idnits 2.17.1 draft-irtf-icnrg-videostreaming-07.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- -- The document has an IETF Trust Provisions (28 Dec 2009) Section 6.c(ii) Publication Limitation clause. If this document is intended for submission to the IESG for publication, this constitutes an error. 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 : ---------------------------------------------------------------------------- ** The document seems to lack an Introduction section. (A line matching the expected section header was found, but with an unexpected indentation: ' 1. Introduction' ) ** The document seems to lack a Security Considerations section. (A line matching the expected section header was found, but with an unexpected indentation: ' 10. Security Considerations' ) ** 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.) (A line matching the expected section header was found, but with an unexpected indentation: ' 11. IANA Considerations' ) ** The document seems to lack an Authors' Addresses Section. ** There is 1 instance of too long lines in the document, the longest one being 110 characters in excess of 72. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == Line 1669 has weird spacing: '...orkshop on I...' == Line 1670 has weird spacing: '...cations over ...' == Line 1674 has weird spacing: '...amework for ...' == Line 1697 has weird spacing: '... Paper on D...' == Line 1699 has weird spacing: '...Systems and ...' == (1 more instance...) == The document doesn't use any RFC 2119 keywords, yet seems to have RFC 2119 boilerplate text. == The document seems to contain a disclaimer for pre-RFC5378 work, but was first submitted on or after 10 November 2008. The disclaimer is usually necessary only for documents that revise or obsolete older RFCs, and that take significant amounts of text from those RFCs. If you can contact all authors of the source material and they are willing to grant the BCP78 rights to the IETF Trust, you can and should remove the disclaimer. Otherwise, the disclaimer is needed and you can ignore this comment. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (February 16, 2016) is 2992 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- -- Missing reference section? '2' on line 1590 looks like a reference -- Missing reference section? '3' on line 1596 looks like a reference -- Missing reference section? '4' on line 1603 looks like a reference -- Missing reference section? '6' on line 1612 looks like a reference -- Missing reference section? 'RFC2119' on line 201 looks like a reference -- Missing reference section? '7' on line 1619 looks like a reference -- Missing reference section? '25' on line 1696 looks like a reference -- Missing reference section? '1' on line 1586 looks like a reference -- Missing reference section? '5' on line 1607 looks like a reference -- Missing reference section? '8' on line 1623 looks like a reference -- Missing reference section? '9' on line 1630 looks like a reference -- Missing reference section? '30' on line 1719 looks like a reference -- Missing reference section? '10' on line 1634 looks like a reference -- Missing reference section? '11' on line 1638 looks like a reference -- Missing reference section? '12' on line 1645 looks like a reference -- Missing reference section? '13' on line 1647 looks like a reference -- Missing reference section? '14' on line 1652 looks like a reference -- Missing reference section? 'RFC6972' on line 1580 looks like a reference -- Missing reference section? '15' on line 1654 looks like a reference -- Missing reference section? '16' on line 1657 looks like a reference -- Missing reference section? '17' on line 1661 looks like a reference -- Missing reference section? '31' on line 1723 looks like a reference -- Missing reference section? '22' on line 1682 looks like a reference -- Missing reference section? '23' on line 1686 looks like a reference -- Missing reference section? '18' on line 1665 looks like a reference -- Missing reference section? '19' on line 1667 looks like a reference -- Missing reference section? '20' on line 1673 looks like a reference -- Missing reference section? '21' on line 1678 looks like a reference -- Missing reference section? '24' on line 1691 looks like a reference -- Missing reference section? '26' on line 1702 looks like a reference -- Missing reference section? '27' on line 1706 looks like a reference -- Missing reference section? '28' on line 1711 looks like a reference -- Missing reference section? '29' on line 1715 looks like a reference Summary: 5 errors (**), 0 flaws (~~), 9 warnings (==), 35 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 1 ICNRG C. Westphal, Ed. 2 Internet Draft Huawei 3 Intended status: Informational S. Lederer 4 Expires: August 15, 2016 C. Timmerer 5 D. Posch 6 Alpen-Adria University Klagenfurt 7 Aytac Azgin 8 S. Liu 9 Huawei 10 C. Mueller 11 Bitmovin 12 A.Detti 13 University of Rome Tor Vergata 14 D. Corujo 15 University of Aveiro 16 J. Wang 17 City University of Hong-Kong 18 Marie-Jose Montpetit 19 Niall Murray 20 Athlone Institute of Technology 22 February 16, 2016 24 Adaptive Video Streaming over ICN 25 draft-irtf-icnrg-videostreaming-07.txt 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 This Internet-Draft is submitted in full conformance with the 33 provisions of BCP 78 and BCP 79. This document may not be modified, 34 and derivative works of it may not be created, and it may not be 35 published except as an Internet-Draft. 37 This Internet-Draft is submitted in full conformance with the 38 provisions of BCP 78 and BCP 79. This document may not be modified, 39 and derivative works of it may not be created, except to publish it 40 as an RFC and to translate it into languages other than English. 42 This document may contain material from IETF Documents or IETF 43 Contributions published or made publicly available before November 44 10, 2008. The person(s) controlling the copyright in some of this 45 material may not have granted the IETF Trust the right to allow 46 modifications of such material outside the IETF Standards Process. 47 Without obtaining an adequate license from the person(s) controlling 48 the copyright in such materials, this document may not be modified 49 outside the IETF Standards Process, and derivative works of it may 50 not be created outside the IETF Standards Process, except to format 51 it for publication as an RFC or to translate it into languages other 52 than English. 54 Internet-Drafts are working documents of the Internet Engineering 55 Task Force (IETF), its areas, and its working groups. Note that 56 other groups may also distribute working documents as Internet- 57 Drafts. 59 Internet-Drafts are draft documents valid for a maximum of six 60 months and may be updated, replaced, or obsoleted by other documents 61 at any time. It is inappropriate to use Internet-Drafts as 62 reference material or to cite them other than as "work in progress." 64 The list of current Internet-Drafts can be accessed at 65 http://www.ietf.org/ietf/1id-abstracts.txt 67 The list of Internet-Draft Shadow Directories can be accessed at 68 http://www.ietf.org/shadow.html 70 This Internet-Draft will expire on August 16, 2016. 72 Copyright Notice 74 Copyright (c) 2016 IETF Trust and the persons identified as the 75 document authors. All rights reserved. 77 This document is subject to BCP 78 and the IETF Trust's Legal 78 Provisions Relating to IETF Documents 79 (http://trustee.ietf.org/license-info) in effect on the date of 80 publication of this document. Please review these documents 81 carefully, as they describe your rights and restrictions with 82 respect to this document. 84 Abstract 86 This document considers the consequences of moving the underlying 87 network architecture from the current Internet to an Information- 88 Centric Network (ICN) architecture on video distribution. As most of 89 the traffic in future networks is expected to be video, we consider 90 how to modify the existing video streaming mechanisms. Several 91 important topics related to video distribution over ICN are 92 presented, covering a wide range of scenarios: we look at how to 93 evolve DASH to work over ICN, and leverage the recent ISO/IEC MPEG 94 Dynamic Adaptive Streaming over HTTP (DASH) standard; we consider 95 layered encoding over ICN; P2P mechanisms introduce distinct 96 requirements for video and we look at how to adapt PPSP for ICN; 97 IPTV adds delay constraints, and this will create more stringent 98 requirements over ICN as well. As part of the discussion on video, 99 we discuss DRMs in ICN. Finally, in addition to considering how 100 existing mechanisms would be impacted by ICN, this document lists 101 some research issues to design ICN specific video streaming 102 mechanisms. 104 Table of Contents 105 1. Introduction....................................................... 4 106 2. Conventions used in this document.................................. 5 107 3. Use case scenarios for ICN and Video Streaming..................... 5 108 4. Video download..................................................... 7 109 5. Video streaming and ICN............................................ 7 110 5.1. Introduction to client-driven streaming and DASH ............... 7 111 5.2. Layered Encoding ............................................... 8 112 5.3. Interactions of Video Streaming with ICN ....................... 9 113 5.3.1. Interaction of DASH and ICN ................................ 9 114 5.3.2. Interaction of ICN with Layered Encoding .................. 11 115 5.4. Possible Integration of Video streaming and ICN architecture .. 12 116 5.4.1. DASH over CCN ............................................. 12 117 5.4.2. Testbed, Open Source Tools, and Dataset ................... 14 118 6. P2P video distribution and ICN.................................... 15 119 6.1. Introduction to PPSP .......................................... 15 120 6.2. PPSP over ICN: deployment concepts ............................ 16 121 6.2.1. PPSP short background ..................................... 16 122 6.2.2. From PPSP messages to ICN named-data ...................... 17 123 6.2.3. Support of PPSP interaction through a pull-based ICN API .. 18 124 6.2.4. Abstract layering for PPSP over ICN ....................... 18 125 6.2.5. PPSP interaction with the ICN routing plane ............... 19 126 6.2.6. ICN deployment for PPSP ................................... 20 127 6.3. Impact of MPEG DASH coding schemes ........................... 21 128 7. IPTV and ICN...................................................... 22 129 7.1. IPTV challenges ............................................... 22 130 7.2. ICN benefits for IPTV delivery ................................ 23 131 8. Digital Rights Managements in ICN................................. 25 132 8.1. Broadcast Encryption for DRM in ICN ........................... 25 133 8.2. AAA Based DRM for ICN Networks ................................ 28 134 8.2.1. Overview .................................................. 28 135 8.2.2. Implementation ............................................ 29 136 9. Future Steps for Video in ICN..................................... 30 137 9.1. Large Scale Live Events ....................................... 30 138 9.2. Video Conferencing and Real-Time Communications ............... 30 139 9.3. Store-and-Forward Optimized Rate Adaptation ................... 