idnits 2.17.1 draft-ietf-rmcat-wireless-tests-05.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 286 has weird spacing: '...nternet fix...' == The document doesn't use any RFC 2119 keywords, yet seems to have RFC 2119 boilerplate text. -- The document date (June 21, 2018) is 2135 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Missing Reference: '2D' is mentioned on line 302, but not defined == Missing Reference: '3D' is mentioned on line 302, but not defined == Missing Reference: 'Uplink' is mentioned on line 339, but not defined == Missing Reference: 'Downlink' is mentioned on line 339, but not defined == Missing Reference: 'TBD' is mentioned on line 714, but not defined == Unused Reference: 'I-D.ietf-rmcat-cc-requirements' is defined on line 902, but no explicit reference was found in the text == Outdated reference: A later version (-14) exists of draft-ietf-rmcat-eval-criteria-07 == Outdated reference: A later version (-10) exists of draft-ietf-rmcat-eval-test-06 Summary: 0 errors (**), 0 flaws (~~), 11 warnings (==), 1 comment (--).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: December 23, 2018 X. Zhu 6 J. Fu 7 W. Tan 8 M. Ramalho 9 Cisco Systems 10 June 21, 2018 12 Evaluation Test Cases for Interactive Real-Time Media over Wireless 13 Networks 14 draft-ietf-rmcat-wireless-tests-05 16 Abstract 18 There is an ongoing effort in IETF RMCAT working group to standardize 19 rate adaptation algorithm(s) for real-time interactive communication. 20 To ensure seamless and robust user experience, the proposed rate 21 adaptation algorithm(s) should work well across all access network 22 types. This document describes test cases for evaluating 23 performances of the proposed rate adaptation solutions over LTE and 24 Wi-Fi networks. 26 Status of This Memo 28 This Internet-Draft is submitted in full conformance with the 29 provisions of BCP 78 and BCP 79. 31 Internet-Drafts are working documents of the Internet Engineering 32 Task Force (IETF). Note that other groups may also distribute 33 working documents as Internet-Drafts. The list of current Internet- 34 Drafts is at https://datatracker.ietf.org/drafts/current/. 36 Internet-Drafts are draft documents valid for a maximum of six months 37 and may be updated, replaced, or obsoleted by other documents at any 38 time. It is inappropriate to use Internet-Drafts as reference 39 material or to cite them other than as "work in progress." 41 This Internet-Draft will expire on December 23, 2018. 43 Copyright Notice 45 Copyright (c) 2018 IETF Trust and the persons identified as the 46 document authors. All rights reserved. 48 This document is subject to BCP 78 and the IETF Trust's Legal 49 Provisions Relating to IETF Documents 50 (https://trustee.ietf.org/license-info) in effect on the date of 51 publication of this document. Please review these documents 52 carefully, as they describe your rights and restrictions with respect 53 to this document. Code Components extracted from this document must 54 include Simplified BSD License text as described in Section 4.e of 55 the Trust Legal Provisions and are provided without warranty as 56 described in the Simplified BSD License. 58 Table of Contents 60 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 61 2. Terminologies . . . . . . . . . . . . . . . . . . . . . . . . 3 62 3. Cellular Network Specific Test Cases . . . . . . . . . . . . 3 63 3.1. Varying Network Load . . . . . . . . . . . . . . . . . . 6 64 3.1.1. Network Connection . . . . . . . . . . . . . . . . . 6 65 3.1.2. Simulation Setup . . . . . . . . . . . . . . . . . . 7 66 3.2. Bad Radio Coverage . . . . . . . . . . . . . . . . . . . 8 67 3.2.1. Network connection . . . . . . . . . . . . . . . . . 9 68 3.2.2. Simulation Setup . . . . . . . . . . . . . . . . . . 9 69 3.3. Desired Evaluation Metrics for cellular test cases . . . 10 70 4. Wi-Fi Networks Specific Test Cases . . . . . . . . . . . . . 10 71 4.1. Bottleneck in Wired Network . . . . . . . . . . . . . . . 12 72 4.1.1. Network topology . . . . . . . . . . . . . . . . . . 12 73 4.1.2. Test setup . . . . . . . . . . . . . . . . . . . . . 13 74 4.1.3. Typical test scenarios . . . . . . . . . . . . . . . 14 75 4.1.4. Expected behavior . . . . . . . . . . . . . . . . . . 14 76 4.2. Bottleneck in Wi-Fi Network . . . . . . . . . . . . . . . 15 77 4.2.1. Network topology . . . . . . . . . . . . . . . . . . 15 78 4.2.2. Test setup . . . . . . . . . . . . . . . . . . . . . 15 79 4.2.3. Typical test scenarios . . . . . . . . . . . . . . . 16 80 4.2.4. Expected behavior . . . . . . . . . . . . . . . . . . 17 81 4.3. Other Potential Test Cases . . . . . . . . . . . . . . . 18 82 4.3.1. EDCA/WMM usage . . . . . . . . . . . . . . . . . . . 18 83 4.3.2. Effects of Legacy 802.11b Devices . . . . . . . . . . 18 84 5. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . 