|--+ 265 | B |<======| A | | C |======>| D | 266 +-----+ +-----+ +-----+ +-----+ 268 Figure 3: The exposed terminal problem: when device A 269 communicates with device B, device C is "exposed". 271 Hidden and exposed terminal situations are often observed in multi- 272 hop ad hoc wireless networks. Asymmetry issues with wireless 273 communication may also arise for reasons other than power inequality 274 (e.g., multipath interference). Such problems are often resolved by 275 specific mechanisms below the IP layer, for example, CSMA/CA, which 276 ensures transmission in periods perceived to be unoccupied by other 277 transmissions. However, depending on the link layer technology in 278 use and the position of the devices, such problems may affect the IP 279 layer due to range-limitation and partial overlap . 281 Besides radio range limitations, wireless communications are affected 282 by irregularities in the shape of the geographical area over which 283 devices may effectively communicate (see for instance [MC03], 284 [MI03]). For example, even omnidirectional wireless transmission is 285 typically non-isotropic (i.e. non-circular). Signal strength often 286 suffers frequent and significant variations, which are not a simple 287 function of distance. Instead, it is a complex function of the 288 environment including obstacles, weather conditions, interference, 289 and other factors that change over time. Because wireless 290 communications have to encounter different terrain, path, 291 obstructions, atmospheric conditions and other phenomena, analytical 292 formulation of signal strength is considered intractable [VTC99], and 293 the radio engineering community has thus developed numerous radio 294 propagation models, relying on median values observed in specific 295 environments [SAR03]. 297 The above irregularities also cause communications on multi-hop ad 298 hoc wireless networks to be non-transitive, asymmetric, or time- 299 varying, as described in Section 3.1, and may impact protocols at the 300 IP layer and above. There may be no indication to the IP layer when 301 a previously established communication channel becomes unusable; 302 "link down" triggers are generally absent in multi-hop ad hoc 303 wireless networks, since the absence of detectable radio energy 304 (e.g., in carrier waves) may simply indicate that neighboring devices 305 are not currently transmitting. Such an absence of detectable radio 306 energy does not therefore indicate whether or not transmissions have 307 failed to reach the intended destination. 309 4. Alternative Terminology 311 Many terms have been used in the past to describe the relationship of 312 devices in a multi-hop ad hoc wireless network based on their ability 313 to send or receive packets to/from each other. The terms used in 314 previous sections of this document have been selected because the 315 authors believe they are unambiguous, with respect to the goal of 316 this document (see Section 1). 318 In this section, we exhibit some other terms that describe the same 319 relationship between devices in multi-hop ad hoc wireless networks. 320 In the following, let network N be, again, a multi-hop ad hoc 321 wireless network. Let the set S be, as before, the set of devices 322 that can directly receive packets transmitted by device A through its 323 interface on network N. In other words, any device B belonging to S 324 can detect packets transmitted by A. Then, due to the asymmetric 325 nature of wireless communications: 327 - We may say that device A "reaches" device B. In this 328 terminology, there is no guarantee that B reaches A, even if A 329 reaches B. 331 - We may say that device B "hears" device A. In this terminology, 332 there is no guarantee that A hears B, even if B hears A. 334 - We may say that device A "has a link" to device B. In this 335 terminology, there is no guarantee that B has a link to A, even if 336 A has a link to B. 338 - We may say that device B "is adjacent to" device A. In this 339 terminology, there is no guarantee that A is adjacent to B, even 340 if B is adjacent to A. 342 - We may say that device B "is downstream from" device A. In this 343 terminology, there is no guarantee that A is downstream from B, 344 even if B is downstream from A. 346 - We may say that device B "is a neighbor of" device A. In this 347 terminology, there is no guarantee that A is a neighbor of B, even 348 if B a neighbor of A. As it happens, terminology based on 349 "neighborhood" is quite confusing for multi-hop wireless 350 communications. For example, when B can detect A, but A cannot 351 detect B, it is not clear whether B should be considered a 352 neighbor of A at all, since A would not necessarily be aware that 353 B was a neighbor, as it cannot detect B. It is thus best to avoid 354 the "neighbor" terminology, except for when some level of symmetry 355 has been verified. 357 This list of alternative terminologies is given here for illustrative 358 purposes only, and is not suggested to be complete or even 359 representative of the breadth of terminologies that have been used in 360 various ways to explain the properties mentioned in Section 3. We do 361 not discuss bidirectionality, but as a final observation it is 362 worthwhile to note that bidirectionality is not synonymous with 363 symmetry. For example, the error statistics in either direction are 364 often different for a link that is otherwise considered 365 bidirectional. 