30 140 9.4. Heterogeneous Wireless Environment Dynamics .................. 31 141 9.5. Network Coding for Video Distribution in ICN .................. 33 142 9.6. Synchronization Issues for Video Distribution in ICN .......... 33 143 10. Security Considerations.......................................... 34 144 11. IANA Considerations.............................................. 34 145 12. Conclusions...................................................... 34 146 13. References....................................................... 35 147 13.1. Normative References ......................................... 35 148 13.2. Informative References ....................................... 35 149 14. Authors' Addresses............................................... 38 150 15. Acknowledgements................................................. 39 152 1. Introduction 154 The unprecedented growth of video traffic has triggered a rethinking 155 of how content is distributed, both in terms of the underlying 156 Internet architecture and in terms of the streaming mechanisms to 157 deliver video objects. 159 In particular, the IRTF ICN working group has been chartered to 160 study new architectures centered upon information; the main 161 contributor to Internet traffic (and information dissemination) is 162 video, and this is expected to stay the same in the near future. If 163 ICN is expected to become prominent, it will have to support video 164 streaming efficiently. 166 As such, it is necessary to discuss along two directions: 168 . Can the current video streaming mechanisms be leveraged and 169 adapted to an ICN architecture? 171 . Can (and should) new, ICN-specific video streaming mechanisms 172 be designed to fully take advantage of the new abstractions 173 exposed by the ICN architecture? 175 This document intends to focus on the first question, in an attempt 176 to define the use cases for video streaming and some requirements. 178 This document focuses on a few scenarios, namely Netflix-like video 179 streaming, peer-to-peer video sharing and IPTV, and identifies how 180 the existing protocols can be adapted to an ICN architecture. In 181 doing so, it also identifies the main issues with these protocols in 182 this ICN context. 184 Some documents have started to consider the ICN-specific 185 requirements of dynamic adaptive streaming [2][3][4][6]. 187 In this document, we give a brief overview of the existing solutions 188 for the selected scenarios. We then consider the interactions of 189 such existing mechanisms with the ICN architecture and list some of 190 the interactions any video streaming mechanism will have to 191 consider. We then identify some areas for future research. 193 2. Conventions used in this document 195 In examples, "C:" and "S:" indicate lines sent by the client and 196 server respectively. 198 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 199 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 200 document are to be interpreted as described in RFC-2119 [RFC2119]. 202 In this document, these words will appear with that interpretation 203 only when in ALL CAPS. Lower case uses of these words are not to be 204 interpreted as carrying RFC-2119 significance. 206 In this document, the characters ">>" preceding an indented line(s) 207 indicates a compliance requirement statement using the key words 208 listed above. This convention aids reviewers in quickly identifying 209 or finding the explicit compliance requirements of this RFC. 211 3. Use case scenarios for ICN and Video Streaming 213 For ICN specific descriptions, we refer to the other working group 214 documents. For our purpose, we assume here that ICN means an 215 architecture where content is retrieved by name and with no binding 216 of content to a specific network location. 218 The consumption of multimedia content comes along with timing 219 requirements for the delivery of the content, for both, live and on- 220 demand consumption. Additionally, real-time use cases such as audio- 221 /video conferencing [7], game streaming, etc., come along with more 222 strict timing requirements. Long startup delays, buffering periods 223 or poor quality, etc., should be avoided to achieve a good Quality 224 of Experience (QoE) to the consumer of the content. (For a 225 definition of QoE in the context of video distribution, please refer 226 to [25]. The working definition is: "Quality of Experience (QoE) is 227 the degree of delight or annoyance of the user of an application or 228 service. It results from the fulfillment of his or her expectations 229 with respect to the utility and / or enjoyment of the application or 230 service in the light of the user's personality and current state.") 232 Of course, these requirements are heavily influenced by routing 233 decisions and caching, which are central parts of ICN and which have 234 to be considered when streaming video in such infrastructures. 236 Due to this range of requirements, we find it useful to narrow the 237 focus on four scenarios (more can be included later): 239 - a video download architecture similar to that if iTtune, where the 240 whole file is being downloaded to the client and can be replayed 241 there multiple times; 242 - a video streaming architecture for playing back movies; this is 243 relevant for the naming and caching aspects of ICN, as well as the 244 interaction with the rate adaptation mechanism necessary to 245 deliver the best QoE to the end-user; 246 - a peer-to-peer architecture for sharing videos; this introduces 247 more stringent routing requirements in terms of locating copies of 248 the content, as the location of the peers evolves and peers join 249 and leave the swarm they use to exchange video chunks; 250 - IPTV; this introduces requirements for multicasting and adds 251 stronger delay constraints. 253 Other scenarios, such as video-conferencing and real-time video 254 communications are not explicitly discussed in this document, while 255 they are in scope. Also, events of mass-media distribution, such as 256 a large crowd in a live event, are also adding new requirements to 257 be included in later version. 259 We discuss how the current state-of-the-art protocols in an IP 260 context can be modified for the ICN architecture. The remainder of 261 this document is organized as follows. In the next section, we 262 consider video download. Then in Section 5, we briefly describe DASH 263 [1], and Layered Encoding (MDC, SVC). P2P is the focus of Section 6, 264 where we describe PPSP. Section 7 highlights the requirements of 265 IPTV, while Section 8 describes the issues of DRM. Section 9 lists 266 some research issues to be solved for ICN-specific video delivery 267 mechanisms. 269 Videoconferencing and real-time video communications will be 270 detailed more in future versions of this document; as well as the 271 mass distribution of content at live large-scale even (stadium, 272 concert hall, etc) for which there is no clearly adopted existing 273 protocol. 275 4. Video download 277 Video download, namely the fetching of a video file from a server or 278 a cache down to the user's local storage, is a natural application 279 of ICN. It should be supported natively without requiring any 280 specific considerations. 282 This is supported now by a host of protocols (say, SCP, FTP, or over 283 HTTP), which would need to be replaced by the protocols to retrieve 284 content in ICNs. 286 However, current mechanisms are built atop existing transport 287 protocols. Some ICN proposals (say, CCN or NDN for instance) attempt 288 to leverage the work done upon these transport protocol and it has 289 been proposed to use mechanisms such as the TCP congestion window 290 (and the associated Adaptive Increase, Multiplicative Decrease - 291 AIMD) to decide how many object requests ("interests" in CCN/NDN 292 terminology) should be in flight at any point in time. 294 It should be noted that ICN intrinsically supports different 295 transport mechanisms, which could achieve better performance than 296 TCP, as they subsume TCP into a special case. For instance, one 297 could imagine a link-by-link transport coupled with caching. This is 298 enabled by the ICN architecture, and would facilitate the point-to- 299 point download of video files. 301 5. Video streaming and ICN 303 5.1. Introduction to client-driven streaming and DASH 305 Media streaming over the hypertext transfer protocol (HTTP) and in a 306 further consequence streaming over the transmission control protocol 307 (TCP) has become omnipresent in today's Internet. Content providers 308 such as Netflix, Hulu, and Vudu do not deploy their own streaming 309 equipment but use the existing Internet infrastructure as it is and 310 they simply deploy their own services over the top (OTT). This 311 streaming approach works surprisingly well without any particular 312 support from the underlying network due to the use of efficient 313 video compression, content delivery networks (CDNs), and adaptive 314 video players. Earlier video streaming research mostly recommended 315 to use the user datagram protocol (UDP) combined with the real time 316 transport protocol (RTP). It assumed it would not be possible to 317 transfer multimedia data smoothly with TCP, because of its 318 throughput variations and large retransmission delays. This point of 319 view has significantly evolved today. HTTP streaming, and especially 320 its most simple form known as progressive download, has become very 321 popular over the past few years because it has some major benefits 322 compared to RTP streaming. As a consequence of the consistent use of 323 HTTP for this streaming method, the existing Internet 324 infrastructure, consisting of proxies, caches and CDNs, could be 325 used. Originally, this architecture was designed to support best 326 effort delivery of files and not real time transport of multimedia 327 data. Nevertheless, real time streaming based on HTTP could also 328 take advantage of this architecture, in comparison to RTP, which 329 could not leverage any of the aforementioned components. Another 330 benefit that results from the use of HTTP is that the media stream 331 could easily pass firewalls or network address translation (NAT) 332 gateways, which was definitely a key for the success of HTTP 333 streaming. However, HTTP streaming is not the holy grail of 334 streaming as it also introduces some drawbacks compared to RTP. 335 Nevertheless, in an ICN-based video streaming architecture these 336 aspects also have to be considered. 338 The basic concept of DASH [1] is to use segments of media content, 339 which can be encoded at different resolutions, bit rates, etc., as 340 so-called representations. These segments are served by conventional 341 HTTP Web servers and can be addressed via HTTP GET requests from the 342 client. As a consequence, the streaming system is pull-based and the 343 entire streaming logic is located on the client, which makes it 344 scalable, and allows to adapt the media stream to the client's 345 capabilities. 