18 85 6. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 18 86 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 19 87 8. Security Considerations . . . . . . . . . . . . . . . . . . . 19 88 9. References . . . . . . . . . . . . . . . . . . . . . . . . . 19 89 9.1. Normative References . . . . . . . . . . . . . . . . . . 19 90 9.2. Informative References . . . . . . . . . . . . . . . . . 20 91 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 20 93 1. Introduction 95 Wireless networks (both cellular and Wi-Fi [IEEE802.11] local area 96 network) are an integral part of the Internet. Mobile devices 97 connected to the wireless networks generate huge amount of media 98 traffic in the Internet. Application scenarios range from users 99 having a video call in the bus to media consumption by someone 100 sitting on a living room couch. It is well known that the 101 characteristics and technical challenges for offering multimedia 102 services over wireless are very different from those of providing the 103 same service over a wired network. Even though RMCAT basic test 104 cases as defined in [I-D.ietf-rmcat-eval-test] have covered many 105 effects of the impairments also visible in wireless networks, there 106 remains characteristics and dynamics unique to a given wireless 107 environment. For example, in LTE networks the base station maintains 108 queues per radio bearer per user hence it leads to a different nature 109 of interaction from that over the wired network, where traffic from 110 all users share the same queue. Furthermore, user mobility in a 111 cellular network is different than user mobility in a Wi-Fi network. 112 Therefore, It is important to evaluate performance of the proposed 113 RMCAT candidate solutions separately over cellular mobile networks 114 and over Wi-Fi local networks (i.e., IEEE 802.11xx protocol family ). 116 RMCAT evaluation criteria [I-D.ietf-rmcat-eval-criteria] document 117 provides the guideline for evaluating candidate algorithms and 118 recognizes the importance of testing over wireless access networks. 119 However, it does not describe any specific test cases for evaluating 120 performance of the candidate algorithm. This document describes test 121 cases specifically targeting cellular networks such as LTE networks 122 and Wi-Fi local networks. 124 2. Terminologies 126 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 127 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 128 "OPTIONAL" in this document are to be interpreted as described in BCP 129 14 [RFC2119] [RFC8174] when, and only when, they appear in all 130 capitals, as shown here. 132 3. Cellular Network Specific Test Cases 134 A cellular environment is more complicated than a wireline ditto 135 since it seeks to provide services in the context of variable 136 available bandwidth, location dependencies and user mobilities at 137 different speeds. In a cellular network the user may reach the cell 138 edge which may lead to a significant amount of retransmissions to 139 deliver the data from the base station to the destination and vice 140 versa. These network links or radio links will often act as a 141 bottleneck for the rest of the network which will eventually lead to 142 excessive delays or packet drops. An efficient retransmission or 143 link adaptation mechanism can reduce the packet loss probability but 144 there will still be some packet losses and delay variations. 145 Moreover, with increased cell load or handover to a congested cell, 146 congestion in transport network will become even worse. Besides, 147 there are certain characteristics which make the cellular network 148 different and challenging than other types of access network such as 149 Wi-Fi and wired network. In a cellular network - 151 o The bottleneck is often a shared link with relatively few users. 153 * The cost per bit over the shared link varies over time and is 154 different for different users. 156 * Left over/ unused resource can be grabbed by other greedy 157 users. 159 o Queues are always per radio bearer hence each user can have many 160 of such queues. 162 o Users can experience both Inter and Intra Radio Access Technology 163 (RAT) handovers ("handover" definition in [HO-def-3GPP] ). 165 o Handover between cells, or change of serving cells (see in 166 [HO-LTE-3GPP] and [HO-UMTS-3GPP] ) might cause user plane 167 interruptions which can lead to bursts of packet losses, delay 168 and/or jitter. The exact behavior depends on the type of radio 169 bearer. Typically, the default best effort bearers do not 170 generate packet loss, instead packets are queued up and 171 transmitted once the handover is completed. 173 o The network part decides how much the user can transmit. 175 o The cellular network has variable link capacity per user 177 * Can vary as fast as a period of milliseconds. 