idnits 2.17.1 draft-ietf-rmcat-wireless-tests-06.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 == Line 287 has weird spacing: '...nternet fix...' -- The document date (December 22, 2018) is 1951 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Missing Reference: '2D' is mentioned on line 303, but not defined == Missing Reference: '3D' is mentioned on line 303, but not defined == Missing Reference: 'Uplink' is mentioned on line 340, but not defined == Missing Reference: 'Downlink' is mentioned on line 340, but not defined == Missing Reference: 'TBD' is mentioned on line 715, but not defined == Unused Reference: 'I-D.ietf-rmcat-cc-requirements' is defined on line 918, but no explicit reference was found in the text == Outdated reference: A later version (-14) exists of draft-ietf-rmcat-eval-criteria-08 == Outdated reference: A later version (-10) exists of draft-ietf-rmcat-eval-test-08 Summary: 0 errors (**), 0 flaws (~~), 10 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group Z. Sarker 3 Internet-Draft I. Johansson 4 Intended status: Informational Ericsson AB 5 Expires: June 25, 2019 X. Zhu 6 J. Fu 7 W. Tan 8 M. Ramalho 9 Cisco Systems 10 December 22, 2018 12 Evaluation Test Cases for Interactive Real-Time Media over Wireless 13 Networks 14 draft-ietf-rmcat-wireless-tests-06 16 Abstract 18 The Real-time Transport Protocol (RTP) is used for interactive 19 multimedia communication applications. These applications are 20 typically required to implement congestion control. To ensure 21 seamless and robust user experience, a well-designed RTP-based 22 congestion control algorithm should work well across all access 23 network types. This document describes test cases for evaluating 24 performances of such congestion control algorithms over LTE and Wi-Fi 25 networks. 27 Status of This Memo 29 This Internet-Draft is submitted in full conformance with the 30 provisions of BCP 78 and BCP 79. 32 Internet-Drafts are working documents of the Internet Engineering 33 Task Force (IETF). Note that other groups may also distribute 34 working documents as Internet-Drafts. The list of current Internet- 35 Drafts is at https://datatracker.ietf.org/drafts/current/. 37 Internet-Drafts are draft documents valid for a maximum of six months 38 and may be updated, replaced, or obsoleted by other documents at any 39 time. It is inappropriate to use Internet-Drafts as reference 40 material or to cite them other than as "work in progress." 42 This Internet-Draft will expire on June 25, 2019. 44 Copyright Notice 46 Copyright (c) 2018 IETF Trust and the persons identified as the 47 document authors. All rights reserved. 49 This document is subject to BCP 78 and the IETF Trust's Legal 50 Provisions Relating to IETF Documents 51 (https://trustee.ietf.org/license-info) in effect on the date of 52 publication of this document. Please review these documents 53 carefully, as they describe your rights and restrictions with respect 54 to this document. Code Components extracted from this document must 55 include Simplified BSD License text as described in Section 4.e of 56 the Trust Legal Provisions and are provided without warranty as 57 described in the Simplified BSD License. 59 Table of Contents 61 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 62 2. Terminologies . . . . . . . . . . . . . . . . . . . . . . . . 3 63 3. Cellular Network Specific Test Cases . . . . . . . . . . . . 3 64 3.1. Varying Network Load . . . . . . . . . . . . . . . . . . 6 65 3.1.1. Network Connection . . . . . . . . . . . . . . . . . 6 66 3.1.2. Simulation Setup . . . . . . . . . . . . . . . . . . 7 67 3.2. Bad Radio Coverage . . . . . . . . . . . . . . . . . . . 8 68 3.2.1. Network connection . . . . . . . . . . . . . . . . . 9 69 3.2.2. Simulation Setup . . . . . . . . . . . . . . . . . . 9 70 3.3. Desired Evaluation Metrics for cellular test cases . . . 10 71 4. Wi-Fi Networks Specific Test Cases . . . . . . . . . . . . . 10 72 4.1. Bottleneck in Wired Network . . . . . . . . . . . . . . . 12 73 4.1.1. Network topology . . . . . . . . . . . . . . . . . . 12 74 4.1.2. Test setup . . . . . . . . . . . . . . . . . . . . . 13 75 4.1.3. Typical test scenarios . . . . . . . . . . . . . . . 14 76 4.1.4. Expected behavior . . . . . . . . . . . . . . . . . . 14 77 4.2. Bottleneck in Wi-Fi Network . . . . . . . . . . . . . . . 15 78 4.2.1. Network topology . . . . . . . . . . . . . . . . . . 15 79 4.2.2. Test setup . . . . . . . . . . . . . . . . . . . . . 15 80 4.2.3. Typical test scenarios . . . . . . . . . . . . . . . 16 81 4.2.4. Expected behavior . . . . . . . . . . . . . . . . . . 17 82 4.3. Other Potential Test Cases . . . . . . . . . . . . . . . 18 83 4.3.1. EDCA/WMM usage . . . . . . . . . . . . . . . . . . . 18 84 4.3.2. Effects of Legacy 802.11b Devices . . . . . . . . . . 18 85 5. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . 18 86 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 18 87 7. Security Considerations . . . . . . . . . . . . . . . . . . . 19 88 8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 19 89 9. References . . . . . . . . . . . . . . . . . . . . . . . . . 19 90 9.1. Normative References . . . . . . . . . . . . . . . . . . 19 91 9.2. Informative References . . . . . . . . . . . . . . . . . 20 92 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 21 94 1. Introduction 96 Wireless networks (both cellular and Wi-Fi [IEEE802.11] local area 97 network) are an integral part of the Internet. Mobile devices 98 connected to the wireless networks generate huge amount of media 99 traffic in the Internet. Application scenarios range from users 100 having a video call in the bus to media consumption by someone 101 sitting on a living room couch. It is well known that the 102 characteristics and technical challenges for offering multimedia 103 services over wireless are very different from those of providing the 104 same service over a wired network. Even though RMCAT basic test 105 cases as defined in [I-D.ietf-rmcat-eval-test] have covered many 106 effects of the impairments also visible in wireless networks, there 107 remains characteristics and dynamics unique to a given wireless 108 environment. For example, in LTE networks the base station maintains 109 queues per radio bearer per user hence it leads to a different nature 110 of interaction from that over the wired network, where traffic from 111 all users share the same queue. Furthermore, user mobility in a 112 cellular network is different than user mobility in a Wi-Fi network. 113 Therefore, It is important to evaluate performance of the proposed 114 RMCAT candidate solutions separately over cellular mobile networks 115 and over Wi-Fi local networks (i.e., IEEE 802.11xx protocol family ). 117 RMCAT evaluation criteria [I-D.ietf-rmcat-eval-criteria] document 118 provides the guideline for evaluating candidate algorithms and 119 recognizes the importance of testing over wireless access networks. 120 However, it does not describe any specific test cases for evaluating 121 performance of the candidate algorithm. This document describes test 122 cases specifically targeting cellular networks such as LTE networks 123 and Wi-Fi local networks. 125 2. Terminologies 127 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 128 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 129 "OPTIONAL" in this document are to be interpreted as described in BCP 130 14 [RFC2119] [RFC8174] when, and only when, they appear in all 131 capitals, as shown here. 133 3. Cellular Network Specific Test Cases 135 A cellular environment is more complicated than a wireline ditto 136 since it seeks to provide services in the context of variable 137 available bandwidth, location dependencies and user mobilities at 138 different speeds. In a cellular network the user may reach the cell 139 edge which may lead to a significant amount of retransmissions to 140 deliver the data from the base station to the destination and vice 141 versa. These network links or radio links will often act as a 142 bottleneck for the rest of the network which will eventually lead to 143 excessive delays or packet drops. An efficient retransmission or 144 link adaptation mechanism can reduce the packet loss probability but 145 there will still be some packet losses and delay variations. 146 Moreover, with increased cell load or handover to a congested cell, 147 congestion in transport network will become even worse. Besides, 148 there are certain characteristics which make the cellular network 149 different and challenging than other types of access network such as 150 Wi-Fi and wired network. In a cellular network - 152 o The bottleneck is often a shared link with relatively few users. 154 * The cost per bit over the shared link varies over time and is 155 different for different users. 157 * Left over/ unused resource can be grabbed by other greedy 158 users. 160 o Queues are always per radio bearer hence each user can have many 161 of such queues. 163 o Users can experience both Inter and Intra Radio Access Technology 164 (RAT) handovers ("handover" definition in [HO-def-3GPP] ). 166 o Handover between cells, or change of serving cells (see in 167 [HO-LTE-3GPP] and [HO-UMTS-3GPP] ) might cause user plane 168 interruptions which can lead to bursts of packet losses, delay 169 and/or jitter. The exact behavior depends on the type of radio 170 bearer. Typically, the default best effort bearers do not 171 generate packet loss, instead packets are queued up and 172 transmitted once the handover is completed. 174 o The network part decides how much the user can transmit. 176 o The cellular network has variable link capacity per user 178 * Can vary as fast as a period of milliseconds. 180 * Depends on lots of facts (such as distance, speed, 181 interference, different flows). 