347 In addition to this, the content can be distributed using 348 conventional CDNs and their HTTP infrastructure, which also scales 349 very well. In order to specify the relationship between the 350 contents' media segments and the associated bit rate, resolution, 351 and timeline, the Media Presentation Description (MPD) is used, 352 which is a XML document. The MPD refers to the available media 353 segments using HTTP URLs, which can be used by the client for 354 retrieving them. 356 5.2. Layered Encoding 358 Another approach for video streaming consist in using layered 359 encoding. Namely, scalable video coding formats the video stream 360 into different layers: a base layer which can be decoded to provide 361 the lowest bit rate for the specific stream, and enhancement layers 362 which can be transmitted separately if network conditions allow. The 363 higher layers offer higher resolutions and enhancement of the video 364 quality, while the layered approach allows to adapt to the network 365 conditions. This is used in MPEG-4 scalable profile or H.263+. 367 H264SVC is available, but not much deployed. JPEG2000 has a wavelet 368 transform approach for layered encoding, but has not been deployed 369 much either. 371 It is not clear if the layered approach is fine-grained enough for 372 rate control. 374 5.3. Interactions of Video Streaming with ICN 376 5.3.1. Interaction of DASH and ICN 378 Video streaming, and DASH in particular, have been designed with 379 goals that are aligned with that of most ICN proposals. Namely, it 380 is a client-based mechanism, which requests items (in this case, 381 chunks of a video stream) by name. 383 ICN and MPEG-DASH [1] have several elements in common: 385 - the client-initiated pull approach; 386 - the content being dealt with in pieces (or chunks); 387 - the support of efficient replication and distribution of content 388 pieces within the network; 389 - the scalable, session-free nature of the exchange between the 390 client and the server at the streaming layer: the client is free 391 to request any chunk from any location; 392 - the support for potentially multiple source locations. 394 For the last point, DASH may list multiple source URLs in a 395 manifest, and ICN is agnostic to the location of a copy it is 396 receiving. We do not imply that current video streaming mechanisms 397 attempt to draw the content from multiple sources concurrently. This 398 is a potential benefit of ICN, but is not considered in the current 399 approaches mentioned in this document. 401 As ICN is a promising candidate for the Future Internet (FI) 402 architecture, it is useful to investigate its suitability in 403 combination with multimedia streaming standards like MPEG-DASH. In 404 this context, the purpose of this section is to present the usage of 405 ICN instead of HTTP in MPEG-DASH 407 However, there are some issues that arise from using a dynamic rate 408 adaptation mechanism in an ICN architecture (note that some of the 409 issues are related to caching, and not necessarily unique to ICN): 411 o Naming of the data in DASH does not necessarily follow the ICN 412 convention of any of the ICN proposals. Several chunks of the 413 same video stream might currently go by different names that for 414 instance do not share a common prefix. There is a need to 415 harmonize the naming of the chunks in DASH with the naming 416 conventions of the ICN. The naming convention of using a 417 filename/time/encoding format could for instance be made 418 compatible with the convention of CCN. 420 o While chunks can be retrieved from any server, the rate 421 adaptation mechanism attempts to estimate the available network 422 bandwidth so as to select the proper playback rate and keep its 423 playback buffer at the proper level. Therefore, there is a need 424 to either include some location semantics in the data chunks so 425 as to properly assess the throughput to a specific location; or 426 to design a different mechanism to evaluate the available network 427 bandwidth. 429 o The typical issue of access control and accounting happens in 430 this context, where chunks can be cached in the network outside 431 of the administrative control of the content publisher. It might 432 be a requirement from the owner of the video stream that access 433 to these data chunks needs to be accounted/billed/monitored. 435 o Dynamic streaming multiplies the representations of a given video 436 stream, therefore diminishing the effectiveness of caching: 437 namely, to get a hit for a chunk in the cache, it has to be for 438 the same format and encoding values. Alternatively, to get the 439 same hit rate as for a stream using a single encoding, the cache 440 size must be scaled up to include all the possible 441 representations. 443 o Caching introduces oscillatory dynamics as it may modify the 444 estimation of the available bandwidth between the end user and 445 the repository where it is getting the chunks from. For instance, 446 if an edge cache holds a low resolution representation near the 447 user, the user getting this low resolution chunks will observe a 448 good performance, and will then request higher resolution chunks. 449 If those are hosted on a server with poor performance, then the 450 client would have to switch back to the low representation. This 451 oscillation may be detrimental to the perceived QoE of the user. 453 o The ICN transport mechanism needs to be compatible to some extent 454 with DASH. To take a CCN example, the rate at which interests are 455 issued should be such that the chunks received in return arrive 456 fast enough and with the proper encoding to keep the playback 457 buffer above some threshold. 459 o The usage of multiple network interfaces is possible in ICN, 460 enabling a seamless handover between them. For the combination 461 with DASH, an intelligent strategy which should focus on traffic 462 load balancing between the available links may be necessary. This 463 would increase the effective media throughput of DASH by 464 leveraging the combined available bandwidth of all links, 465 however, it could potentially lead to high variations of the 466 media throughput. 468 o DASH does not define how the MPD is retrieved; hence, this is 469 compatible with CCN. However, the current profiles defined within 470 MPEG-DASH require the MPD to contain HTTP-URLs (incl. http and 471 https URI schemes) to identify segments. To enable a more 472 integrated approach as described in this document, an additional 473 profile for DASH over CCN has to be defined, enabling ICN/CCN- 474 based URIs to identify and request the media segments. 476 We describe in Section 5.4 a potential implementation of a dynamic 477 adaptive video stream over ICN, based upon DASH and CCN [5]. 479 5.3.2. Interaction of ICN with Layered Encoding 481 Issues of interest to an Information-Centric network architecture in 482 the context of layered video streaming include: 484 . Caching of the multiple layers. The caching priority should go 485 to the base layer, and defining caching policy to decide when 486 to cache enhancement layers; 487 . Synchronization of multiple content streams, as the multiple 488 layers may come from different sources in the network (for 489 instance, the base layer might be cached locally while the 490 enhancement layers may be stored in the origin server). There 491 are both intra-layer synchronization (for the layers of the 492 same video or audio stream) but also inter-stream 493 synchronization to address, for synchronizing the audio and the 494 video streams (see Section 9 for other synchronization aspects 495 to be included in the "Future Steps for Video in ICN"); 496 . Naming of the different layers: when the client requests an 497 object, the request can be satisfied with the base layer alone, 498 aggregated with enhancement layers. Should one request be 499 sufficient to provide different streams? In a CCN architecture 500 for instance, this would violate a one interest-one data packet 501 principle and the client would need to specify each layer it 502 would like to receive. In a Pub/Sub architecture, the 503 rendezvous point would have to make a decision as to which 504 layers (or which pointer to which layer's location) to return. 506 5.4. Possible Integration of Video streaming and ICN architecture 508 5.4.1. DASH over CCN 510 DASH is intended to enable adaptive streaming, i.e., each content 511 piece can be provided in different qualities, formats, languages, 512 etc., to cope with the diversity of todays' networks and devices. As 513 this is an important requirement for Future Internet proposals like 514 CCN, the combination of those two technologies seems to be obvious. 515 Since those two proposals are located at different protocol layers - 516 DASH at the application and CCN at the network layer - they can be 517 combined very efficiently to leverage the advantages of both and 518 potentially eliminate existing disadvantages. As CCN is not based on 519 classical host-to-host connections, it is possible to consume 520 content from different origin nodes as well as over different 521 network links in parallel, which can be seen as an intrinsic error 522 resilience feature w.r.t. the network. This is a useful feature of 523 CCN for adaptive multimedia streaming within mobile environments 524 since most mobile devices are equipped with multiple network links 525 like 3G and WiFi. CCN offers this functionality out of the box which 526 is beneficial when used for DASH-based services. In particular, it 527 is possible to enable adaptive video streaming handling both 528 bandwidth and network link changes. That is, CCN handles the network 529 link decision and DASH is implemented on top of CCN to adapt the 530 video stream to the available bandwidth. 532 In principle, there are two options to integrate DASH and CCN: a 533 proxy service acting as a broker between HTTP and CCN as proposed in 534 [6], and the DASH client implementing a native CCN interface. The 535 former transforms an HTTP request to a corresponding interest packet 536 as well as a data packet back to an HTTP response, including 537 reliable transport as offered by TCP. This may be a good compromise 538 to implement CCN in a managed network and to support legacy devices. 539 Since such a proxy is already described in [6] this draft focuses on 540 a more integrated approach, aiming at fully exploiting the potential 541 of a CCN DASH Client. That is, we describe a native CCN interface 542 within the DASH client, which adopts a CCN naming scheme (CCN URIs) 543 to denote segments in the Media Presentation Description (MPD). In 544 this architecture, only the network access component on the client 545 has to be modified and the segment URIs within MPD have to be 546 updated according to the CCN naming scheme. 548 Initially, the DASH client retrieves the MPD containing the CCN URIs 549 of the content representations including the media segments. The 550 naming scheme of the segments may reflect intrinsic features of CCN 551 like versioning and segmentation support. Such segmentation support 552 is already compulsory for multimedia streaming in CCN and, thus, can 553 also be leveraged for DASH-based streaming over CCN. The CCN 554 versioning can be adopted in a further step to signal different 555 representations of the DASH-based content, which enables an implicit 556 adaptation of the requested content to the clients' bandwidth 557 conditions. That is, the interest packet already provides the 558 desired characteristics of a segment (such as bit rate, resolution, 559 etc.) within the content name (or potentially within parameters 560 defined as extra types in the packet formats). Additionally, if 561 bandwidth conditions of the corresponding interfaces or routing 562 paths allow so, DASH media segments could be aggregated 563 automatically by the CCN nodes, which reduces the amount of interest 564 packets needed to request the content. However, such approaches need 565 further research, specifically in terms of additional intelligence 566 and processing power needed at the CCN nodes. 