179 * Depends on lots of facts (such as distance, speed, 180 interference, different flows). 182 * Uses complex and smart link adaptation which makes the link 183 behavior ever more dynamic. 185 * The scheduling priority depends on the estimated throughput. 187 o Both Quality of Service (QoS) and non-QoS radio bearers can be 188 used. 190 Hence, a real-time communication application operating in such a 191 cellular network need to cope with shared bottleneck link and 192 variable link capacity, event likes handover, non-congestion related 193 loss, abrupt change in bandwidth (both short term and long term) due 194 to handover, network load and bad radio coverage. Even though 3GPP 195 define QoS bearers [QoS-3GPP] to ensure high quality user experience, 196 adaptive real-time applications are desired. 198 Different mobile operators deploy their own cellular network with 199 their own set of network functionalities and policies. Usually, a 200 mobile operator network includes 2G, EDGE, 3G and 4G radio access 201 technologies. Looking at the specifications of such radio 202 technologies it is evident that only 3G and 4G radio technologies can 203 support the high bandwidth requirements from real-time interactive 204 video applications. The future real-time interactive application 205 will impose even greater demand on cellular network performance which 206 makes 4G (and beyond radio technologies) more suitable access 207 technology for such genre of application. 209 The key factors to define test cases for cellular network are 211 o Shared and varying link capacity 213 o Mobility 215 o Handover 217 However, for cellular network it is very hard to separate such events 218 from one another as these events are heavily related. Hence instead 219 of devising separate test cases for all those important events we 220 have divided the test case in two categories. It should be noted 221 that in the following test cases the goal is to evaluate the 222 performance of candidate algorithms over radio interface of the 223 cellular network. Hence it is assumed that the radio interface is 224 the bottleneck link between the communicating peers and that the core 225 network does not add any extra congestion in the path. Also the 226 combination of multiple access technologies such as one user has LTE 227 connection and another has Wi-Fi connection is kept out of the scope 228 of this document. However, later those additional scenarios can also 229 be added in this list of test cases. While defining the test cases 230 we assumed a typical real-time telephony scenario over cellular 231 networks where one real-time session consists of one voice stream and 232 one video stream. We recommend that an LTE network simulator is used 233 for the test cases defined in this document, for example-NS-3 LTE 234 simulator [LTE-simulator]. 236 3.1. Varying Network Load 238 The goal of this test is to evaluate the performance of the candidate 239 congestion control algorithm under varying network load. The network 240 load variation is created by adding and removing network users a.k.a. 241 User Equipments (UEs) during the simulation. In this test case, each 242 of the user/UE in the media session is an RMCAT compliant endpoint. 243 The arrival of users follows a Poisson distribution, which is 244 proportional to the length of the call, so that the number of users 245 per cell is kept fairly constant during the evaluation period. At 246 the beginning of the simulation there should be enough amount of time 247 to warm-up the network. This is to avoid running the evaluation in 248 an empty network where network nodes are having empty buffers, low 249 interference at the beginning of the simulation. This network 250 initialization period is therefore excluded from the evaluation 251 period. 253 This test case also includes user mobility and competing traffic. 254 The competing traffics includes both same kind of flows (with same 255 adaptation algorithms) and different kind of flows (with different 256 service and congestion control). The investigated congestion control 257 algorithms should show maximum possible network utilization and 258 stability in terms of rate variations, lowest possible end to end 259 frame latency, network latency and Packet Loss Rate (PLR) at 260 different cell load level. 262 3.1.1. Network Connection 264 Each mobile user is connected to a fixed user. The connection 265 between the mobile user and fixed user consists of a LTE radio 266 access, an Evolved Packet Core (EPC) and an Internet connection. The 267 mobile user is connected to the EPC using LTE radio access technology 268 which is further connected to the Internet. The fixed user is 269 connected to the Internet via wired connection with no bottleneck 270 (practically infinite bandwidth). The Internet and wired connection 271 in this setup does not add any network impairments to the test, it 272 only adds 10ms of one-way transport propagation delay. 