183 * Uses complex and smart link adaptation which makes the link 184 behavior ever more dynamic. 186 * The scheduling priority depends on the estimated throughput. 188 o Both Quality of Service (QoS) and non-QoS radio bearers can be 189 used. 191 Hence, a real-time communication application operating in such a 192 cellular network need to cope with shared bottleneck link and 193 variable link capacity, event likes handover, non-congestion related 194 loss, abrupt change in bandwidth (both short term and long term) due 195 to handover, network load and bad radio coverage. Even though 3GPP 196 define QoS bearers [QoS-3GPP] to ensure high quality user experience, 197 adaptive real-time applications are desired. 199 Different mobile operators deploy their own cellular network with 200 their own set of network functionalities and policies. Usually, a 201 mobile operator network includes 2G, EDGE, 3G and 4G radio access 202 technologies. Looking at the specifications of such radio 203 technologies it is evident that only 3G and 4G radio technologies can 204 support the high bandwidth requirements from real-time interactive 205 video applications. The future real-time interactive application 206 will impose even greater demand on cellular network performance which 207 makes 4G (and beyond radio technologies) more suitable access 208 technology for such genre of application. 210 The key factors to define test cases for cellular network are 212 o Shared and varying link capacity 214 o Mobility 216 o Handover 218 However, for cellular network it is very hard to separate such events 219 from one another as these events are heavily related. Hence instead 220 of devising separate test cases for all those important events we 221 have divided the test case in two categories. It should be noted 222 that in the following test cases the goal is to evaluate the 223 performance of candidate algorithms over radio interface of the 224 cellular network. Hence it is assumed that the radio interface is 225 the bottleneck link between the communicating peers and that the core 226 network does not add any extra congestion in the path. Also the 227 combination of multiple access technologies such as one user has LTE 228 connection and another has Wi-Fi connection is kept out of the scope 229 of this document. However, later those additional scenarios can also 230 be added in this list of test cases. While defining the test cases 231 we assumed a typical real-time telephony scenario over cellular 232 networks where one real-time session consists of one voice stream and 233 one video stream. We recommend that an LTE network simulator is used 234 for the test cases defined in this document, for example-NS-3 LTE 235 simulator [LTE-simulator]. 237 3.1. Varying Network Load 239 The goal of this test is to evaluate the performance of the candidate 240 congestion control algorithm under varying network load. The network 241 load variation is created by adding and removing network users a.k.a. 242 User Equipments (UEs) during the simulation. In this test case, each 243 of the user/UE in the media session is an RMCAT compliant endpoint. 244 The arrival of users follows a Poisson distribution, which is 245 proportional to the length of the call, so that the number of users 246 per cell is kept fairly constant during the evaluation period. At 247 the beginning of the simulation there should be enough amount of time 248 to warm-up the network. This is to avoid running the evaluation in 249 an empty network where network nodes are having empty buffers, low 250 interference at the beginning of the simulation. This network 251 initialization period is therefore excluded from the evaluation 252 period. 254 This test case also includes user mobility and competing traffic. 255 The competing traffics includes both same kind of flows (with same 256 adaptation algorithms) and different kind of flows (with different 257 service and congestion control). The investigated congestion control 258 algorithms should show maximum possible network utilization and 259 stability in terms of rate variations, lowest possible end to end 260 frame latency, network latency and Packet Loss Rate (PLR) at 261 different cell load level. 263 3.1.1. Network Connection 265 Each mobile user is connected to a fixed user. The connection 266 between the mobile user and fixed user consists of a LTE radio 267 access, an Evolved Packet Core (EPC) and an Internet connection. The 268 mobile user is connected to the EPC using LTE radio access technology 269 which is further connected to the Internet. The fixed user is 270 connected to the Internet via wired connection with no bottleneck 271 (practically infinite bandwidth). The Internet and wired connection 272 in this setup does not add any network impairments to the test, it 273 only adds 10ms of one-way transport propagation delay. 