568 After requesting the MPD, the DASH client will start to request 569 particular segments. Therefore, CCN interest packets are generated 570 by the CCN access component and forwarded to the available 571 interfaces. Within the CCN, these interest packets leverage the 572 efficient interest aggregation for, e.g., popular content, as well 573 as the implicit multicast support. Finally, the interest packets are 574 satisfied by the corresponding data packets containing the video 575 segment data, which are stored on the origin server or any CCN node, 576 respectively. With an increasing popularity of the content, it will 577 be distributed across the network resulting in lower transmission 578 delays and reduced bandwidth requirements for origin servers and 579 content providers respectively. 581 With the extensive usage of in-network caching, new drawbacks are 582 introduced since the streaming logic is located at the client, i.e., 583 clients are not aware of each other and the network infrastructure 584 and cache states. Furthermore, negative effects are introduced when 585 multiple clients are competing for a bottleneck and when caching is 586 influencing this bandwidth competition. As mentioned above, the 587 clients request individual portions of the content based on 588 available bandwidth, which is calculated using throughput 589 estimations. This uncontrolled distribution of the content 590 influences the adaptation process of adaptive streaming clients. The 591 impact of this falsified throughput estimation could be tremendous 592 and leads to a wrong adaptation decision which may impact the 593 Quality of Experience (QoE) at the client, as shown in [8]. In ICN, 594 the client does not have the knowledge from which source the 595 requested content is actually served or how many origin servers of 596 the content are available, as this is transparent and depends on the 597 name-based routing. This introduces the challenge that the 598 adaptation logic of the adaptive streaming client is not aware of 599 the event when the ICN routing decides to switch to a different 600 origin server or content is coming through a different 601 link/interface. As most algorithms implementing the adaption logic 602 are using bandwidth measurements and related heuristics, the 603 adaptation decisions are no longer valid when changing origin 604 servers (or links) and potentially cause playback interruptions and, 605 consequently, stalling. Additionally, ICN supports the usage of 606 multiple interfaces and a seamless handover between them, which 607 again comes together with bandwidth changes, e.g., switching between 608 fixed and wireless, 3G/4G and WiFi networks, support of different 609 types of servers (say with/without SSD, or hardware acceleration), 610 etc. Considering these characteristics of ICN, adaptation algorithms 611 merely based on bandwidth measurements are not appropriate anymore, 612 as potentially each segment can be transferred from another ICN node 613 or interface, all with different bandwidth condition. Thus, 614 adaptation algorithms taking into account these intrinsic 615 characteristics of ICN are preferred over algorithms based on mere 616 bandwidth measurements. 618 5.4.2. Testbed, Open Source Tools, and Dataset 620 For the evaluations of DASH over CCN, a testbed with open source 621 tools and datasets is provided in [9]. In particular, it provides 622 two client player implementations, (i) a libdash extension for DASH 623 over CCN and (ii) a VLC plugin implementing DASH over CCN. For both 624 implementations the CCNx implementation has been used as a basis. 626 The general architecture of libdash is organized in modules, so that 627 the library implements a MPD parser and an extensible connection 628 manager. The library provides object-oriented interfaces for these 629 modules to access the MPD and the downloadable segments. These 630 components are extended to support DASH over CCN and available in a 631 separate development branch of the github project available at 632 http://www.github.com/bitmovin/libdash. libdash comes together with 633 a fully featured DASH player with a QT-based frontend, demonstrating 634 the usage of libdash and providing a scientific evaluation platform. 635 As an alternative, patches for the DASH plugin of the VLC player are 636 provided. These patches can be applied to the latest source code 637 checkout of VLC resulting in a DASH over CCN-enabled VLC player. 639 Finally, a DASH over CCN dataset is provided in form of a CCNx 640 repository. It includes 15 different quality representation of the 641 well-known Big Buck Bunny Movie, ranging from 100 kbps up to 4500 642 kbps. The content is split into segments of two seconds, and 643 described by an associated MPD using the presented naming scheme in 644 Section 4.1. This repository can be downloaded from [9], and is also 645 provided by a public accessible CCNx node. Associated routing 646 commands for the CCNx namespaces of the content are provided via 647 scripts coming together with the dataset and can be used as a public 648 testbed. 650 6. P2P video distribution and ICN 652 Another form of distributing content - and video in particular- 653 which ICNs need to support is Peer-to-Peer distribution (P2P). We 654 see now how an existing protocol such as PPSP can be modified to 655 work in an ICN environment. 657 6.1. Introduction to PPSP 659 P2P video Streaming (PPS) is a popular approach to redistribute live 660 media over Internet. The proposed P2PVS solutions can be roughly 661 classified in two classes: 663 - Push/Tree based 665 - Pull/Mesh based 667 The Push/Tree based solution creates an overlay network among peers 668 that has a tree shape [30]. Using a progressive encoding (e.g. 669 Multiple Description Coding or H.264 Scalable Video Coding), 670 multiple trees could be set up to support video rate adaptation. On 671 each tree an enhancement stream is sent. The higher the number of 672 received streams, the higher the video quality. A peer controls the 673 video rate by fetching or not the streams delivered over the 674 distribution trees. 676 The Pull/Mesh based solution is inspired by the BitTorrent file 677 sharing mechanism. A Tracker collects information about the state of 678 the swarm (i.e. set of participating peers). A peer forms a mesh 679 overlay network with a subset of peers, and exchange data with them. 680 A peer announces what data items it disposes and requests missing 681 data items that are announced by connected peers. In case of live 682 streaming, the involved data set includes only a recent window of 683 data items published by the source. Also in this case, the use of a 684 progressive encoding can be exploited for video rate adaptation. 686 Pull/Mesh based P2PVS solutions are the more promising candidate for 687 the ICN deployment, since most of ICN approach provides a pull-based 688 API [5][10][11][12]. In addition, Pull/Mesh based P2PVS are more 689 robust than Push/Tree based one [13] and the Peer to Peer Streaming 690 Protocol (PPSP) working group [14] is also proposing a Pull/Mesh 691 based solution. 693 +------------------------------------------------+ 694 | | 695 | +--------------------------------+ | 696 | | Tracker | | 697 | +--------------------------------+ | 698 | | ^ ^ | 699 |Tracker | | Tracker |Tracker | 700 |Protocol| | Protocol |Protocol | 701 | | | | | 702 | V | | | 703 | +---------+ Peer +---------+ | 704 | | Peer |<----------->| Peer | | 705 | +---------+ Protocol +---------+ | 706 | | ^ | 707 | | |Peer | 708 | | |Protocol | 709 | V | | 710 | +---------------+ | 711 | | Peer | | 712 | +---------------+ | 713 | | 714 +------------------------------------------------+ 715 Figure 1: PPSP System Architecture (source [RFC6972]) 717 Figure 1 reports the PPSP architecture presented in [RFC6972]. PEERs 718 announce and share video chunks and a TRACKER maintains a list of 719 PEERs participating in a specific audio/video channel or in the 720 distribution of a streaming file. The tracker functionality may be 721 centralized in a server or distributed over the PEERs. PPSP 722 standardize the Peer and Tracker Protocols, which can run directly 723 over UDP or TCP. 725 This document discusses some preliminary concepts about the 726 deployment of PPSP on top of an ICN that exposes a pull-based API, 727 meanwhile considering the impact of MPEG DASH streaming format. 729 6.2. PPSP over ICN: deployment concepts 731 6.2.1. PPSP short background 733 PPSP specifies peer protocol (PPSPP) [15] and tracker protocol 734 (PPSP-TP)[16]. 736 Some of the operations carried out by the tracker protocol are the 737 followings. When a peer wishes to join the streaming session it 738 contacts the Tracker (CONNECT message), obtains a PEER_ID and a list 739 of PEER_IDs (and IP addresses) of other peers that are participating 740 to the SWARM and that the tracker has singled out for the requesting 741 peer (this may be a subset of the all peers of the SWARM). In 742 addition to this join operation, a peer may contact the tracker to 743 request to renew the list of participating peers (FIND message), to 744 periodically update its status to the tracker (STAT_REPORT message), 745 etc. 747 Some of the operations carried out by the peer protocol are the 748 following. Using the list of peers delivered by the tracker, a peer 749 establishes a session with them (HANDSHAKE message). A peer 750 periodically announces to neighboring peers which chunks it has 751 available for download (HAVE message). Using these announcements, a 752 peer requests missing chunks from neighboring peers (REQUEST 753 messages), which will send back them (DATA message). 755 6.2.2. From PPSP messages to ICN named-data 757 An ICN provides users with data items exposed by names. The bundle 758 name and data item is usually referred as named-data, named-content, 759 etc. To transfer PPSP messages though an ICN the messages should be 760 wrapped as named-data items, and receivers should request them by 761 name. 763 A PPSP entity receives messages from peers and/or tracker. Some 764 operations require gathering the messages generated by another 765 specific host (peer or tracker). For instance, if a peer A wishes to 766 gain information about video chunks available from peer B, the 767 former shall fetch the PPSP HAVE messages specifically generated by 768 the later. We refer to these kinds of named-data as "located-named- 769 data", since they should be gathered from a specific location (e.g. 770 peer B). 772 For other PPSP operations, like to fetch a DATA message (i.e. a 773 video chunk), what it is relevant for a peer is just to receive the 774 requested content, independently from who is the endpoint that 775 generate the data. We refer to this information with the generic 776 term "named-data". 778 The naming scheme differentiates named-data and located-named-data 779 items. In case of named-data, the naming scheme only includes a 780 content identifier (e.g. the name of the video chunk), without any 781 prefix identifying who provides the content. For instance, a DATA 782 message containing the video chunk #1 may be named as 783 "ccnx:/swarmID/chunk/chunkID", where swarmID is a unique identifier 784 of the streaming session, "chunk" is a keyword and chunkID is the 785 chunk identifier (e.g. a integer number). 