idnits 2.17.1 draft-ietf-issll-isslow-05.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** Looks like you're using RFC 2026 boilerplate. This must be updated to follow RFC 3978/3979, as updated by RFC 4748. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- ** Missing expiration date. The document expiration date should appear on the first and last page. ** The document seems to lack a 1id_guidelines paragraph about 6 months document validity -- however, there's a paragraph with a matching beginning. Boilerplate error? == No 'Intended status' indicated for this document; assuming Proposed Standard Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** The document seems to lack an IANA Considerations section. (See Section 2.2 of https://www.ietf.org/id-info/checklist for how to handle the case when there are no actions for IANA.) ** The document seems to lack separate sections for Informative/Normative References. All references will be assumed normative when checking for downward references. ** The abstract seems to contain references ([1]), which it shouldn't. Please replace those with straight textual mentions of the documents in question. Miscellaneous warnings: ---------------------------------------------------------------------------- == Line 306 has weird spacing: '...tion is able ...' == Line 308 has weird spacing: '...at only initi...' -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (April 1999) is 9142 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: '5' is defined on line 547, but no explicit reference was found in the text == Outdated reference: A later version (-03) exists of draft-ietf-mmusic-confarch-00 -- Possible downref: Normative reference to a draft: ref. '1' -- Possible downref: Non-RFC (?) normative reference: ref. '3' ** Obsolete normative reference: RFC 1717 (ref. '5') (Obsoleted by RFC 1990) ** Obsolete normative reference: RFC 1889 (ref. '6') (Obsoleted by RFC 3550) == Outdated reference: A later version (-06) exists of draft-ietf-issll-isslow-mcml-05 == Outdated reference: A later version (-05) exists of draft-ietf-issll-isslow-rtf-04 ** Obsolete normative reference: RFC 2509 (ref. '10') (Obsoleted by RFC 3544) -- Possible downref: Non-RFC (?) normative reference: ref. '13' -- Possible downref: Non-RFC (?) normative reference: ref. '14' -- Possible downref: Non-RFC (?) normative reference: ref. '15' Summary: 9 errors (**), 0 flaws (~~), 7 warnings (==), 7 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 INTERNET-DRAFT Carsten Bormann 3 Expires: October 1999 Universitaet Bremen TZI 4 April 1999 6 Providing integrated services over low-bitrate links 7 draft-ietf-issll-isslow-05.txt 9 Status of this memo 11 This document is an Internet-Draft and is in full conformance with 12 all provisions of Section 10 of RFC2026. 14 Internet-Drafts are working documents of the Internet Engineering 15 Task Force (IETF), its areas, and its working groups. Note that 16 other groups may also distribute working documents as Internet- 17 Drafts. 19 Internet-Drafts are draft documents valid for a maximum of six months 20 and may be updated, replaced, or obsoleted by other documents at any 21 time. It is inappropriate to use Internet- Drafts as reference 22 material or to cite them other than as "work in progress." 24 The list of current Internet-Drafts can be accessed at 25 http://www.ietf.org/ietf/1id-abstracts.txt 27 The list of Internet-Draft Shadow Directories can be accessed at 28 http://www.ietf.org/shadow.html 30 Abstract 32 This document describes an architecture for providing integrated 33 services over low-bitrate links, such as modem lines, ISDN B- 34 channels, and sub-T1 links. It covers only the lower parts of the 35 Internet Multimedia Conferencing Architecture [1]; additional 36 components required for application services such as Internet 37 Telephony (e.g., a session initiation protocol) are outside the scope 38 of this document. The main components of the architecture are: a 39 real-time encapsulation format for asynchronous and synchronous low- 40 bitrate links, a header compression architecture optimized for real- 41 time flows, elements of negotiation protocols used between routers 42 (or between hosts and routers), and announcement protocols used by 43 applications to allow this negotiation to take place. 45 1. Introduction 47 As an extension to the ``best-effort'' services the Internet is well- 48 known for, additional types of services (``integrated services'') 49 that support the transport of real-time multimedia information are 50 being developed for, and deployed in the Internet. Important 51 elements of this development are: 53 - parameters for forwarding mechanisms that are appropriate for 54 real-time information [11, 12], 56 - a setup protocol that allows establishing special forwarding 57 treatment for real-time information flows (RSVP [4]), 59 - a transport protocol for real-time information (RTP/RTCP [6]). 