274 The path from the fixed user to mobile user is defines as "Downlink" 275 and the path from mobile user to the fixed user is defined as 276 "Uplink". We assume that only uplink or downlink is congested for 277 the mobile users. Hence, we recommend that the uplink and downlink 278 simulations are run separately. 280 uplink 281 ++))) +--------------------------> 282 ++-+ ((o)) 283 | | / \ +-------+ +------+ +---+ 284 +--+ / \----+ +-----+ +----+ | 285 / \ +-------+ +------+ +---+ 286 UE BS EPC Internet fixed 287 <--------------------------+ 288 downlink 290 Figure 1: Simulation Topology 292 3.1.2. Simulation Setup 294 The values enclosed within " [ ] " for the following simulation 295 attributes follow the notion set in [I-D.ietf-rmcat-eval-test]. The 296 desired simulation setup as follows- 298 1. Radio environment 300 A. Deployment and propagation model : 3GPP case 1[Deployment] 302 B. Antenna: Multiple-Input and Multiple-Output (MIMO), [2D, 3D] 304 C. Mobility: [3km/h, 30km/h] 306 D. Transmission bandwidth: 10Mhz 308 E. Number of cells: multi cell deployment (3 Cells per Base 309 Station (BS) * 7 BS) = 21 cells 311 F. Cell radius: 166.666 Meters 313 G. Scheduler: Proportional fair with no priority 315 H. Bearer: Default bearer for all traffic. 317 I. Active Queue Management (AQM) settings: AQM [on,off] 319 2. End to end Round Trip Time (RTT): [ 40, 150] 321 3. User arrival model: Poisson arrival model 323 4. User intensity: 325 * Downlink user intensity: {0.7, 1.4, 2.1, 2.8, 3.5, 4.2, 4.9, 326 5.6, 6.3, 7.0, 7.7, 8.4, 9,1, 9.8, 10.5} 328 * Uplink user intercity : {0.7, 1.4, 2.1, 2.8, 3.5, 4.2, 4.9, 329 5.6, 6.3, 7.0} 331 5. Simulation duration: 91s 333 6. Evaluation period : 30s-60s 335 7. Media traffic 337 1. Media type: Video 339 a. Media direction: [Uplink, Downlink] 341 b. Number of Media source per user: One (1) 343 c. Media duration per user: 30s 345 d. Media source: same as define in section 4.3 of 346 [I-D.ietf-rmcat-eval-test] 348 2. Media Type : Audio 350 a. Media direction: Uplink and Downlink 352 b. Number of Media source per user: One (1) 354 c. Media duration per user: 30s 356 d. Media codec: Constant BitRate (CBR) 358 e. Media bitrate : 20 Kbps 360 f. Adaptation: off 362 8. Other traffic model: 364 * Downlink simulation: Maximum of 4Mbps/cell (web browsing or 365 FTP traffic) 367 * Unlink simulation: Maximum of 2Mbps/cell (web browsing or FTP 368 traffic) 370 3.2. Bad Radio Coverage 372 The goal of this test is to evaluate the performance of candidate 373 congestion control algorithm when users visit part of the network 374 with bad radio coverage. The scenario is created by using larger 375 cell radius than previous test case. In this test case each of the 376 user/UE in the media session is an RMCAT compliant endpoint. The 377 arrival of users follows a Poisson distribution, which is 378 proportional to the length of the call, so that the number of users 379 per cell is kept fairly constant during the evaluation period. At 380 the beginning of the simulation there should be enough amount of time 381 to warm-up the network. This is to avoid running the evaluation in 382 an empty network where network nodes are having empty buffers, low 383 interference at the beginning of the simulation. This network 384 initialization period is therefore excluded from the evaluation 385 period. 387 This test case also includes user mobility and competing traffic. 388 The competing traffics includes same kind of flows (with same 389 adaptation algorithms) . The investigated congestion control 390 algorithms should show maximum possible network utilization and 391 stability in terms of rate variations, lowest possible end to end 392 frame latency, network latency and Packet Loss Rate (PLR) at 393 different cell load level. 395 3.2.1. Network connection 397 Same as defined in Section 3.1.1 399 3.2.2. Simulation Setup 401 The desired simulation setup is same as Varying Network Load test 402 case defined in Section 3.1 except following changes- 404 1. Radio environment : Same as defined in Section 3.1.2 except 405 followings 407 A. Deployment and propagation model : 3GPP case 3[Deployment] 409 B. Cell radius: 577.3333 Meters 411 C. Mobility: 3km/h 413 2. User intensity = {0.7, 1.4, 2.1, 2.8, 3.5, 4.2, 4.9, 5.6, 6.3, 414 7.0} 416 3. Media traffic model: Same as defined in Section 3.1.2 418 4. Other traffic model: None 420 3.3. Desired Evaluation Metrics for cellular test cases 422 RMCAT evaluation criteria document [I-D.ietf-rmcat-eval-criteria] 423 defines metrics to be used to evaluate candidate algorithms. 