787 In case of located-named-data, the naming scheme includes a 788 location-prefix, which uniquely identifies the host generating the 789 data item. This prefix may be the PEER_ID in case the host was a 790 peer or a tracker identifier in case the host was the tracker. For 791 instance, a HAVE message generated by a peer B may be named as 792 "ccnx:/swarmID/peer/PEER_ID/HAVE", where "peer" is a keyword, 793 PEER_ID_B is the identifier of peer B and HAVE is a keyword. 795 6.2.3. Support of PPSP interaction through a pull-based ICN 796 API 798 The PPSP procedures are based both on pull and push interactions. 799 For instance, the distribution of chunks availability can be 800 classified as a push-based operation, since a peer sends an 801 "unsolicited" information (HAVE message) to neighboring peers. 802 Conversely the procedure used to receive video chunks can be 803 classified as pull-based, since it is supported by a 804 request/response interaction (i.e. REQUEST, DATA messages). 806 As we said, we refer to an ICN architecture which provides a pull- 807 based API. Accordingly, the mapping of PPSP pull-based procedure is 808 quite simple. For instance, using the CCN architecture [5] a PPSP 809 DATA message may be carried by a CCN Data message and a REQUEST 810 message can transferred by a CCN Interest. 812 Conversely, the support of push-based PPSP operations may be more 813 difficult. We need of an adaptation functionality that carries out a 814 push-based operation using the underlying pull-based service 815 primitives. For instance, a possible approach is to use the 816 request/response (i.e. Interest/Data) four ways handshakes proposed 817 in [7]. Another possibility is that receivers periodically send out 818 request messages of the named-data that neighbors will push and, 819 when available, sender inserts the pushed data within a response 820 message. 822 6.2.4. Abstract layering for PPSP over ICN 824 +-----------------------------------+ 825 | Application | 826 +-----------------------------------+ 827 | PPSP (TCP/IP) | 828 +-----------------------------------+ 829 | ICN - PPSP Adaptation Layer (AL) | 830 +-----------------------------------+ 831 | ICN Architecture | 832 +-----------------------------------+ 833 Figure 2: Mediator approach 835 Figure 2 provides a possible abstract layering for PPSP over ICN. 836 The Adaptation Layer acts as a mediator (proxy) between legacy PPSP 837 entities based on TCP/IP and the ICN architecture. In facts, the 838 role the mediator is to use ICN to transfer PPSP legacy messages. 840 This approach makes possible to merely reuse TCP/IP P2P applications 841 whose software includes also PPSP functionality. This "all-in-one" 842 development approach may be rather common since the PPSP-Application 843 interface is not going to be specified. Moreover, if the Operating 844 System will provide libraries that expose a PPSP API, these will be 845 initially based on a underlying TCP/IP API. Also in this case, the 846 mediator approach would make possible to easily reuse both the PPSP 847 libraries and the Application on top of an ICN. 849 +-----------------------------------+ 850 | Application | 851 +-----------------------------------+ 852 | ICN-PPSP | 853 +-----------------------------------+ 854 | ICN Architecture | 855 +-----------------------------------+ 857 Figure 3: Clean-slate approach 859 Figure 3 sketches a clean-slate layering approach in which the 860 application directly includes or interacts with a PPSP version based 861 on ICN. Likely such a PPSP_ICN integration could yield a simplier 862 development, also because it does not require implementing a TCP/IP 863 to ICN translation as in the Mediator approach. However, the clean- 864 slate approach requires developing the application (in case of 865 embedded PPSP functionality) or the PPSP library from scratch, 866 without exploiting what might already exist for TCP/IP. 868 Overall, the Mediator approach may be considered as the first step 869 of a migration path towards ICN native PPSP applications. 871 6.2.5. PPSP interaction with the ICN routing plane 873 Upon the ICN API a user (peer) requests a content and the ICN sends 874 it back. The content is gathered by the ICN from any source, which 875 could be the closest peer that disposes of the named-data item, an 876 in-network cache, etc. Actually, "where" to gather the content is 877 controlled by an underlying ICN routing plane, which sets up the ICN 878 forwarding tables (e.g. CCN FIB [5]). 880 A cross-layer interaction between the ICN routing plane and the PPSP 881 may be required to support a PPSP session. Indeed, ICN shall forward 882 request messages (e.g. CCN Interest) towards the proper peer that 883 can handle them. Depending on the layering approach, this cross- 884 layer interaction is controlled either by the Adaptation Layer or by 885 the ICN-PPSP. For example, if a peer A receives a HAVE message 886 indicating that peer B disposes of the video chunk named 887 "ccnx:/swarmID/chunk/chunkID", then former should insert in its ICN 888 forwarding table an entry for the prefix 889 "ccnx:/swarmID/chunk/chunkID" whose next hop locator (e.g. IP 890 address) is the network address of peer B [17]. 892 6.2.6. ICN deployment for PPSP 894 The ICN functionality that supports a PPSP session may be "isolated" 895 or "integrated" with the one of a public ICN. 897 In the isolated case, a PPSP session is supported by an instance of 898 an ICN (e.g. deployed on top of IP), whose functionalities operate 899 only on the limited set of nodes participating to the swarm, i.e. 900 peers and the tracker. This approach resembles the one followed by 901 current P2P application, which usually form an overlay network among 902 peers of a P2P application. And intermediate public IP routers do 903 not carry out P2P functionalities. 905 In the integrated case, the nodes of a public ICN may be involved in 906 the forwarding and in-network caching procedures. In doing so, the 907 swarm may benefit from the presence of in-network caches so limiting 908 uplink traffic on peers and inter-domain traffic too. These are 909 distinctive advantages of using PPSP over a public ICN, rather than 910 over TCP/IP. In addition, such advantages aren't likely manifested 911 in the case of isolated deployment. 913 However, the possible interaction between the PPSP and the routing 914 layer of a public ICN may be dramatic, both in terms of explosion of 915 the forwarding tables and in terms of security. These issues 916 specifically take place for those ICN architectures for which the 917 name resolution (i.e. name to next-hop) occurs en-route, like the 918 CCN architecture. 920 For instance, using the CCN architecture, to fetch a named-data item 921 offered by a peer A the on-path public ICN entities have to route 922 the request messages towards the peer A. This implies that the ICN 923 forwarding tables of public ICN nodes may contain many entries, e.g. 925 one entry per video chunk, and these entries are difficult to be 926 aggregated since peers avail sparse parts of a big content, whose 927 names have a same prefix (e.g. "ccnx:/swarmID"). Another possibility 928 is to wrap all PPSP messages into a located-named-data. In this case 929 the forwarding tables should contain "only" the PEER_ID prefixes 930 (e.g. "ccnx:/swarmID/peer/PEER_ID"), so scaling down the number of 931 entries from number of chunks to number of peers. However, in this 932 case the ICN mechanisms recognize a same video chunk offered by 933 different peers as different contents, so vanishing caching and 934 multicasting ICN benefits. Moreover, in any case routing entries 935 should be updated either the base of the availability of named-data 936 items on peers or on the presence of peers, and these events in a 937 P2P session is rapidly changing so possibly hampering the 938 convergence of the routing plane. Finally, since peers have an 939 impact on the ICN forwarding table of public nodes, this may open 940 obvious security issues. 942 6.3. Impact of MPEG DASH coding schemes 944 The introduction of video rate adaptation may valuably decrease the 945 effectiveness of P2P cooperation and of in-network caching, 946 depending of the kind of the video coding used by the MPEG DASH 947 stream. 949 In case of a MPEG DASH streaming with MPEG AVC encoding, a same 950 video chunk is independently encoded at different rates and the 951 encoding output is a different file for each rate. For instance, in 952 case of a video encoded at three different rates R1,R2,R3, for each 953 segment S we have three distinct files: S.R1, S.R2, S.R3. These 954 files are independent of each other. To fetch a segment coded at R2 955 kbps, a peer shall request the specific file S.R2. Receiver-driven 956 algorithms, implemented by the video client, usually handle the 957 estimation of the best coding rate. 959 The independence among files associated to different encoding rates 960 and the heterogeneity of peer bandwidths, may dramatically reduce 961 the interaction among peers, the effectiveness of in-network caching 962 (in case of integrated deployment), and consequently the ability of 963 PPSP to offload the video server (i.e. a seeder peer). Indeed, a 964 peer A may select a coding rate (e.g. R1) different from the one 965 selected by a peer B (e.g. R2) and this prevents the former to fetch 966 video chunks from the later, since peer B avails of chunks coded at 967 a rate different from the ones needed by A. To overcome this issue, 968 a common distributed rate selection algorithm could force peers to 969 select the same coding rate [17]; nevertheless this approach may be 970 not feasible in the in case of many peers. 972 The use of SVC encoding (Annex G extension of the H.264/MPEG-4 AVC 973 video compression standard) should make rate adaptation possible, 974 meanwhile neither reducing peer collaborations nor the in-network 975 caching effectiveness. For a single video chunk, a SVC encoder 976 produces different files for the different rates (roughly "layers"), 977 and these files are progressively related each other. Starting from 978 a base-layer which provides the minimum rate encoding, the next 979 rates are encoded as an "enhancement layer" of the previous one. For 980 instance, in case the video is coded with three rates R1 (base- 981 layer), R2 (enhancement-layer n.1), R3 (enhancement-layer n.2), then 982 for each DASH segment we have three files S.R1, S.R2 and S.R3. The 983 file S.R1 is the segment coded at the minimum rate (base-layer). The 984 file S.R2 enhances S.R1, so as S.R1 and S.R2 can be combined to 985 obtain a segment coded at rate R2. To get a segment coded at rate 986 R2, a peer shall fetch both S.R1 and S.R2. This progressive 987 dependence among files that encode a same segment at different rates 988 makes peer cooperation possible, also in case peers player have 989 autonomously selected different coding rates. For instance, if peer 990 A has selected the rate R1, the downloaded files S.R1 are useful 991 also for a peer B that has selected the rate R2, and vice versa. 