61 In addition to these elements at the network and transport levels of 62 the Internet Multimedia Conferencing Architecture [1], further 63 components are required to define application services such as 64 Internet Telephony, e.g., protocols for session initiation and 65 control. These components are outside the scope of this document. 67 Up to now, the newly developed services could not (or only very 68 inefficiently) be used over forwarding paths that include low-bitrate 69 links such as 14.4, 33.6, and 56 kbit/s modems, 56 and 64 kbit/s ISDN 70 B-channels, or even sub-T1 links. The encapsulation formats used on 71 these links are not appropriate for the simultaneous transport of 72 arbitrary data and real-time information that has to meet stringent 73 delay requirements. Transmission of a 1500 byte packet on a 28.8 74 kbit/s modem link makes this link unavailable for the transmission of 75 real-time information for about 400 ms. This adds a worst-case delay 76 that causes real-time applications to operate with round-trip delays 77 on the order of at least a second -- unacceptable for real-time 78 conversation. In addition, the header overhead associated with the 79 protocol stacks used is prohibitive on low-bitrate links, where 80 compression down to a few dozen bytes per real-time information 81 packet is often desirable. E.g., the overhead of at least 44 82 (4+20+8+12) bytes for HDLC/PPP, IP, UDP, and RTP completely 83 overshadows typical audio payloads such as the 19.75 bytes needed for 84 a G.723.1 ACELP audio frame -- a 14.4 kbit/s link is completely 85 consumed by this header overhead alone at 40 real-time frames per 86 second total (i.e., at 25 ms packetization delay for one stream or 50 87 ms for two streams, with no space left for data, yet). While the 88 header overhead can be reduced by combining several real-time 89 information frames into one packet, this increases the delay incurred 90 while filling that packet and further detracts from the goal of real- 91 time transfer of multi-media information over the Internet. 93 This document describes an approach for addressing these problems. 94 The main components of the architecture are: 96 - a real-time encapsulation format for asynchronous and 97 synchronous low-bitrate links, 99 - a header compression architecture optimized for real-time flows, 101 - elements of negotiation protocols used between routers (or 102 between hosts and routers), and 104 - announcement protocols used by applications to allow this 105 negotiation to take place. 107 2. Design Considerations 109 The main design goal for an architecture that addresses real-time 110 multimedia flows over low-bitrate links is that of minimizing the 111 end-to-end delay. More specifically, the worst case delay (after 112 removing possible outliers, which are equivalent to packet losses 113 from an application point of view) is what determines the playout 114 points selected by the applications and thus the delay actually 115 perceived by the user. 117 In addition, any such architecture should obviously undertake every 118 attempt to maximize the bandwidth actually available to media data; 119 overheads must be minimized. 121 An important component of the integrated services architecture is the 122 provision of reservations for real-time flows. One of the problems 123 that systems on low-bitrate links (routers or hosts) face when 124 performing admission control for such reservations is that they must 125 translate the bandwidth requested in the reservation to the one 126 actually consumed on the link. Methods such as data compression 127 and/or header compression can reduce the requirements on the link, 128 but admission control can only make use of the reduced requirements 129 in its calculations if it has enough information about the data 130 stream to know how effective the compression will be. One goal of 131 the architecture therefore is to provide the integrated services 132 admission control with this information. A beneficial side effect 133 may be to allow the systems to perform better compression than would 134 be possible without this information. This may make it worthwhile to 135 provide this information even when it is not intended to make a 136 reservation for a real-time flow. 138 3. The Need for a Concerted Approach 140 Many technical approaches come to mind for addressing these problems, 141 in particular a new form of low-delay encapsulation to address delay 142 and header compression methods to address overhead. This section 143 shows that these techniques should be combined to solve the problem. 