424 However, looking at the nature and distinction of cellular networks 425 we recommend at minimum following metrics to be used to evaluate the 426 performance of the candidate algorithms for the test cases defined in 427 this document. 429 The desired metrics are- 431 o Average cell throughput (for all cells), shows cell utilizations. 433 o Application sending and receiving bitrate, goodput. 435 o Packet Loss Rate (PLR). 437 o End to end Media frame delay. For video, this means the delay 438 from capture to display. 440 o Transport delay. 442 o Algorithm stability in terms of rate variation. 444 4. Wi-Fi Networks Specific Test Cases 446 Given the prevalence of Internet access links over Wi-Fi, it is 447 important to evaluate candidate RMCAT congestion control solutions 448 over test cases that include Wi-Fi access lines. Such evaluations 449 should also highlight the inherent different characteristics of Wi-Fi 450 networks in contrast to wired networks: 452 o The wireless radio channel is subject to interference from nearby 453 transmitters, multipath fading, and shadowing, causing 454 fluctuations in link throughput and sometimes an error-prone 455 communication environment 457 o Available network bandwidth is not only shared over the air 458 between cocurrent users, but also between uplink and downlink 459 traffic due to the half duplex nature of wireless transmission 460 medium. 462 o Packet transmissions over Wi-Fi are susceptible to contentions and 463 collisions over the air. Consequently, traffic load beyond a 464 certain utilization level over a Wi-Fi network can introduce 465 frequent collisions and significant network overhead. This, in 466 turn, leads to excessive delay, retransmissions, packet losses and 467 lower effective bandwidth for applications. 469 o The IEEE 802.11 standard (i.e., Wi-Fi) supports multi-rate 470 transmission capabilities by dynamically choosing the most 471 appropriate modulation scheme for a given received singal 472 strength. A different choice of physical-layer rate leads to 473 different application-layer throughput. 475 o Presence of legancy 802.11b networks can significantly slow down 476 the the rest of a modern Wi-Fi Network, since it takes longer to 477 transmit the same packet over a slower link than over a faster 478 link. [Editor's note: maybe include a reference here instead.] 480 o Handover from one Wi-Fi Access Point (AP) to another may lead to 481 packet delay and losses during the process. 483 o IEEE 802.11e defined EDCA/WMM (Enhanced DCF Channel Access/Wi-Fi 484 Multi-Media) to give voice and video streams higher priority over 485 pure data applications (e.g., file transfers). 487 In summary, presence of Wi-Fi access links in different network 488 topologies can exert different impact on the network performance in 489 terms of application-layer effective throughput, packet loss rate, 490 and packet delivery delay. These, in turn, influence the behavior of 491 end-to-end real-time multimedia congestion control. 493 Throughout this draft, unless otherwise mentioned, test cases are 494 described using 802.11n due to its wide availability in real-world 495 networks. Statistics collected from enterprise Wi-Fi networks show 496 that the dominant physical modes are 802.11n and 802.11ac, accounting 497 for 73.6% and 22.5% of enterprise network users, respectively. 499 Typically, a Wi-Fi access network connects to a wired infrastructure. 500 Either the wired or the Wi-Fi segment of the network could be the 501 bottleneck. In the following sections, we describe basic test cases 502 for both scenarios separately. The same set of performance metrics 503 as in [I-D.ietf-rmcat-eval-test]) should be collected for each test 504 case. 506 While all test cases described below can be carried out using 507 simulations, e.g. based on [ns-2] or [ns-3], it is also recommended 508 to perform testbed-based evaluations using Wi-Fi access points and 509 endpoints running up-to-date IEEE 802.11 protocols. [Editor's Note: 510 need to add some more discussions on the pros and cons of simulation- 511 based vs. testbed-based evaluations. Will be good to provide 512 recommended testbed configurations. ] 514 4.1. Bottleneck in Wired Network 516 The test scenarios below are intended to mimic the set up of video 517 conferencing over Wi-Fi connections from the home. Typically, the 518 Wi-Fi home network is not congested and the bottleneck is present 519 over the wired home access link. Although it is expected that test 520 evaluation results from this section are similar to those from test 521 cases defined for wired networks (see [I-D.ietf-rmcat-eval-test]), it 522 is worthwhile to run through these tests as sanity checks. 