993 7. IPTV and ICN 995 7.1. IPTV challenges 997 IPTV refers to the delivery of quality content broadcast over the 998 Internet, and is typically associated with strict quality 999 requirements, i.e., with a perceived latency of less than 500 ms and 1000 a packet loss rate that is multiple orders lower than the current 1001 loss rates experienced in the most commonly used access networks 1002 (see [31]). We can summarize the major challenges for the delivery 1003 of IPTV service as follows. 1005 Channel change latency represents a major concern for the IPTV 1006 service. Perceived latency during channel change should be less than 1007 500ms. To achieve this objective over the IP infrastructure, we have 1008 multiple choices: 1010 (i) receiving fast unicast streams from a dedicated server (most 1011 effective but not resource efficient); 1012 (ii) connecting to other peers in the network (efficiency depends 1013 on peer support, effective and resource efficient, if also 1014 supported with a dedicated server); 1016 (iii) connecting to multiple multicast sessions at once (effective 1017 but not resource efficient, and depends on the accuracy of 1018 the prediction model used to track user activity). 1020 The second major challenge is the error recovery. Typical IPTV 1021 service requirements dictate the mean time between artifacts to be 1022 approximately 2 hours (see [31]). This suggests the perceived loss 1023 rate to be around or less than 10^-7. Current IP-based solutions 1024 rely on the following proactive and reactive recovery techniques: 1025 (i) joining the FEC multicast stream corresponding to the perceived 1026 packet loss rate (not efficient as the recovery strength is chosen 1027 based on worst-case loss scenarios), (ii) making unicast recovery 1028 requests to dedicated servers (requires active support from the 1029 service provider), (iii) probing peers to acquire repair packets 1030 (finding matching peers and enabling their cooperation is another 1031 challenge). 1033 7.2. ICN benefits for IPTV delivery 1035 ICN presents significant advantages for the delivery of IPTV 1036 traffic. For instance, ICN inherently supports multicast and allows 1037 for quick recovery from packet losses (with the help of in-network 1038 caching). Similarly, peer support is also provided in the shape of 1039 in-network caches that typically act as the middleman between two 1040 peers, enabling therefore earlier access to IPTV content. 1042 However, despite these advantages, delivery of IPTV service over 1043 Information Centric Networks brings forth new challenges. We can 1044 list some of these challenges as follows: 1046 . Messaging overhead: ICN is a pull-based architecture and relies 1047 on a unique balance between requests and responses. A user 1048 needs to make a request for each data packet. In the case of 1049 IPTV, with rates up to, and likely to be, above 15Mbps, we 1050 observe significant traffic upstream to bring those streams. As 1051 the number of streams increases (including the same session at 1052 different quality levels and other formats), so does the burden 1053 on the routers. Even if the majority of requests are aggregated 1054 at the core, routers close to the edge (where we observe the 1055 biggest divergence in user requests) will experience a 1056 significant increase in overhead to process these requests. The 1057 same is true at the user side, as the uplink usage multiplies 1058 in the number of sessions a user requests (for instance, to 1059 minimize the impact of bandwidth fluctuations). 1060 . Cache control: As the IPTV content expires at a rapid rate 1061 (with a likely expiry threshold of 1s), we need solutions to 1062 effectively flush out such content to also prevent degradation 1063 impact on other cached content, with the help of intelligently 1064 chosen naming conventions. However, to allow for fast recovery 1065 and optimize access time to sessions (from current or new 1066 users), the timing of such expirations needs to be adaptive to 1067 network load and user demand. However, we also need to support 1068 quick access to earlier content, whenever needed, for instance, 1069 when the user accesses the rewind feature (note that in-network 1070 caches will not be of significant help in such scenarios due to 1071 overhead required to maintain such content). 1072 . Access accuracy: To receive the up-to-date session data, users 1073 need to be aware of such information at the time of their 1074 request. Unlike IP multicast, since the users join a session 1075 indirectly, session information is critical to minimize 1076 buffering delays and reduce the startup latency. Without such 1077 information, and without any active cooperation from the 1078 intermediate routers, stale data can seriously undermine the 1079 efficiency of content delivery. Furthermore, finding a cache 1080 does not necessarily equate to joining a session, as the look- 1081 ahead latency for the initial content access point may have a 1082 shorter lifetime than originally intended. For instance, if the 1083 user that has initiated the indirect multicast leaves the 1084 session early, the requests from the remaining users need to 1085 experience an additional latency of one RTT as they travel 1086 towards the content source. If the startup latency is chosen 1087 depending on the closeness to the intermediate router, going to 1088 the content source in-session can lead to undesired pauses. 1090 It should be noted that IPTV includes more than just multicast. Many 1091 implementations include "trick plays" (fast forward, pause, rewind) 1092 that often transform a multicast session into multiple unicast 1093 sessions. In this context, ICN is beneficial, as the caching offers 1094 an implicit multicast, but without tight synchronization constraints 1095 in between two different users. One user may rewind, and start 1096 playing forward again, drawing from a nearby cache of the content 1097 recently viewed by another user (whereas in a strict multicast 1098 session, the opportunity of one user lagging off behind would be 1099 more difficult to implement). 1101 8. Digital Rights Managements in ICN 1103 This section discusses the need for Digital Rights Management (DRM) 1104 functionalities for multimedia streaming over ICN. It focuses on two 1105 possible approaches: modifying AAA to support DRM in ICN, and using 1106 Broadcast Encryption. 1108 It is assumed that ICN will be used heavily for digital content 1109 dissemination. It is vital to consider DRM for digital content 1110 distribution. In today's Internet there are two predominant classes 1111 of business models for on-demand video streaming. The first model is 1112 based on advertising revenues. Non-copyright protected (usually 1113 user-generated content, UGC) is offered by large infrastructure 1114 providers like Google (YouTube) at no charge. The infrastructure is 1115 financed by spliced advertisements into the content. In this context 1116 DRM considerations may not be required, since producers of UGC may 1117 only strive for the maximum possible dissemination. Some producers 1118 of UGC are mainly interested to share content with their families, 1119 friends, colleges or others and have no intention to make profit. 1120 However, the second class of business models requires DRM, because 1121 they are primarily profit oriented. For example, large on-demand 1122 streaming platforms like Netflix establish business models based on 1123 subscriptions. Consumers may have to pay a monthly fee in order to 1124 get access to copyright protected content like TV series, movies or 1125 music. This model may be ad-supported and free to the content 1126 consumer, like YouTube Channels or Spotify. But the creator of the 1127 content expects some remuneration for his work. From the perspective 1128 of the service providers and the copyright owners, only clients that 1129 pay the fee (explicitly or implicitly through ad placement) should 1130 be able to access and consume the content. Anyway, the challenge is 1131 to find an efficient and scalable way of access control to digital 1132 content, which is distributed in information-centric networks. 1134 8.1. Broadcast Encryption for DRM in ICN 1136 The section discusses Broadcast Encryption (BE) as a suitable basis 1137 for DRM functionalities in conformance to the ICN communication 1138 paradigm. Especially when network inherent caching is considered the 1139 advantage of BE will be highlighted. 1141 In ICN, data packets can be cached inherently in the network and any 1142 network participant can request a copy of these packets. This makes 1143 it very difficult to implement an access control for content that is 1144 distributed via ICN. A naive approach is to encrypt the transmitted 1145 data for each consumer with a distinct key. This prohibits everyone 1146 other than the intended consumers to decrypt and consume the data. 1147 However, this approach is not suitable for ICN's communication 1148 paradigm since it would reduce the benefits gained from the inherent 1149 network caching. Even if multiple consumers request the same content 1150 the requested data for each consumer would differ using this 1151 approach. A better but still insufficient idea is to use a single 1152 key for all consumers. This does not destruct the benefits of ICN's 1153 caching ability. The drawback is that if one of the consumers 1154 illegally distributes the key, the system is broken and any entity 1155 in the network can access the data. Changing the key after such an 1156 event is useless since the provider has no possibility to identify 1157 the illegal distributer. Therefore this person cannot be stopped 1158 from distributing the new key again. In addition to this issue other 1159 challenges have to be considered. Subscriptions expire after a 1160 certain time and then it has to be ensured that these consumers 1161 cannot access the content anymore. For a provider that serves 1162 millions of daily consumers (e.g. Netflix) there could be a 1163 significant number of expiring subscriptions per day. Publishing a 1164 new key every time a subscription expires would require an 1165 unsuitable amount of computational power just to re-encrypt the 1166 collection of audio-visual content. 1168 A possible approach to solve these challenges is Broadcast 1169 Encryption (BE) [22] as proposed in [23]. From this point on, this 1170 section will focus only on BE as an enabler for DRM functionality in 1171 the use case of ICN video streaming. This subsection continues with 1172 the explanation of how BE works and shows how BE can be used to 1173 implement an access control scheme in the context of content 1174 distribution in ICN. 1176 BE actually carries a misleading name. One might expect a concrete 1177 encryption scheme. However, it belongs to the family of key- 1178 management schemes (KMS). KMS are responsible for the generation, 1179 exchange, storage and replacement of cryptographic keys. The most 1180 interesting characteristics of Broadcast Encryption Schemes (BES) 1181 are: 1183 . A BES typically uses a global trusted entity called the 1184 licensing agent (LA), which is responsible for spreading a set 1185 of pre-generated secrets among all participants. Each 1186 participant gets a distinct subset of secrets assigned from the 1187 LA. 1188 . The participants can agree on a common session key, which is 1189 chosen by the LA. The LA broadcasts an encrypted message that 1190 includes the key. Participants with a valid set of secrets can 1191 derive the session-key from this message. 