145 3.1. Real-Time Encapsulation 147 The purpose of defining a real-time link-layer encapsulation protocol 148 is to be able to introduce newly arrived real-time packets into the 149 link-layer data stream without having to wait for the currently 150 transmitted (possibly large) packet to end. Obviously, a real-time 151 encapsulation must be part of any complete solution as the problem of 152 delays induced by large frames on the link can only be solved on this 153 layer. 155 To be able to switch to a real-time packet quickly in an interface 156 driver, it is first necessary to identify packets that belong to 157 real-time flows. This can be done using a heuristic approach (e.g., 158 favor the transmission of highly periodic flows of small packets 159 transported in IP/UDP, or use the IP precedence fields in a specific 160 way defined within an organization). Preferably, one also could make 161 use of a protocol defined for identifying flows that require special 162 treatment, i.e. RSVP. Of the two service types defined for use with 163 RSVP now, the guaranteed service will only be available in certain 164 environments; for this and various other reasons, the service type 165 chosen for many adaptive audio/video applications will most likely be 166 the controlled-load service. Controlled-load does not provide 167 control parameters for target delay; thus it does not unambiguously 168 identify those packet streams that would benefit most from being 169 transported in a real-time encapsulation format. This calls for a 170 way to provide additional parameters in integrated services flow 171 setup protocols to control the real-time encapsulation. 173 Real-time encapsulation is not sufficient on its own, however: Even 174 if the relevant flows can be appropriately identified for real-time 175 treatment, most applications simply cannot operate properly on low- 176 bitrate links with the header overhead implied by the combination of 177 HDLC/PPP, IP, UDP, and RTP, i.e. they absolutely require header 178 compression. 180 3.2. Header Compression 182 Header compression can be performed in a variety of elements and at a 183 variety of levels in the protocol architecture. As many vendors of 184 Internet Telephony products for PCs ship applications, the approach 185 that is most obvious to them is to reduce overhead by performing 186 header compression at the application level, i.e. above transport 187 protocols such as UDP[1]. 189 Generally, header compression operates by installing state at both 190 ends of a path that allows the receiving end to reconstruct 191 information omitted at the sending end. Many good techniques for 192 header compression (RFC 1144, [2]) operate on the assumption that the 193 path will not reorder the frames generated. This assumption does not 194 hold for end-to-end compression; therefore additional overhead is 195 required for resequencing state changes and for compressed packets 196 making use of these state changes. 198 Assume that a very good application level header compression solution 199 for RTP flows could be able to save 11 out of the 12 bytes of an RTP 200 header [3]. Even this perfect solution only reduces the total header 201 _________________________ 202 [1] or actually by using a non-standard, efficiently coded head- 203 er in the first place. 205 overhead by 1/4. It would have to be deployed in all applications, 206 even those that operate on systems that are attached to higher- 207 bitrate links. 209 Because of this limited effectiveness, the AVT group that is 210 responsible for RTP within the IETF has decided to not further pursue 211 application level header compression. 213 For router and IP stack vendors, the obvious approach is to define 214 header compression that can be negotiated between peer routers. 216 Advanced header compression techniques now being defined in the IETF 217 [2] certainly can relieve the link from significant parts of the 218 IP/UDP overhead (i.e., most of 28 of the 44 bytes mentioned above). 220 One of the design principles of the new IP header compression 221 developed in conjunction with IPv6 is that it stops at layers the 222 semantics of which cannot be inferred from information in lower layer 223 (outer) headers. Therefore, this header compression technique alone 224 cannot compress the data that is contained within UDP packets. 226 Any additional header compression technique runs into a problem: If 227 it assumes specific application semantics (i.e., those of RTP and a 228 payload data format) based on heuristics, it runs the risk of being 229 triggered falsely and (e.g. in case of packet loss) reconstructing 230 packets that are catastrophically incorrect for the application 231 actually being used. A header compression technique that can be 232 operated based on heuristics but does not cause incorrect 233 decompression even if the heuristics failed is described in [7]; a 234 companion document describes the mapping of this technique to PPP 235 [10]. 