524 4.1.1. Network topology 526 Figure 2 shows topology of the network for Wi-Fi test cases. The 527 test contains multiple mobile nodes (MNs) connected to a common Wi-Fi 528 access point (AP) and their corresponding wired clients on fixed 529 nodes (FNs). Each connection carries either RMCAT or TCP traffic 530 flow. Directions of the flows can be uplink, downlink, or bi- 531 directional. 533 uplink 534 +----------------->+ 535 +------+ +------+ 536 | MN_1 |)))) /=====| FN_1 | 537 +------+ )) // +------+ 538 . )) // . 539 . )) // . 540 . )) // . 541 +------+ +----+ +-----+ +------+ 542 | MN_N | ))))))) | | | |========| FN_N | 543 +------+ | | | | +------+ 544 | AP |=========| FN0 | 545 +----------+ | | | | +----------+ 546 | MN_tcp_1 | )))) | | | |======| MN_tcp_1 | 547 +----------+ +----+ +-----+ +----------+ 548 . )) \\ . 549 . )) \\ . 550 . )) \\ . 551 +----------+ )) \\ +----------+ 552 | MN_tcp_M |))) \=====| MN_tcp_M | 553 +----------+ +----------+ 554 +<-----------------+ 555 downlink 557 Figure 2: Network topology for Wi-Fi test cases 559 4.1.2. Test setup 561 o Test duration: 120s 563 o Wi-Fi network characteristics: 565 * Radio propagation model: Log-distance path loss propagation 566 model [NS3WiFi] 568 * PHY- and MAC-layer configuration: IEEE 802.11n 570 * MCS Index at 11: 16-QAM 1/2, Raw Data Rate@52Mbps 572 o Wired path characteristics: 574 * Path capacity: 1Mbps 576 * One-Way propagation delay: 50ms. 578 * Maximum end-to-end jitter: 30ms 580 * Bottleneck queue type: Drop tail. 582 * Bottleneck queue size: 300ms. 584 * Path loss ratio: 0%. 586 o Application characteristics: 588 * Media Traffic: 590 + Media type: Video 592 + Media direction: See Section 4.1.3 594 + Number of media sources (N): See Section 4.1.3 596 + Media timeline: 598 - Start time: 0s. 600 - End time: 119s. 602 * Competing traffic: 604 + Type of sources: long-lived TCP or CBR over UDP 606 + Traffic direction: See Section 4.1.3 607 + Number of sources (M): See Section 4.1.3 609 + Congestion control: Default TCP congestion control [TBD] or 610 CBR over UDP 612 + Traffic timeline: See Section 4.1.3 614 4.1.3. Typical test scenarios 616 o Single uplink RMCAT flow: N=1 with uplink direction and M=0. 618 o One pair of bi-directional RMCAT flows: N=2 (with one uplink flow 619 and one downlink flow); M=0. 621 o One pair of bi-directional RMCAT flows, one on-off CBR over UDP 622 flow on uplink: N=2 (with one uplink flow and one downlink flow); 623 M=1 (uplink). CBR flow on time at 0s-60s, off time at 60s-119s. 625 o One pair of bi-directional RMCAT flows, one off-on CBR over UDP 626 flow on uplink: N=2 (with one uplink flow and one downlink flow); 627 M=1 (uplink). UDP off time: 0s-60s, on time: 60s-119s. 629 o One RMCAT flow competing against one long-live TCP flow over 630 uplink: N=1 (uplink) and M = 1(uplink), TCP start time at 0s and 631 end time at 119s. 633 4.1.4. Expected behavior 635 o Single uplink RMCAT flow: the candidate algorithm is expected to 636 detect the path capacity constraint, to converge to bottleneck 637 link capacity and to adapt the flow to avoid unwanted oscillation 638 when the sending bit rate is approaching the bottleneck link 639 capacity. No excessive rate oscillations should be present. 641 o Bi-directional RMCAT flows: It is expected that the candidate 642 algorithm is able to converge to the bottleneck capacity of the 643 wired path on both directions despite presence of measurment noise 644 over the Wi-Fi connection. In the presence of background TCP or 645 CBR over UDP traffic, the rate of RMCAT flows should adapt in a 646 timely manner to changes in the available bottleneck bandwidth. 648 o One RMCAT flow competing with long-live TCP flow over uplink: the 649 candidate algorithm should be able to avoid congestion collapse, 650 and to stablize at a fair share of the bottleneck link capacity. 652 4.2. Bottleneck in Wi-Fi Network 654 These test cases assume that the wired portion along the media path 655 is well-provisioned whereas the bottleneck exists over the Wi-Fi 656 access network. This is to mimic the application scenarios typically 657 encountered by users in an enterprise environment or at a coffee 658 house. 660 4.2.1. Network topology 662 Same as defined in Section 4.1.1 664 4.2.2. Test setup 666 o Test duration: 120s 668 o Wi-Fi network characteristics: 670 * Radio propagation model: Log-distance path loss propagation 671 model [NS3WiFi] 673 * PHY- and MAC-layer configuration: IEEE 802.11n 675 * MCS Index at 11: 16-QAM 1/2, Raw Data Rate at 52Mbps 677 o Wired path characteristics: 679 * Path capacity: 100Mbps 681 * One-Way propagation delay: 50ms. 683 * Maximum end-to-end jitter: 30ms 685 * Bottleneck queue type: Drop tail. 687 * Bottleneck queue size: 300ms. 689 * Path loss ratio: 0%. 