275 The path from the fixed user to mobile user is defines as "Downlink" 276 and the path from mobile user to the fixed user is defined as 277 "Uplink". We assume that only uplink or downlink is congested for 278 the mobile users. Hence, we recommend that the uplink and downlink 279 simulations are run separately. 281 uplink 282 ++))) +--------------------------> 283 ++-+ ((o)) 284 | | / \ +-------+ +------+ +---+ 285 +--+ / \----+ +-----+ +----+ | 286 / \ +-------+ +------+ +---+ 287 UE BS EPC Internet fixed 288 <--------------------------+ 289 downlink 291 Figure 1: Simulation Topology 293 3.1.2. Simulation Setup 295 The values enclosed within " [ ] " for the following simulation 296 attributes follow the notion set in [I-D.ietf-rmcat-eval-test]. The 297 desired simulation setup as follows- 299 1. Radio environment 301 A. Deployment and propagation model : 3GPP case 1[Deployment] 303 B. Antenna: Multiple-Input and Multiple-Output (MIMO), [2D, 3D] 305 C. Mobility: [3km/h, 30km/h] 307 D. Transmission bandwidth: 10Mhz 309 E. Number of cells: multi cell deployment (3 Cells per Base 310 Station (BS) * 7 BS) = 21 cells 312 F. Cell radius: 166.666 Meters 314 G. Scheduler: Proportional fair with no priority 316 H. Bearer: Default bearer for all traffic. 318 I. Active Queue Management (AQM) settings: AQM [on,off] 320 2. End to end Round Trip Time (RTT): [ 40, 150] 322 3. User arrival model: Poisson arrival model 324 4. User intensity: 326 * Downlink user intensity: {0.7, 1.4, 2.1, 2.8, 3.5, 4.2, 4.9, 327 5.6, 6.3, 7.0, 7.7, 8.4, 9,1, 9.8, 10.5} 329 * Uplink user intercity : {0.7, 1.4, 2.1, 2.8, 3.5, 4.2, 4.9, 330 5.6, 6.3, 7.0} 332 5. Simulation duration: 91s 334 6. Evaluation period : 30s-60s 336 7. Media traffic 338 1. Media type: Video 340 a. Media direction: [Uplink, Downlink] 342 b. Number of Media source per user: One (1) 344 c. Media duration per user: 30s 346 d. Media source: same as define in section 4.3 of 347 [I-D.ietf-rmcat-eval-test] 349 2. Media Type : Audio 351 a. Media direction: Uplink and Downlink 353 b. Number of Media source per user: One (1) 355 c. Media duration per user: 30s 357 d. Media codec: Constant BitRate (CBR) 359 e. Media bitrate : 20 Kbps 361 f. Adaptation: off 363 8. Other traffic model: 365 * Downlink simulation: Maximum of 4Mbps/cell (web browsing or 366 FTP traffic) 368 * Unlink simulation: Maximum of 2Mbps/cell (web browsing or FTP 369 traffic) 371 3.2. Bad Radio Coverage 373 The goal of this test is to evaluate the performance of candidate 374 congestion control algorithm when users visit part of the network 375 with bad radio coverage. The scenario is created by using larger 376 cell radius than previous test case. In this test case each of the 377 user/UE in the media session is an RMCAT compliant endpoint. The 378 arrival of users follows a Poisson distribution, which is 379 proportional to the length of the call, so that the number of users 380 per cell is kept fairly constant during the evaluation period. At 381 the beginning of the simulation there should be enough amount of time 382 to warm-up the network. This is to avoid running the evaluation in 383 an empty network where network nodes are having empty buffers, low 384 interference at the beginning of the simulation. This network 385 initialization period is therefore excluded from the evaluation 386 period. 388 This test case also includes user mobility and competing traffic. 389 The competing traffics includes same kind of flows (with same 390 adaptation algorithms) . The investigated congestion control 391 algorithms should show maximum possible network utilization and 392 stability in terms of rate variations, lowest possible end to end 393 frame latency, network latency and Packet Loss Rate (PLR) at 394 different cell load level. 396 3.2.1. Network connection 398 Same as defined in Section 3.1.1 400 3.2.2. Simulation Setup 402 The desired simulation setup is same as Varying Network Load test 403 case defined in Section 3.1 except following changes- 405 1. Radio environment : Same as defined in Section 3.1.2 except 406 followings 408 A. Deployment and propagation model : 3GPP case 3[Deployment] 410 B. Cell radius: 577.3333 Meters 412 C. Mobility: 3km/h 414 2. User intensity = {0.7, 1.4, 2.1, 2.8, 3.5, 4.2, 4.9, 5.6, 6.3, 415 7.0} 417 3. Media traffic model: Same as defined in Section 3.1.2 419 4. Other traffic model: None 421 3.3. Desired Evaluation Metrics for cellular test cases 423 RMCAT evaluation criteria document [I-D.ietf-rmcat-eval-criteria] 424 defines metrics to be used to evaluate candidate algorithms. 425 However, looking at the nature and distinction of cellular networks 426 we recommend at minimum following metrics to be used to evaluate the 427 performance of the candidate algorithms for the test cases defined in 428 this document. 430 The desired metrics are- 432 o Average cell throughput (for all cells), shows cell utilizations. 434 o Application sending and receiving bitrate, goodput. 436 o Packet Loss Rate (PLR). 