237 With all of these techniques, the total IP/UDP/RTP header overhead 238 for an audio stream can be reduced to two bytes per packet. This 239 technology need only be deployed at bottleneck links; high-speed 240 links can transfer the real-time streams without routers or switches 241 expending CPU cycles to perform header compression. 243 4. Principles of Real-Time Encapsulation for Low-Bitrate Links 245 The main design goal for a real-time encapsulation is to minimize the 246 delay incurred by real-time packets that become available for sending 247 while a long data packet is being sent. To achieve this, the 248 encapsulation must be able to either abort or suspend the transfer of 249 the long data packet. As an additional goal is to minimize the 250 overhead required for the transmission of packets from periodic 251 flows, this strongly argues for being able to suspend a packet, i.e. 252 segment it into parts between which the real-time packets can be 253 transferred. 255 4.1. Using existing IP fragmentation 257 Transmitting only part of a packet, to allow higher-priority traffic 258 to intervene and then resuming its transmission later on, is a kind 259 of fragmentation. Fragmentation is an existing functionality of the 260 IP layer: An IPv4 header already contains fields that allow a large 261 IP datagram to be fragmented into small parts. A sender's ``real- 262 time PPP'' implementation might simply indicate a small MTU to its IP 263 stack and thus cause all larger datagrams to be fragmented down to a 264 size that allows the access delay goals to be met[2]. (Also, a PPP 265 implementation can negotiate down the MTU of its peer, causing the 266 peer to fragment to a small size, which might be considered a crude 267 form of negotiating an access delay goal with the peer system -- if 268 that system supports priority queueing at the fragment level.) 270 Unfortunately, a full, 20 byte IP header is needed for each fragment 271 (larger when IP options are used). This limits the minimum size of 272 fragments that can be used without too much overhead. (Also, the 273 size of non-final fragments must be a multiple of 8 bytes, further 274 limiting the choice.) With path MTU discovery, IP level 275 fragmentation causes TCP implementations to use small MSSs -- this 276 further increases the per-packet overhead to 40 bytes per fragment. 278 In any case, fragmentation at the IP level persists on the path 279 further down to the datagram receiver, increasing the transmission 280 overheads and router load throughout the network. With its high 281 overhead and the adverse effect on the Internet, IP level 282 fragmentation can only be a stop-gap mechanism when no other 283 fragmentation protocol is available in the peer implementation. 285 4.2. Link-Layer Mechanisms 287 Cell-oriented multiplexing techniques such as ATM that introduce 288 regular points where cells from a different packet can be 289 interpolated are too inefficient for low-bitrate links; also, they 290 are not supported by chips used to support the link layer in low- 291 bitrate routers and host interfaces. 293 Instead, the real-time encapsulation should as far as possible make 294 use of the capabilities of the chips that have been deployed. On 295 synchronous lines, these chips support HDLC framing; on asynchronous 296 lines, an asynchronous variant of HDLC that usually is implemented in 297 software is being used. Both variants of HDLC provide a delimiting 298 mechanism to indicate the end of a frame over the link. The obvious 299 solution to the segmentation problem is to combine this mechanism 300 with an indication of whether the delimiter terminates or suspends 301 the current packet. 303 This indication could be in an octet appended to each frame 304 _________________________ 305 [2] This assumes that the IP stack is able to priority-tag frag- 306 ments, or that the PPP implementation is able to correlate the 307 fragments to the initial one that carries the information relevant 308 for prioritizing, or that only initial fragments can be high- 309 priority. 311 information field; however, seven out of eight bits of the octet 312 would be wasted. Instead, the bit could be carried at the start of 313 the next frame in conjunction with multiplexing information (PPP 314 protocol identifier etc.) that will be required here anyway. Since 315 the real-time flows will in general be periodic, this multiplexing 316 information could convey (part of) the compressed form of the header 317 for the packet. If packets from the real-time flow generally are of 318 constant length (or have a defined maximum length that is often 319 used), the continuation of the suspended packet could be immediately 320 attached to it, without expending a further frame delimiter, i.e., 321 the interpolation of the real-time packet would then have zero 322 overhead. Since packets from low-delay real-time flows generally 323 will not require the ability to be further suspended, the 324 continuation bit could be reserved for the non-real-time packet 325 stream. 