367 5. Security Considerations 369 Section 18 of RFC 3819 [RFC3819] provides an excellent overview of 370 security considerations at the subnetwork layer. Beyond the material 371 there, multi-hop ad hoc wireless networking (i) is not limited to 372 subnetwork layer operation, and (ii) makes use of wireless 373 communications. 375 On one hand, a detailed description of security implications of 376 wireless communications in general is outside of the scope of this 377 document. Notably, however, eavesdropping on a wireless link is much 378 easier than for wired media (although significant progress has been 379 made in the field of wireless monitoring of wired transmissions). As 380 a result, traffic analysis attacks can be even more subtle and 381 difficult to defeat in this context. Furthermore, such 382 communications over a shared media are particularly prone to theft of 383 service and denial of service (DoS) attacks. 385 On the other hand, the potential multi-hop aspect of the networks we 386 consider in this document goes beyond traditional scope of subnetwork 387 design. In practice, unplanned relaying of network traffic (both 388 user traffic and control traffic) happens routinely. Due to the 389 physical nature of wireless media, Man in the Middle (MITM) attacks 390 are facilitated, which may significantly alter network performance. 391 This highlights the need to stick to the "end-to-end principle": L3 392 security, end-to-end, becomes a primary goal, independently of 393 securing layer-2 and layer-1 protocols (though L2 and L1 security can 394 indeed help to reach this goal). 396 6. IANA Considerations 398 This document does not have any IANA actions. 400 7. Informative References 402 [RFC2501] Corson, S. and J. Macker, "Mobile Ad hoc Networking 403 (MANET): Routing Protocol Performance Issues and 404 Evaluation Considerations", RFC 2501, 405 DOI 10.17487/RFC2501, January 1999, 406 . 408 [RFC3561] Perkins, C., Belding-Royer, E., and S. Das, "Ad hoc On- 409 Demand Distance Vector (AODV) Routing", RFC 3561, 410 DOI 10.17487/RFC3561, July 2003, 411 . 413 [RFC3626] Clausen, T., Ed. and P. Jacquet, Ed., "Optimized Link 414 State Routing Protocol (OLSR)", RFC 3626, 415 DOI 10.17487/RFC3626, October 2003, 416 . 418 [RFC3684] Ogier, R., Templin, F., and M. Lewis, "Topology 419 Dissemination Based on Reverse-Path Forwarding (TBRPF)", 420 RFC 3684, DOI 10.17487/RFC3684, February 2004, 421 . 423 [RFC3819] Karn, P., Ed., Bormann, C., Fairhurst, G., Grossman, D., 424 Ludwig, R., Mahdavi, J., Montenegro, G., Touch, J., and L. 425 Wood, "Advice for Internet Subnetwork Designers", BCP 89, 426 RFC 3819, DOI 10.17487/RFC3819, July 2004, 427 . 429 [RFC4728] Johnson, D., Hu, Y., and D. Maltz, "The Dynamic Source 430 Routing Protocol (DSR) for Mobile Ad Hoc Networks for 431 IPv4", RFC 4728, DOI 10.17487/RFC4728, February 2007, 432 . 434 [RFC5449] Baccelli, E., Jacquet, P., Nguyen, D., and T. Clausen, 435 "OSPF Multipoint Relay (MPR) Extension for Ad Hoc 436 Networks", RFC 5449, DOI 10.17487/RFC5449, February 2009, 437 . 439 [RFC6250] Thaler, D., "Evolution of the IP Model", RFC 6250, 440 DOI 10.17487/RFC6250, May 2011, 441 . 443 [DoD01] Freebersyser, J. and B. Leiner, "A DoD perspective on 444 mobile ad hoc networks", Addison Wesley C. E. Perkins, 445 Ed., 2001, pp. 29--51, 2001. 447 [FUNKFEUER] 448 "Austria Wireless Community Network, 449 http://www.funkfeuer.at", 2013. 451 [MC03] Corson, S. and J. Macker, "Mobile Ad hoc Networking: 452 Routing Technology for Dynamic, Wireless Networks", IEEE 453 Press Mobile Ad hoc Networking, Chapter 9, 2003. 455 [SAR03] Sarkar, T., Ji, Z., Kim, K., Medour, A., and M. Salazar- 456 Palma, "A Survey of Various Propagation Models for Mobile 457 Communication", IEEE Press Antennas and Propagation 458 Magazine, Vol. 45, No. 3, 2003. 460 [VTC99] Kim, D., Chang, Y., and J. Lee, "Pilot power control and 461 service coverage support in CDMA mobile systems", IEEE 462 Press Proceedings of the IEEE Vehicular Technology 463 Conference (VTC), pp.1464-1468, 1999. 465 [MI03] Kotz, D., Newport, C., and C. Elliott, "The Mistaken 466 Axioms of Wireless-Network Research", Dartmouth College 467 Computer Science Technical Report TR2003-467, 2003. 469 [FREIFUNK] 470 "Freifunk Wireless Community Networks, 471 http://www.freifunk.net", 2013. 473 Appendix A. Acknowledgements 475 This document stems from discussions with the following people, in 476 alphabetical order: Jari Arkko, Teco Boot, Carlos Jesus Bernardos 477 Cano, Ian Chakeres, Thomas Clausen, Robert Cragie, Christopher 478 Dearlove, Ralph Droms, Brian Haberman, Ulrich Herberg, Paul Lambert, 479 Kenichi Mase, Thomas Narten, Erik Nordmark, Alexandru Petrescu, Stan 480 Ratliff, Zach Shelby, Shubhranshu Singh, Fred Templin, Dave Thaler, 481 Mark Townsley, Ronald Velt in't, and Seung Yi. 483 Authors' Addresses 485 Emmanuel Baccelli 486 INRIA 488 EMail: Emmanuel.Baccelli@inria.fr 489 URI: http://www.emmanuelbaccelli.org/ 491 Charles E. Perkins 492 Futurewei 494 Phone: +1-408-330-4586 495 EMail: charlie.perkins@huawei.com