1193 . The number of participants in the system can change 1194 dynamically. Entities may join or leave the communication group 1195 at any time. If a new entity joins the LA passes on a valid set 1196 of secrets to that entity. If an entity leaves (or is forced to 1197 leave) the LA revokes the entity's subset of keys, which means 1198 that it cannot derive the correct session key anymore when the 1199 LA distributes a new key. 1200 . Traitors (entities that reveal their secrets) can be traced and 1201 excluded from ongoing communication. The algorithms and 1202 preconditions to identify a traitor vary between concrete BES. 1204 This listing already illustrates why BE is suitable to control the 1205 access to data that is distributed via an information-centric 1206 network. BE enables the usage of a single session key for 1207 confidential data transmission between a dynamically changing subset 1208 or network participants. ICN caches can be utilized since the data 1209 is encrypted only with a single key known by all legitimate clients. 1210 Furthermore, traitors can be identified and removed from the system. 1211 The issue of re-encryption still exists, because the LA will 1212 eventually update the session key when a participant should be 1213 excluded. However, this disadvantage can be relaxed in some way if 1214 the following points are considered: 1216 . The updates of the session key can be delayed until a set of 1217 compromised secretes has been gathered. Note that secrets may 1218 become compromised because of two reasons. First, a traitor 1219 could have illegally revealed the secret. Second, the 1220 subscription of an entity expired. Delayed revocation 1221 temporarily enables some non-legitimate entities to consume 1222 content. However, this should not be a severe problem in home 1223 entertainment scenarios. Updating the session key in regular 1224 (not too short) intervals is a good tradeoff. The longer the 1225 interval last the less computational resources are required for 1226 content re-encryption and the better the cache utilization in 1227 the ICN will be. To evict old data from ICN caches that has 1228 been encrypted with the prior session key the publisher could 1229 indicate a lifetime for transmitted packets. 1230 . Content should be re-encrypted dynamically at request time. 1231 This has the benefit that untapped content is not re-encrypted 1232 if the content is not requested during two session key updates 1233 and therefore no resources are wasted. Furthermore, if the 1234 updates are triggered in non-peak times the maximum amount of 1235 resource needed at one point in time can be lowered 1236 effectively, since in peak times generally more diverse content 1237 is requested. 1238 . Since the amount of required computational resources may vary 1239 strongly from time to time it would be beneficial for any 1240 streaming provider to use cloud-based services to be able to 1241 dynamically adapt the required resources to the current needs. 1242 Regarding to a lack of computation time or bandwidth the cloud 1243 service could be used to scale up to overcome shortages. 1245 Figure 4 show the potential usage of BE in a multimedia delivery 1246 frameworks that builds upon ICN infrastructure and uses the concept 1247 of dynamic adaptive streaming, e.g., DASH. BE would be implemented 1248 on the top to have an efficient and scalable way of access control 1249 to the multimedia content. 1251 +--------Multimedia Delivery Framework--------+ 1252 | | 1253 | Technologies Properties | 1254 | +----------------+ +----------------+ | 1255 | | Broadcast |<--->| Controlled | | 1256 | | Encryption | | Access | | 1257 | +----------------+ +----------------+ | 1258 | |Dynamic Adaptive|<--->| Multimedia | | 1259 | | Streaming | | Adaptation | | 1260 | +----------------+ +----------------+ | 1261 | | ICN |<--->| Cachable | | 1262 | | Infrastructure | | Data Chunks | | 1263 | +----------------+ +----------------+ | 1264 +---------------------------------------------+ 1266 Figure 4: A potential multimedia framework using BE. 1268 8.2 . AAA Based DRM for ICN Networks 1270 8.2.1. Overview 1272 Recently, a novel approach to Digital Rights Management (DRM) has 1273 emerged to link DRM to usual network management operations, hence 1274 linking DRM to authentication, authorization, and accounting (AAA) 1275 services. ICN provides the abstraction of an architecture where 1276 content is requested by name and could be served from anywhere. In 1277 DRM, the content provider (the origin of the content) allows the 1278 destination (the end user account) to use the content. The content 1279 provider and content storage/cache are at two different entities in 1280 ICC and for traditional DRM only source and destination count and 1281 not the intermediate storage. The proposed solution allows the 1282 provider of the caching to be involved in the DRM policies using 1283 well known AAA mechanisms. It is important to note that this 1284 solution is compatible with the proposes the Broadcast Encryption 1285 (BE) proposed earlier in this draft. The BE proposes a technology as 1286 this solution is more operational. 1288 8.2.2. Implementation 1290 With the proposed AAA-based DRM, when content is requested by name 1291 from a specific destination, the request could link back to both the 1292 content provider and the caching provider via traditional AAA 1293 mechanisms, and trigger the appropriate DRM policy independently 1294 from where the content is stored. In this approach the caching, DRM 1295 and AAA remain independent entities but can work together through 1296 ICN mechanisms. The proposed solution enables extending the 1297 traditional DRM done by the content provider to jointly being done 1298 by content provider and network/caching provider. 1300 The solution is based on the concept of a "token". The content 1301 provider authenticates the end user and issues an encrypted token to 1302 authenticate the a named content ID or IDs that the user can access. 1303 The token will be shared with the network provider and used as the 1304 interface to the AAA protocols. At this point all content access is 1305 under the control of the network provider and the ICN. The 1306 controllers and switches can manage the content requests and handle 1307 mobility. The content can be accessed from anywhere as long as 1308 the token remains valid or the content is available in the network. 1309 In such a scheme the content provider does not need to be contacted 1310 every time a named content is requested. This reduces the load of 1311 the content provider network and creates a DRM mechanism that is 1312 much more appropriate for the distributed caching and peer-to-peer 1313 storage characteristic of ICN networks. In particular, the content 1314 requested by name can be served from anywhere under the only 1315 condition that the storage/cache can verify that the token is valid 1316 for content access. 1318 The solution is also fully customizable to both content and network 1319 provider's needs as the tokens can be issued based on user accounts, 1320 location and hardware (MAC address for example) linking it naturally 1321 to legacy authentication mechanisms. In addition, since both content 1322 and network providers are involved in DRM policies pollution attacks 1323 and other illegal requests for the content can be more easily 1324 detected. The proposed AAA-based DRM is currently under full 1325 development. 1327 9. Future Steps for Video in ICN 1329 The explosion of online video services, along with their increased 1330 consumption by mobile wireless terminals, further exacerbates the 1331 challenges of Video Adaptation leveraging ICN mechanisms. The 1332 following sections present a series of research items derived from 1333 these challenges, further introducing next steps for the subject. 1335 9.1. Large Scale Live Events 1337 An active area of investigation and a potential use case where ICN 1338 would provide significant benefits, is that of distributing content, 1339 and video in particular, using local communications in large scale 1340 events such as sports event in a stadium, a concert or a large 1341 demonstration. 1343 Such use-case involves locating content that is generated on the fly 1344 and requires discovery mechanisms in addition to sharing mechanisms. 1345 The scalability of the distribution becomes important as well. 1347 9.2. Video Conferencing and Real-Time Communications 1349 Current protocols for video-conferencing have been designed, and 1350 this document needs to take input from them to identify the key 1351 research issues. Real-time communications add timing constraints 1352 (both in terms of delay and in terms of synchronization) to the 1353 scenario discussed above. 1355 AR and VR (and immersive multimedia experiences in general) are 1356 clearly an area of further investigation, as they involve combining 1357 multiple streams of data from multiple users into a coherent whole. 1358 This raises issues of multi-source multi-destination multimedia 1359 streams that ICN may be equipped to deal with in a more natural 1360 manner than IP that is inherently unicast. 1362 9.3. Store-and-Forward Optimized Rate Adaptation 1364 One of the benefits of ICN is to allow the network to insert caching 1365 in the middle of the data transfer. This can be used to reduce the 1366 overall bandwidth demands over the network by caching content for 1367 future re-use. But it provides more opportunities for optimizing 1368 video streams. 1370 Consider for instance the following scenario: a client is connected 1371 via an ICN network to a server. Let's say the client is connected 1372 wirelessly to a node that has a caching capability, which is 1373 connected through a WAN to the server. Assume further that the 1374 capacity of each of the links (both the wireless and the WAN logical 1375 links) vary with time. 1377 If the rate adaptation is provided in an end-to-end manner, as in 1378 current mechanisms like DASH, then the maximal rate that can be 1379 supported at the client is that of the minimal bandwidth on each 1380 link. 1382 For instance, if during time period 1, the wireless capacity is 1 1383 and the wired capacity is 2, and during time period 2, the wireless 1384 is 2 due to some hotspot, and the wired is 1 due to some congestion 1385 in the network, then the best end-to-end rate that can be achieved 1386 is 1 during each period. 1388 However, if the cache is used during time period 1 to pre-fetch 2 1389 units of data, then during period 2, there is 1 unit of data at the 1390 cache, and another unit of data, which can be streamed from the 1391 server, and the rate that can be achieved is therefore 2 units of 1392 data. In this case, the average bandwidth rises from 1 to 1.5 over 1393 the 2 periods. 1395 This straw man example illustrate a) the benefit of ICN for 1396 increasing the throughput of the network, and b) the need for the 1397 special rate adaptation mechanisms to be designed so as to take 1398 advantage of this gain. End-to-end rate adaptation cannot take 1399 advantage of the cache availability. 