327 One real-time encapsulation format with these (and other) functions 328 is described in ITU-T H.223 [13], the multiplex used by the H.324 329 modem-based videophone standard [14]. It was investigated whether 330 compatibility could be achieved with this specification, which will 331 be used in future videophone-enabled (H.324 capable) modems. 332 However, since the multiplexing capabilities of H.223 are limited to 333 15 schedules (definitions of sequences of packet types that can be 334 identified in a multiplex header), for general Internet usage a 335 superset or a more general encapsulation would have been required. 336 Also, a PPP-style negotiation protocol was needed instead of using 337 (and necessarily extending) ITU-T H.245 [15] for setting the 338 parameters of the multiplex. In the PPP context, the interactions 339 with the encapsulations for data compression and link layer 340 encryption needed to be defined (including operation in the presence 341 of padding). But most important, H.223 requires synchronous HDLC 342 chips that can be configured to send frames without an attached CRC, 343 which is not possible with all chips deployed in commercially 344 available routers; so complete compatibility was unachievable. 346 Instead of adopting H.223, it was decided to pursue an approach that 347 is oriented towards compatibility both with existing hardware and 348 existing software (in particular PPP) implementations. The next 349 subsection groups these implementations according to their 350 capabilities. 352 4.3. Implementation models 354 This section introduces a number of terms for types of 355 implementations that are likely to emerge. It is important to have 356 these different implementation models in mind as there is no single 357 approach that fits all models best. 359 4.3.1. Sender types 361 There are two fundamental approaches to real-time transmission on 362 low-bitrate links: 364 Sender type 1 365 The PPP real-time framing implementation is able to control the 366 transmission of each byte being transmitted with some known, 367 bounded delay (e.g., due to FIFOs). For example, this is 368 generally true of PC host implementations, which directly access 369 serial interface chips byte by byte or by filling a very small 370 FIFO. For type 1 senders, a suspend/resume type approach will 371 be typically used: When a long frame is to be sent, the attempt 372 is to send it undivided; only if higher priority packets come up 373 during the transmission will the lower-priority long frame be 374 suspended and later resumed. This approach allows the minimum 375 variation in access delay for high-priority packets; also, 376 fragmentation overhead is only incurred when actually needed. 378 Sender type 2 379 With type 2 senders, the interface between the PPP real-time 380 framing implementation and the transmission hardware is not in 381 terms of streams of bytes, but in terms of frames, e.g., in the 382 form of multiple (prioritized) send queues directly supported by 383 hardware. This is often true of router systems for synchronous 384 links, in particular those that have to support a large number 385 of low-bitrate links. As type 2 senders have no way to suspend 386 a frame once it has been handed down for transmission, they 387 typically will use a queues-of-fragments approach, where long 388 packets are always split into units that are small enough to 389 maintain the access delay goals for higher-priority traffic. 390 There is a trade-off between the variation in access delay 391 resulting from a large fragment size and the overhead that is 392 incurred for every long packet by choosing a small fragment 393 size. 395 4.3.2. Receiver types 397 Although the actual work of formulating transmission streams for 398 real-time applications is performed at the sender, the ability of the 399 receiver to immediately make use of the information received depends 400 on its characteristics: 402 Receiver type 1 403 Type 1 receivers have full control over the stream of bytes 404 received within PPP frames, i.e., bytes received are available 405 immediately to the PPP real-time framing implementation (with 406 some known, bounded delay e.g. due to FIFOs etc.). 408 Receiver type 2 409 With type 2 receivers, the PPP real-time framing implementation 410 only gets hold of a frame when it has been received completely, 411 i.e., the final flag has been processed (typically by some HDLC 412 chip that directly fills a memory buffer). 