438 o End to end Media frame delay. For video, this means the delay 439 from capture to display. 441 o Transport delay. 443 o Algorithm stability in terms of rate variation. 445 4. Wi-Fi Networks Specific Test Cases 447 Given the prevalence of Internet access links over Wi-Fi, it is 448 important to evaluate candidate RMCAT congestion control solutions 449 over test cases that include Wi-Fi access lines. Such evaluations 450 should also highlight the inherent different characteristics of Wi-Fi 451 networks in contrast to wired networks: 453 o The wireless radio channel is subject to interference from nearby 454 transmitters, multipath fading, and shadowing, causing 455 fluctuations in link throughput and sometimes an error-prone 456 communication environment 458 o Available network bandwidth is not only shared over the air 459 between cocurrent users, but also between uplink and downlink 460 traffic due to the half duplex nature of wireless transmission 461 medium. 463 o Packet transmissions over Wi-Fi are susceptible to contentions and 464 collisions over the air. Consequently, traffic load beyond a 465 certain utilization level over a Wi-Fi network can introduce 466 frequent collisions and significant network overhead. This, in 467 turn, leads to excessive delay, retransmissions, packet losses and 468 lower effective bandwidth for applications. 470 o The IEEE 802.11 standard (i.e., Wi-Fi) supports multi-rate 471 transmission capabilities by dynamically choosing the most 472 appropriate modulation scheme for a given received singal 473 strength. A different choice of physical-layer rate leads to 474 different application-layer throughput. 476 o Presence of legancy 802.11b networks can significantly slow down 477 the the rest of a modern Wi-Fi Network, since it takes longer to 478 transmit the same packet over a slower link than over a faster 479 link. [Editor's note: maybe include a reference here instead.] 481 o Handover from one Wi-Fi Access Point (AP) to another may lead to 482 packet delay and losses during the process. 484 o IEEE 802.11e defined EDCA/WMM (Enhanced DCF Channel Access/Wi-Fi 485 Multi-Media) to give voice and video streams higher priority over 486 pure data applications (e.g., file transfers). 488 In summary, presence of Wi-Fi access links in different network 489 topologies can exert different impact on the network performance in 490 terms of application-layer effective throughput, packet loss rate, 491 and packet delivery delay. These, in turn, influence the behavior of 492 end-to-end real-time multimedia congestion control. 494 Throughout this draft, unless otherwise mentioned, test cases are 495 described using 802.11n due to its wide availability in real-world 496 networks. Statistics collected from enterprise Wi-Fi networks show 497 that the dominant physical modes are 802.11n and 802.11ac, accounting 498 for 73.6% and 22.5% of enterprise network users, respectively. 500 Typically, a Wi-Fi access network connects to a wired infrastructure. 501 Either the wired or the Wi-Fi segment of the network could be the 502 bottleneck. In the following sections, we describe basic test cases 503 for both scenarios separately. The same set of performance metrics 504 as in [I-D.ietf-rmcat-eval-test]) should be collected for each test 505 case. 507 While all test cases described below can be carried out using 508 simulations, e.g. based on [ns-2] or [ns-3], it is also recommended 509 to perform testbed-based evaluations using Wi-Fi access points and 510 endpoints running up-to-date IEEE 802.11 protocols. [Editor's Note: 511 need to add some more discussions on the pros and cons of simulation- 512 based vs. testbed-based evaluations. Will be good to provide 513 recommended testbed configurations. ] 515 4.1. Bottleneck in Wired Network 517 The test scenarios below are intended to mimic the set up of video 518 conferencing over Wi-Fi connections from the home. Typically, the 519 Wi-Fi home network is not congested and the bottleneck is present 520 over the wired home access link. Although it is expected that test 521 evaluation results from this section are similar to those from test 522 cases defined for wired networks (see [I-D.ietf-rmcat-eval-test]), it 523 is worthwhile to run through these tests as sanity checks. 525 4.1.1. Network topology 527 Figure 2 shows topology of the network for Wi-Fi test cases. The 528 test contains multiple mobile nodes (MNs) connected to a common Wi-Fi 529 access point (AP) and their corresponding wired clients on fixed 530 nodes (FNs). Each connection carries either RMCAT or TCP traffic 531 flow. Directions of the flows can be uplink, downlink, or bi- 532 directional. 534 uplink 535 +----------------->+ 536 +------+ +------+ 537 | MN_1 |)))) /=====| FN_1 | 538 +------+ )) // +------+ 539 . )) // . 540 . )) // . 541 . )) // . 