691 o Application characteristics: 693 * Media Traffic: 695 + Media type: Video 697 + Media direction: See Section 4.2.3 699 + Number of media sources (N): See Section 4.2.3 700 + Media timeline: 702 - Start time: 0s. 704 - End time: 119s. 706 * Competing traffic: 708 + Type of sources: long-lived TCP or CBR over UDP 710 + Number of sources (M): See Section 4.2.3 712 + Traffic direction: See Section 4.2.3 714 + Congestion control: Default TCP congestion control [TBD] or 715 CBR over UDP 717 + Traffic timeline: See Section 4.2.3 719 4.2.3. Typical test scenarios 721 This section describes a few test scenarios that are deemed as 722 important for understanding the behavior of a RMCAT candidate 723 solution over a Wi-Fi network. 725 o Multiple RMCAT Flows Sharing the Wireless Downlink: N=16 (all 726 downlink); M = 0. This test case is for studying the impact of 727 contention on competing RMCAT flows. For an 802.11n network, 728 given the MCS Index of 11 and the corresponding raw data rate of 729 52Mbps, the total application-layer throughput (assuming 730 reasonable distance, low interference and infrequent contentions 731 caused by competing streams) is around 20Mbps. Consequently, a 732 total of N=16 RMCAT flows are needed to saturate the wireless 733 interface in this experiment. Evaluation of a given candidate 734 solution should focus on whether downlink RMCAT flows can stablize 735 at a fair share of total application-layer throughput. 737 o Multiple RMCAT Flows Sharing the Wireless Uplink: N = 16 (all 738 downlink); M = 0. When multiple clients attempt to transmit video 739 packets uplink over the wireless interface, they introduce more 740 frequent contentions and potential collisions. Per-flow 741 throughput is expected to be lower than that in the previous 742 downlink-only scenario. Evaluation of a given candidate solution 743 should focus on whether uplink flows can stablize at a fair share 744 of application-layer throughput. 746 o Multiple Bi-directional RMCAT Flows: N = 16 (8 uplink and 8 747 downlink); M = 0. The goal of this test is to evaluate 748 performance of the candidate solution in terms of bandwidth 749 fairness between uplink and downlink flow. 751 o Multiple Bi-directional RMCAT Flows with on-off CBR traffic: N = 752 16 (8 uplink and 8 downlink); M = 5(uplink). The goal of this 753 test is to evaluate adaptation behavior of the candidate solution 754 when its available bandwidth changes due to departure of 755 background traffic. The background traffic consists of several 756 (e.g., M=5) CBR flows transported over UDP, which are ON at times 757 t=0-60s and are OFF at times t=61-120s. 759 o Multiple Bi-directional RMCAT Flows with off-on CBR traffic: N = 760 16 (8 uplink and 8 downlink); M = 5(uplink). The goal of this 761 test is to evaluate adaptation behavior of the candidate solution 762 when its available bandwidth changes due to arrival of background 763 traffic. The background traffic consists of several (e.g., M=5) 764 parallel CBR flows transported over UDP, which are OFF at times 765 t=0-60s and are ON at times t=61-120s. 767 o Multiple RMCAT flows in the presence of background TCP traffic. 768 The goal of this test is to evaluate how RMCAT flows compete 769 against TCP over a congested Wi-Fi network for a given candidate 770 solution. TCP start time: 0s, end time: 119s. [Editor's Note: 771 need to add the number of recommended RMCAT and TCP flows] 773 o Varying number of RMCAT flows. The goal of this test is to 774 evaluate how a candidate RMCAT solution responds to varying 775 traffic load/demand over a congested Wi-Fi network. [Editor's 776 Note: need to specify recommended arrival/departure pattern of 777 RMCAT flows] 779 4.2.4. Expected behavior 781 o Multiple downlink RMCAT flows: each RMCAT flow should get its fair 782 share of the total bottleneck link bandwidth. Overall bandwidth 783 usage should not be significantly lower than that experienced by 784 the same number of concurrent downlink TCP flows. In other words, 785 the performance of multiple concurrent TCP flows will be used as a 786 performance benchmark for this test scenario. The end-to-end 787 delay and packet loss ratio experienced by each flow should be 788 within acceptable range for real-time multimedia applications. 790 o Multiple uplink RMCAT flows: overall bandwidth usage shared by all 791 RMCAT flows should not be significantly lower than that 792 experienced by the same number of concurrent uplink TCP flows. In 793 other words, the performance of multiple concurrent TCP flows will 794 be used as a performance benchmark for this test scenario. 796 o Multiple bi-directional RMCAT flows with dynamic background 797 traffic carry CBR flows over UDP: RMCAT flows should adapt in a 798 timely fashion to the resulting changes in available bandwidth. 