414 4.4. Conclusion 416 As a result of the diversity in capabilities of current 417 implementations, there are now two specifications for real-time 418 encapsulation: One, the multi-class extension to the PPP multi-link 419 protocol, is providing the solution for the queues-of-fragments 420 approach by extending the single-stream PPP multi-link protocol by 421 multiple classes [8]. The other encapsulation, PPP in a real-time 422 oriented HDLC-like framing, builds on this specification end extends 423 it by a way to dynamically delimit multiple fragments within one HDLC 424 frame [9], providing the solution for the suspend/resume type 425 approach. 427 5. Principles of Header Compression for Real-Time Flows 429 A good baseline for a discussion about header compression is in the 430 new IP header compression specification that was designed in 431 conjunction with the development of IPv6 [2]. The techniques used 432 there can reduce the 28 bytes of IPv4/UDP header to about 6 bytes 433 (depending on the number of concurrent streams); with the remaining 4 434 bytes of HDLC/PPP overhead and 12 bytes for RTP the total header 435 overhead can be about halved but still exceeds the size of a G.723.1 436 ACELP frame. Note that, in contrast to IP header compression, the 437 environment discussed here assumes the existence of a full-duplex PPP 438 link and thus can rely on negotiation where IP header compression 439 requires repeated transmission of the same information. (The use of 440 the architecture of the present document with link layer multicasting 441 has not yet been examined.) 443 Additional design effort was required for RTP header compression. 444 Applying the concepts of IP header compression, of the (at least) 12 445 bytes in an RTP header, 7 bytes (timestamp, sequence, and marker bit) 446 would qualify as RANDOM; DELTA encoding cannot generally be used 447 without further information since the lower layer header does not 448 unambiguously identify the semantics and there is no TCP checksum 449 that can be relied on to detect incorrect decompression. Only a more 450 semantics-oriented approach can provide better compression (just as 451 RFC 1144 can provide very good compression of TCP headers by making 452 use of semantic knowledge of TCP and its checksumming method). 454 For RTP packets, differential encoding of the sequence number and 455 timestamps is an efficient approach for certain cases of payload data 456 formats. E.g., speech flows generally have sequence numbers and 457 timestamp fields that increase by 1 and by the frame size in 458 timestamp units, resp.; the CRTP (compressed RTP) specification makes 459 use of this relationship by encoding these fields only when the 460 second order difference is non-zero [7]. 462 6. Announcement Protocols Used by Applications 464 As argued, the compressor can operate best if it can make use of 465 information that clearly identifies real-time streams and provides 466 information about the payload data format in use. 468 If these systems are routers, this consent must be installed as 469 router state; if these systems are hosts, it must be known to their 470 networking kernels. Sources of real-time information flows are 471 already describing characteristics of these flows to their kernels 472 and to the routers in the form of TSpecs in RSVP PATH messages [4]. 473 Since these messages make use of the router alert option, they are 474 seen by all routers on the path; path state about the packet stream 475 is normally installed at each of these routers that implement RSVP. 476 Additional RSVP objects could be defined that are included in PATH 477 messages by those applications that desire good performance over low- 478 bitrate links; these objects would be coded to be ignored by routers 479 that are not interested in them (class number 11bbbbbb). 481 Note that the path state is available in the routers even when no 482 reservation is made; this allows informed compression of best-effort 483 traffic. It is not quite clear, though, how path state could be 484 teared down quickly when a source ceases to transmit. 486 7. Elements of Hop-By-Hop Negotiation Protocols 488 The IP header compression specification attempts to account for 489 simplex and multicast links by providing information about the 490 compressed streams only in the forward direction. E.g., a full 491 IP/UDP header must be sent after F_MAX_TIME (currently 3 seconds), 492 which is a negligible total overhead (e.g. one full header every 150 493 G.723.1 packets), but must be considered carefully in scheduling the 494 real-time transmissions. Both simplex and multicast links are not 495 prevailing in the low-bitrate environment (although multicast 496 functionality may become more important with wireless systems); in 497 this document, we therefore assume full-duplex capability. 