542 +------+ +----+ +-----+ +------+ 543 | MN_N | ))))))) | | | |========| FN_N | 544 +------+ | | | | +------+ 545 | AP |=========| FN0 | 546 +----------+ | | | | +----------+ 547 | MN_tcp_1 | )))) | | | |======| MN_tcp_1 | 548 +----------+ +----+ +-----+ +----------+ 549 . )) \\ . 550 . )) \\ . 551 . )) \\ . 552 +----------+ )) \\ +----------+ 553 | MN_tcp_M |))) \=====| MN_tcp_M | 554 +----------+ +----------+ 555 +<-----------------+ 556 downlink 558 Figure 2: Network topology for Wi-Fi test cases 560 4.1.2. Test setup 562 o Test duration: 120s 564 o Wi-Fi network characteristics: 566 * Radio propagation model: Log-distance path loss propagation 567 model [NS3WiFi] 569 * PHY- and MAC-layer configuration: IEEE 802.11n 571 * MCS Index at 11: 16-QAM 1/2, Raw Data Rate@52Mbps 573 o Wired path characteristics: 575 * Path capacity: 1Mbps 577 * One-Way propagation delay: 50ms. 579 * Maximum end-to-end jitter: 30ms 581 * Bottleneck queue type: Drop tail. 583 * Bottleneck queue size: 300ms. 585 * Path loss ratio: 0%. 587 o Application characteristics: 589 * Media Traffic: 591 + Media type: Video 593 + Media direction: See Section 4.1.3 595 + Number of media sources (N): See Section 4.1.3 597 + Media timeline: 599 - Start time: 0s. 601 - End time: 119s. 603 * Competing traffic: 605 + Type of sources: long-lived TCP or CBR over UDP 607 + Traffic direction: See Section 4.1.3 608 + Number of sources (M): See Section 4.1.3 610 + Congestion control: Default TCP congestion control [TBD] or 611 CBR over UDP 613 + Traffic timeline: See Section 4.1.3 615 4.1.3. Typical test scenarios 617 o Single uplink RMCAT flow: N=1 with uplink direction and M=0. 619 o One pair of bi-directional RMCAT flows: N=2 (with one uplink flow 620 and one downlink flow); M=0. 622 o One pair of bi-directional RMCAT flows, one on-off CBR over UDP 623 flow on uplink: N=2 (with one uplink flow and one downlink flow); 624 M=1 (uplink). CBR flow on time at 0s-60s, off time at 60s-119s. 626 o One pair of bi-directional RMCAT flows, one off-on CBR over UDP 627 flow on uplink: N=2 (with one uplink flow and one downlink flow); 628 M=1 (uplink). UDP off time: 0s-60s, on time: 60s-119s. 630 o One RMCAT flow competing against one long-live TCP flow over 631 uplink: N=1 (uplink) and M = 1(uplink), TCP start time at 0s and 632 end time at 119s. 634 4.1.4. Expected behavior 636 o Single uplink RMCAT flow: the candidate algorithm is expected to 637 detect the path capacity constraint, to converge to bottleneck 638 link capacity and to adapt the flow to avoid unwanted oscillation 639 when the sending bit rate is approaching the bottleneck link 640 capacity. No excessive rate oscillations should be present. 642 o Bi-directional RMCAT flows: It is expected that the candidate 643 algorithm is able to converge to the bottleneck capacity of the 644 wired path on both directions despite presence of measurment noise 645 over the Wi-Fi connection. In the presence of background TCP or 646 CBR over UDP traffic, the rate of RMCAT flows should adapt in a 647 timely manner to changes in the available bottleneck bandwidth. 649 o One RMCAT flow competing with long-live TCP flow over uplink: the 650 candidate algorithm should be able to avoid congestion collapse, 651 and to stablize at a fair share of the bottleneck link capacity. 653 4.2. Bottleneck in Wi-Fi Network 655 These test cases assume that the wired portion along the media path 656 is well-provisioned whereas the bottleneck exists over the Wi-Fi 657 access network. This is to mimic the application scenarios typically 658 encountered by users in an enterprise environment or at a coffee 659 house. 661 4.2.1. Network topology 663 Same as defined in Section 4.1.1 665 4.2.2. Test setup 667 o Test duration: 120s 669 o Wi-Fi network characteristics: 671 * Radio propagation model: Log-distance path loss propagation 672 model [NS3WiFi] 674 * PHY- and MAC-layer configuration: IEEE 802.11n 676 * MCS Index at 11: 16-QAM 1/2, Raw Data Rate at 52Mbps 678 o Wired path characteristics: 680 * Path capacity: 100Mbps 682 * One-Way propagation delay: 50ms. 684 * Maximum end-to-end jitter: 30ms 686 * Bottleneck queue type: Drop tail. 688 * Bottleneck queue size: 300ms. 690 * Path loss ratio: 0%. 692 o Application characteristics: 694 * Media Traffic: 696 + Media type: Video 698 + Media direction: See Section 4.2.3 700 + Number of media sources (N): See Section 4.2.3 701 + Media timeline: 703 - Start time: 0s. 705 - End time: 119s. 707 * Competing traffic: 709 + Type of sources: long-lived TCP or CBR over UDP 711 + Number of sources (M): See Section 4.2.3 713 + Traffic direction: See Section 4.2.3 715 + Congestion control: Default TCP congestion control [TBD] or 716 CBR over UDP 718 + Traffic timeline: See Section 4.2.3 720 4.2.3. Typical test scenarios 722 This section describes a few test scenarios that are deemed as 723 important for understanding the behavior of a RMCAT candidate 724 solution over a Wi-Fi network. 