800 o Multiple bi-directional RMCAT flows with TCP traffic: overall 801 bandwidth usage shared by all RMCAT flows should not be 802 significantly lower than those achieved by the same number of bi- 803 directional TCP flows. In other words, the performance of 804 multiple concurrent TCP flows will be used as a performance 805 benchmark for this test scenario. All downlink RMCAT flows are 806 expected to obtain similar bandwidth with respect to each other. 808 4.3. Other Potential Test Cases 810 4.3.1. EDCA/WMM usage 812 EDCA/WMM is prioritized QoS with four traffic classes (or Access 813 Categories) with differing priorities. RMCAT flows should achieve 814 better performance (i.e., lower delay, fewer packet losses) with 815 EDCA/WMM enabled when competing against non-interactive background 816 traffic (e.g., file transfers). When most of the traffic over Wi-Fi 817 is dominated by media, however, turning on WMM may actually degrade 818 performance since all media flows now attempt to access the wireless 819 transmission medium more aggressively, thereby causing more frequent 820 collisions and collision-induced losses. This is a topic worthy of 821 further investigation. 823 4.3.2. Effects of Legacy 802.11b Devices 825 When there is 802.11b devices connected to modern 802.11 network, it 826 may affect the performance of the whole network. Additional test 827 cases can be added to evaluate the affects of legancy devices on the 828 performance of RMCAT congestion control algorithm. 830 5. Conclusion 832 This document defines a collection of test cases that are considered 833 important for cellular and Wi-Fi networks. Moreover, this document 834 also provides a framework for defining additional test cases over 835 wireless cellular/Wi-Fi networks. 837 6. Acknowledgements 839 We would like to thank Tomas Frankkila, Magnus Westerlund, Kristofer 840 Sandlund for their valuable comments while writing this draft. 842 7. IANA Considerations 844 This memo includes no request to IANA. 846 8. Security Considerations 848 Security issues have not been discussed in this memo. 850 9. References 852 9.1. Normative References 854 [Deployment] 855 TS 25.814, 3GPP., "Physical layer aspects for evolved 856 Universal Terrestrial Radio Access (UTRA)", October 2006, 857 . 860 [HO-def-3GPP] 861 TR 21.905, 3GPP., "Vocabulary for 3GPP Specifications", 862 December 2009, . 865 [HO-LTE-3GPP] 866 TS 36.331, 3GPP., "E-UTRA- Radio Resource Control (RRC); 867 Protocol specification", December 2011, 868 . 871 [HO-UMTS-3GPP] 872 TS 25.331, 3GPP., "Radio Resource Control (RRC); Protocol 873 specification", December 2011, 874 . 877 [I-D.ietf-rmcat-eval-criteria] 878 Singh, V., Ott, J., and S. Holmer, "Evaluating Congestion 879 Control for Interactive Real-time Media", draft-ietf- 880 rmcat-eval-criteria-07 (work in progress), May 2018. 882 [NS3WiFi] "Wi-Fi Channel Model in NS3 Simulator", 883 . 886 [QoS-3GPP] 887 TS 23.203, 3GPP., "Policy and charging control 888 architecture", June 2011, . 891 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 892 Requirement Levels", BCP 14, RFC 2119, 893 DOI 10.17487/RFC2119, March 1997, 894 . 896 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 897 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 898 May 2017, . 900 9.2. Informative References 902 [I-D.ietf-rmcat-cc-requirements] 903 Jesup, R. and Z. Sarker, "Congestion Control Requirements 904 for Interactive Real-Time Media", draft-ietf-rmcat-cc- 905 requirements-09 (work in progress), December 2014. 907 [I-D.ietf-rmcat-eval-test] 908 Sarker, Z., Singh, V., Zhu, X., and M. Ramalho, "Test 909 Cases for Evaluating RMCAT Proposals", draft-ietf-rmcat- 910 eval-test-06 (work in progress), June 2018. 912 [IEEE802.11] 913 "Standard for Information technology--Telecommunications 914 and information exchange between systems Local and 915 metropolitan area networks--Specific requirements Part 11: 916 Wireless LAN Medium Access Control (MAC) and Physical 917 Layer (PHY) Specifications", 2012. 919 [LTE-simulator] 920 "NS-3, A discrete-Event Network Simulator", 921 . 924 [ns-2] "The Network Simulator - ns-2", 925 . 927 [ns-3] "The Network Simulator - ns-3", . 929 Authors' Addresses 931 Zaheduzzaman Sarker 932 Ericsson AB 933 Laboratoriegraend 11 934 Luleae 97753 935 Sweden 937 Phone: +46 107173743 938 Email: zaheduzzaman.sarker@ericsson.com 939 Ingemar Johansson 940 Ericsson AB 941 Laboratoriegraend 11 942 Luleae 97753 943 Sweden 945 Phone: +46 10 7143042 946 Email: ingemar.s.johansson@ericsson.com 948 Xiaoqing Zhu 949 Cisco Systems 950 12515 Research Blvd., Building 4 951 Austin, TX 78759 952 USA 954 Email: xiaoqzhu@cisco.com 956 Jiantao Fu 957 Cisco Systems 958 707 Tasman Drive 959 Milpitas, CA 95035 960 USA 962 Email: jianfu@cisco.com 964 Wei-Tian Tan 965 Cisco Systems 966 725 Alder Drive 967 Milpitas, CA 95035 968 USA 970 Email: dtan2@cisco.com 972 Michael A. Ramalho 973 Cisco Systems, Inc. 974 8000 Hawkins Road 975 Sarasota, FL 34241 976 USA 978 Phone: +1 919 476 2038 979 Email: mramalho@cisco.com