499 As compression techniques will improve, a negotiation between the two 500 peers on the link would provide the best flexibility in 501 implementation complexity and potential for extensibility. The peer 502 routers/hosts can decide which real-time packet streams are to be 503 compressed, which header fields are not to be sent at all, which 504 multiplexing information should be used on the link, and how the 505 remaining header fields should be encoded. PPP, a well-tried suite 506 of negotiation protocols, is already used on most of the low-bitrate 507 links and seems to provide the obvious approach. Cooperation from 508 PPP is also needed to negotiate the use of real-time encapsulations 509 between systems that are not configured to automatically do so. 510 Therefore, PPP options that can be negotiated at the link setup (LCP) 511 phase are included in [8], [9], and [10]. 513 8. Security Considerations 515 Header compression protocols that make use of assumptions about 516 application protocols need to be carefully analyzed whether it is 517 possible to subvert other applications by maliciously or 518 inadvertently enabling their use. 520 It is generally not possible to do significant hop-by-hop header 521 compression on encrypted streams. With certain security policies, it 522 may be possible to run an encrypted tunnel to a network access server 523 that does header compression on the decapsulated packets and sends 524 them over an encrypted link encapsulation; see also the short mention 525 of interactions between real-time encapsulation and encryption in 526 section 4 above. If the security requirements permit, a special RTP 527 payload data format that encrypts only the data may preferably be 528 used. 530 9. References 532 [1] Mark Handley, Jon Crowcroft, Carsten Bormann, Joerg Ott. ``The 533 Internet Multimedia Conferencing Architecture'', Work in 534 Progress (draft-ietf-mmusic-confarch-00.txt), July 1997. 536 [2] M. Degermark, B. Nordgren, S. Pink, ``IP Header Compression'', 537 RFC 2507, February 1999. 539 [3] Scott Petrack, Ed Ellesson, ``Framework for C/RTP: Compressed 540 RTP Using Adaptive Differential Header Compression'', 541 contribution to the mailing list rem-conf@es.net, February 1996. 543 [4] R. Braden, Ed., L. Zhang, S. Berson, S. Herzog, S. Jamin, 544 ``Resource ReSerVation Protocol (RSVP) -- Version 1 Functional 545 Specification'', RFC 2205, September 1997. 547 [5] K. Sklower, B. Lloyd, G. McGregor, D. Carr, T. Coradetti, ``The 548 PPP Multilink Protocol (MP)'', RFC 1990, August 1996 (obsoletes 549 RFC1717). 551 [6] H. Schulzrinne, S. Casner, R. Frederick, V. Jacobson, ``RTP: A 552 Transport Protocol for Real-Time Applications'', RFC 1889, 553 January 1996. 555 [7] S. Casner, V. Jacobson, ``Compressing IP/UDP/RTP Headers for 556 Low-Speed Serial Links'', RFC 2508, February 1999. 558 [8] C. Bormann, ``The Multi-Class Extension to Multi-Link PPP'', 559 Work in Progress (draft-ietf-issll-isslow-mcml-05.txt), August 560 1998. 562 [9] C. Bormann, ``PPP in a real-time oriented HDLC-like framing'', 563 Work in Progress (draft-ietf-issll-isslow-rtf-04.txt), August 564 1998. 566 [10] M. Engan, S. Casner, C. Bormann, ``IP Header Compression over 567 PPP'', RFC 2509, February 1999. 569 [11] J. Wroclawski. ``Specification of the Controlled-Load Network 570 Element Service'', RFC 2211, September 1997. 572 [12] S. Shenker, C. Partridge, R. Guerin. ``Specification of 573 Guaranteed Quality of Service'', RFC 2212, September 1997. 575 [13] ITU-T Recommendation H.223, ``Multiplexing protocol for low bit 576 rate multimedia communication'', International Telecommunication 577 Union, Telecommunication Standardization Sector (ITU-T), March 578 1996. 580 [14] ITU-T Recommendation H.324, ``Terminal for low bit rate 581 multimedia communication'', International Telecommunication 582 Union, Telecommunication Standardization Sector (ITU-T), March 583 1996. 585 [15] ITU-T Recommendation H.245, ``Control protocol for multimedia 586 communication'', International Telecommunication Union, 587 Telecommunication Standardization Sector (ITU-T), March 1996. 589 10. Author's Address 591 Carsten Bormann 592 Universitaet Bremen FB3 TZI 593 Postfach 330440 594 D-28334 Bremen, GERMANY 595 cabo@tzi.org 596 phone +49.421.218-7024 597 fax +49.421.218-7000 599 Acknowledgements 601 Much of the early discussion that led to this document was done with 602 Scott Petrack and Cary Fitzgerald. Steve Casner, Mikael Degermark, 603 Steve Jackowski, Dave Oran, the other members of the ISSLL subgroup 604 on low bitrate links (ISSLOW), and in particular the ISSLL WG co- 605 chairs Eric Crawley and John Wroclawski have helped in making this 606 architecture a reality.