726 o Multiple RMCAT Flows Sharing the Wireless Downlink: N=16 (all 727 downlink); M = 0. This test case is for studying the impact of 728 contention on competing RMCAT flows. For an 802.11n network, 729 given the MCS Index of 11 and the corresponding raw data rate of 730 52Mbps, the total application-layer throughput (assuming 731 reasonable distance, low interference and infrequent contentions 732 caused by competing streams) is around 20Mbps. Consequently, a 733 total of N=16 RMCAT flows are needed to saturate the wireless 734 interface in this experiment. Evaluation of a given candidate 735 solution should focus on whether downlink RMCAT flows can stablize 736 at a fair share of total application-layer throughput. 738 o Multiple RMCAT Flows Sharing the Wireless Uplink: N = 16 (all 739 downlink); M = 0. When multiple clients attempt to transmit video 740 packets uplink over the wireless interface, they introduce more 741 frequent contentions and potential collisions. Per-flow 742 throughput is expected to be lower than that in the previous 743 downlink-only scenario. Evaluation of a given candidate solution 744 should focus on whether uplink flows can stablize at a fair share 745 of application-layer throughput. 747 o Multiple Bi-directional RMCAT Flows: N = 16 (8 uplink and 8 748 downlink); M = 0. The goal of this test is to evaluate 749 performance of the candidate solution in terms of bandwidth 750 fairness between uplink and downlink flow. 752 o Multiple Bi-directional RMCAT Flows with on-off CBR traffic: N = 753 16 (8 uplink and 8 downlink); M = 5(uplink). The goal of this 754 test is to evaluate adaptation behavior of the candidate solution 755 when its available bandwidth changes due to departure of 756 background traffic. The background traffic consists of several 757 (e.g., M=5) CBR flows transported over UDP, which are ON at times 758 t=0-60s and are OFF at times t=61-120s. 760 o Multiple Bi-directional RMCAT Flows with off-on CBR traffic: N = 761 16 (8 uplink and 8 downlink); M = 5(uplink). The goal of this 762 test is to evaluate adaptation behavior of the candidate solution 763 when its available bandwidth changes due to arrival of background 764 traffic. The background traffic consists of several (e.g., M=5) 765 parallel CBR flows transported over UDP, which are OFF at times 766 t=0-60s and are ON at times t=61-120s. 768 o Multiple RMCAT flows in the presence of background TCP traffic. 769 The goal of this test is to evaluate how RMCAT flows compete 770 against TCP over a congested Wi-Fi network for a given candidate 771 solution. TCP start time: 0s, end time: 119s. [Editor's Note: 772 need to add the number of recommended RMCAT and TCP flows] 774 o Varying number of RMCAT flows. The goal of this test is to 775 evaluate how a candidate RMCAT solution responds to varying 776 traffic load/demand over a congested Wi-Fi network. [Editor's 777 Note: need to specify recommended arrival/departure pattern of 778 RMCAT flows] 780 4.2.4. Expected behavior 782 o Multiple downlink RMCAT flows: each RMCAT flow should get its fair 783 share of the total bottleneck link bandwidth. Overall bandwidth 784 usage should not be significantly lower than that experienced by 785 the same number of concurrent downlink TCP flows. In other words, 786 the performance of multiple concurrent TCP flows will be used as a 787 performance benchmark for this test scenario. The end-to-end 788 delay and packet loss ratio experienced by each flow should be 789 within acceptable range for real-time multimedia applications. 791 o Multiple uplink RMCAT flows: overall bandwidth usage shared by all 792 RMCAT flows should not be significantly lower than that 793 experienced by the same number of concurrent uplink TCP flows. In 794 other words, the performance of multiple concurrent TCP flows will 795 be used as a performance benchmark for this test scenario. 797 o Multiple bi-directional RMCAT flows with dynamic background 798 traffic carry CBR flows over UDP: RMCAT flows should adapt in a 799 timely fashion to the resulting changes in available bandwidth. 801 o Multiple bi-directional RMCAT flows with TCP traffic: overall 802 bandwidth usage shared by all RMCAT flows should not be 803 significantly lower than those achieved by the same number of bi- 804 directional TCP flows. In other words, the performance of 805 multiple concurrent TCP flows will be used as a performance 806 benchmark for this test scenario. All downlink RMCAT flows are 807 expected to obtain similar bandwidth with respect to each other. 809 4.3. Other Potential Test Cases 811 4.3.1. EDCA/WMM usage 813 EDCA/WMM is prioritized QoS with four traffic classes (or Access 814 Categories) with differing priorities. RMCAT flows should achieve 815 better performance (i.e., lower delay, fewer packet losses) with 816 EDCA/WMM enabled when competing against non-interactive background 817 traffic (e.g., file transfers). When most of the traffic over Wi-Fi 818 is dominated by media, however, turning on WMM may actually degrade 819 performance since all media flows now attempt to access the wireless 820 transmission medium more aggressively, thereby causing more frequent 821 collisions and collision-induced losses. This is a topic worthy of 822 further investigation. 824 4.3.2. Effects of Legacy 802.11b Devices 826 When there is 802.11b devices connected to modern 802.11 network, it 827 may affect the performance of the whole network. Additional test 828 cases can be added to evaluate the affects of legancy devices on the 829 performance of RMCAT congestion control algorithm. 831 5. Conclusion 833 This document defines a collection of test cases that are considered 834 important for cellular and Wi-Fi networks. Moreover, this document 835 also provides a framework for defining additional test cases over 836 wireless cellular/Wi-Fi networks. 838 6. IANA Considerations 840 This memo includes no request to IANA. 842 7. Security Considerations 844 The security considerations in [I-D.ietf-rmcat-eval-criteria] and the 845 relevant congestion control algorithms apply. The principles for 846 congestion control are described in [RFC2914], and in particular any 847 new method MUST implement safeguards to avoid congestion collapse of 848 the Internet. 850 The evaluation of the test cases are intended to be run in a 851 controlled lab environment. Hence, the applications, simulators and 852 network nodes ought to be well-behaved and should not impact the 853 desired results. It is important to take appropriate caution to 854 avoid leaking non-responsive traffic from unproven congestion 855 avoidance techniques onto the open Internet. 857 8. Acknowledgements 859 We would like to thank Tomas Frankkila, Magnus Westerlund, Kristofer 860 Sandlund for their valuable comments while writing this draft. 862 9. References 864 9.1. Normative References 866 [Deployment] 867 TS 25.814, 3GPP., "Physical layer aspects for evolved 868 Universal Terrestrial Radio Access (UTRA)", October 2006, 869 . 872 [HO-def-3GPP] 873 TR 21.905, 3GPP., "Vocabulary for 3GPP Specifications", 874 December 2009, . 877 [HO-LTE-3GPP] 878 TS 36.331, 3GPP., "E-UTRA- Radio Resource Control (RRC); 879 Protocol specification", December 2011, 880 . 883 [HO-UMTS-3GPP] 884 TS 25.331, 3GPP., "Radio Resource Control (RRC); Protocol 885 specification", December 2011, 886 . 889 [I-D.ietf-rmcat-eval-criteria] 890 Singh, V., Ott, J., and S. Holmer, "Evaluating Congestion 891 Control for Interactive Real-time Media", draft-ietf- 892 rmcat-eval-criteria-08 (work in progress), November 2018. 894 [NS3WiFi] "Wi-Fi Channel Model in NS3 Simulator", 895 . 898 [QoS-3GPP] 899 TS 23.203, 3GPP., "Policy and charging control 900 architecture", June 2011, . 903 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 904 Requirement Levels", BCP 14, RFC 2119, 905 DOI 10.17487/RFC2119, March 1997, 906 . 908 [RFC2914] Floyd, S., "Congestion Control Principles", BCP 41, 909 RFC 2914, DOI 10.17487/RFC2914, September 2000, 910 . 912 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 913 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 914 May 2017, . 916 9.2. Informative References 918 [I-D.ietf-rmcat-cc-requirements] 919 Jesup, R. and Z. Sarker, "Congestion Control Requirements 920 for Interactive Real-Time Media", draft-ietf-rmcat-cc- 921 requirements-09 (work in progress), December 2014. 923 [I-D.ietf-rmcat-eval-test] 924 Sarker, Z., Singh, V., Zhu, X., and M. Ramalho, "Test 925 Cases for Evaluating RMCAT Proposals", draft-ietf-rmcat- 926 eval-test-08 (work in progress), November 2018. 928 [IEEE802.11] 929 "Standard for Information technology--Telecommunications 930 and information exchange between systems Local and 931 metropolitan area networks--Specific requirements Part 11: 932 Wireless LAN Medium Access Control (MAC) and Physical 933 Layer (PHY) Specifications", 2012. 935 [LTE-simulator] 936 "NS-3, A discrete-Event Network Simulator", 937 . 940 [ns-2] "The Network Simulator - ns-2", 941 . 943 [ns-3] "The Network Simulator - ns-3", . 945 Authors' Addresses 947 Zaheduzzaman Sarker 948 Ericsson AB 949 Laboratoriegraend 11 950 Luleae 97753 951 Sweden 953 Phone: +46 107173743 954 Email: zaheduzzaman.sarker@ericsson.com 956 Ingemar Johansson 957 Ericsson AB 958 Laboratoriegraend 11 959 Luleae 97753 960 Sweden 962 Phone: +46 10 7143042 963 Email: ingemar.s.johansson@ericsson.com 965 Xiaoqing Zhu 966 Cisco Systems 967 12515 Research Blvd., Building 4 968 Austin, TX 78759 969 USA 971 Email: xiaoqzhu@cisco.com 973 Jiantao Fu 974 Cisco Systems 975 707 Tasman Drive 976 Milpitas, CA 95035 977 USA 979 Email: jianfu@cisco.com 980 Wei-Tian Tan 981 Cisco Systems 982 725 Alder Drive 983 Milpitas, CA 95035 984 USA 986 Email: dtan2@cisco.com 988 Michael A. Ramalho 989 Cisco Systems, Inc. 990 8000 Hawkins Road 991 Sarasota, FL 34241 992 USA 994 Phone: +1 919 476 2038 995 Email: mramalho@cisco.com