1401 9.4. Heterogeneous Wireless Environment Dynamics 1403 With the ever-growing increase in online services being accessed by 1404 mobile devices, operators have been deploying different overlapping 1405 wireless access networking technologies. In this way, in the same 1406 area, user terminals are within range of different cellular, Wi-Fi 1407 or even WiMAX networks. Moreover, with the advent of the Internet of 1408 Things (e.g., surveillance cameras feeding video footage), this list 1409 can be further complemented with more specific short-range 1410 technologies, such as Bluetooth or ZigBee. 1412 In order to leverage from this plethora of connectivity 1413 opportunities, user terminals are coming equipped with different 1414 wireless access interfaces, providing them with extended 1415 connectivity opportunities. In this way, such devices become able to 1416 select the type of access which best suits them according to 1417 different criteria, such as available bandwidth, battery 1418 consumption, access do different link conditions according to the 1419 user profile or even access to different content. Ultimately, these 1420 aspects contribute to the Quality of Experience perceived by the 1421 end-user, which is of utmost importance when it comes to video 1422 content. 1424 However, the fact that these users are mobile and using wireless 1425 technologies, also provides a very dynamic setting, where the 1426 current optimal link conditions at a specific moment might not last 1427 or be maintained while the user moves. These aspects have been amply 1428 analyzed in recently finished projects such as FP7 MEDIEVAL [18], 1429 where link events reporting on wireless conditions and available 1430 alternative connection points were combined with vide requirements 1431 and traffic optimization mechanisms, towards the production of a 1432 joint network and mobile terminal mobility management decision. 1433 Concretely, in [19] link information about the deterioration of the 1434 wireless signal was sent towards a mobility management controller in 1435 the network. This input was combined with information about the user 1436 profile, as well as of the current video service requirements, and 1437 used to trigger the decrease or increase of scalable video layers, 1438 adjusting the video to the ongoing link conditions. Incrementally, 1439 the video could also be adjusted when a new better connectivity 1440 opportunity presents itself. 1442 In this way, regarding Video Adaptation, ICN mechanisms can leverage 1443 from their intrinsic multiple source support capability and go 1444 beyond the monitoring of the status of the current link, thus 1445 exploiting the availability of different connectivity possibilities 1446 (e.g., different "interfaces"). Moreover, information obtained from 1447 the mobile terminal's point of view of its network link, as well as 1448 information from the network itself (i.e., load, policies, and 1449 others), can generate scenarios where such information is combined 1450 in a joint optimization procedure allowing the content to be forward 1451 to users using the best available connectivity option (e.g., 1452 exploiting management capabilities supported by ICN intrinsic 1453 mechanisms as in [20]). 1455 In fact, ICN base mechanisms can further be exploited in enabling 1456 new deployment scenarios such as preparing the network for mass 1457 requests from users attending a large multimedia event (i.e., 1458 concert, sports), allowing video to be adapted according to content, 1459 user and network requirements and operation capabilities in a 1460 dynamic way. 1462 The enablement of such scenarios require further research, with the 1463 main points highlighted as follows: 1465 . Development of a generic video services (and obviously content) 1466 interface allowing the definition and mapping of their 1467 requirements (and characteristics) into the current capabilities 1468 of the network; 1470 . How to define a scalable mechanism allowing either the video 1471 application at the terminal, or some kind of network management 1472 entity, to adapt the video content in a dynamic way; 1474 . How to develop the previous research items using intrinsic ICN 1475 mechanisms (i.e., naming and strategy layers); 1477 . Leverage intelligent pre-caching of content to prevent stalls and 1478 poor quality phases, which lead to bad Quality of Experience of 1479 the user. This includes in particular the usage in mobile 1480 environments, which are characterized by severe bandwidth changes 1481 as well as connection outages, as shown in [21]; 1483 . How to take advantage of the multi-path opportunities over the 1484 heterogeneous wireless interfaces. 1486 9.5. Network Coding for Video Distribution in ICN 1488 An interesting research area for combining heterogeneous sources is 1489 to use network coding [24]. Network coding allows to asynchronously 1490 combine multiple sources by having each of them send information 1491 that is not duplicated by the other but can be combined to retrieve 1492 the video stream. 1494 However, this creates issues in ICN in terms of defining the proper 1495 rate adaptation for the video stream; securing the encoded data; 1496 caching the encoded data; timeliness of the encoded data; overhead 1497 of the network coding operations both in network resources and in 1498 added buffering delay, etc. 1500 Network coding has shown promise in reducing buffering events in 1501 unicast, multicast and P2P setting. [26] considers strategies using 1502 network coding to enhance QoE for multimedia communications. Network 1503 coding can be applied to multiple streams, but also within a single 1504 stream as an equivalent of a composable erasure code. Clearly, there 1505 is a need for further investigation of network coding in ICN, 1506 potentially as a topic of activity in the research group. 1508 9.6. Synchronization Issues for Video Distribution in ICN 1510 ICN de-couples the fetching of video chunks from the location of 1511 these chunks. This means an audio chunk may be received from one 1512 network element (cache/storage/server) while a video chunk may be 1513 received from another one while another chunk (say, the next one, or 1514 another layer from the same video stream) may come from a third 1515 element. This introduces disparity in the retrieval times and 1516 locations of the different elements of a video stream that need to 1517 be played at the same (or almost same) time. Synchronization of such 1518 delivery and playback may require specific synchronization tools for 1519 video delivery in ICN. 1521 Other synchronization aspects involve: 1523 - synchronizing within a single stream, for instance the consecutive 1524 chunks of a single stream, or the multiple layers of a layered 1525 scheme, when sources and transport layers may be different. Re- 1526 ordering the packets of a stream distributed over multiple sources 1527 at the video client, or ensuring that multiple chunks coming from 1528 multiple sources arrive within an acceptable time window; 1529 - synchronizing multiple streams, such as the audio and video 1530 components of a video stream, which can be received from 1531 independent sources; 1532 - synchronizing multiple streams from multiple sources to multiple 1533 destinations, such as mass distribution of live events. For 1534 instance, for live video streams or video-conferencing, some level 1535 of synchronization is required so that people watching the stream 1536 view the same events at the same time. 1538 Some of these issues were addressed in [27] in the context of social 1539 video consumption. Network coding, with traffic engineering, is 1540 considered as a potential solution for synchronization issues. Other 1541 approaches could be considered that are specific for ICN as well. 1543 Traffic engineering in ICN [28,29] may be required to provide proper 1544 synchronization of multiple streams. 1546 10. Security Considerations 1548 This is informational. There are no specific security considerations 1549 outside of those mentioned in the text. 1551 11. IANA Considerations 1553 This document does not require any IANA action. 1555 12. Conclusions 1557 This draft proposed adaptive video streaming for ICN, identified 1558 potential problems and presented the combination of CCN with DASH as 1559 a solution. As both concepts, DASH and CCN, maintain several 1560 elements in common, like, e.g., the content in different versions 1561 being dealt with in segments, combination of both technologies seems 1562 useful. Thus, adaptive streaming over CCN can leverage advantages 1563 such as, e.g., efficient caching and intrinsic multicast support of 1564 CCN, routing based on named data URIs, intrinsic multi-link and 1565 multi-source support, etc. 1567 In this context, the usage of CCN with DASH in mobile environments 1568 comes together with advantages compared to today's solutions, 1569 especially for devices equipped with multiple network interfaces. 1570 The retrieval of data over multiple links in parallel is a useful 1571 feature, specifically for adaptive multimedia streaming, since it 1572 offers the possibility to dynamically switch between the available 1573 links depending on their bandwidth capabilities, transparent to the 1574 actual DASH client. 1576 13. References 1578 13.1. Normative References 1580 [RFC6972] Y. Zhang, N. Zong, "Problem Statement and Requirements of 1581 the Peer-to-Peer Streaming Protocol (PPSP)", RFC6972, July 1582 2013 1584 13.2. Informative References 1586 [1] ISO/IEC DIS 23009-1.2, Information technology - Dynamic 1587 adaptive streaming over HTTP (DASH) - Part 1: Media 1588 presentation description and segment formats 1590 [2] Lederer, S., Mueller, C., Rainer, B., Timmerer, C., 1591 Hellwagner, H., "An Experimental Analysis of Dynamic Adaptive 1592 Streaming over HTTP in Content Centric Networks", in 1593 Proceedings of the IEEE International Conference on Multimedia 1594 and Expo 2013, San Jose, USA, July, 2013 1596 [3] Liu, Y., Geurts, J., Point, J., Lederer, S., Rainer, B., 1597 Mueller, C., Timmerer, C., Hellwagner, H., "Dynamic Adaptive 1598 Streaming over CCN: A Caching and Overhead Analysis", in 1599 Proceedings of the IEEE international Conference on 1600 Communication (ICC) 2013 - Next-Generation Networking 1601 Symposium, Budapest, Hungary, June, 2013 1603 [4] Grandl, R., Su, K., Westphal, C., "On the Interaction of 1604 Adaptive Video Streaming with Content-Centric Networks", 1605 eprint arXiv:1307.0794, July 2013. 1607 [5] V. Jacobson, D. Smetters, J. Thornton, M. Plass, N. Briggs and 1608 R. Braynard, "Networking named content", in Proc. of the 5th 1609 int. Conf. on Emerging Networking Experiments and Technologies 1610 (CoNEXT '09). ACM, New York, NY, USA, 2009, pp. 1-12. 1612 [6] A. Detti, M. Pomposini, N. Blefari-Melazzi, S. Salsano and A. 1613 Bragagnini, "Offloading cellular networks with Information- 1614 Centric Networking: The case of video streaming", In Proc. of 1615 the Int. Symp. on a World of Wireless, Mobile and Multimedia 1616 Networks (WoWMoM '12), IEEE, San Francisco, CA, USA, 1-3, 1617 2012. 1619 [7] V. Jacobson, D. K. Smetters, N. H. Briggs, M. F. Plass, P. 1620 Stewart, J. D. Thornton, and R. L. Braynard, "VoCCN: Voice 1621 over content-centric networks," in ACM ReArch Workshop, 2009 1623 [8] Christopher Mueller, Stefan Lederer and Christian Timmerer, A 1624 proxy effect analysis and fair adaptation algorithm for 1625 multiple competing dynamic adaptive streaming over HTTP 1626 clients, In Proceedings of the Conference on Visual 1627 Communications and Image Processing (VCIP) 2012, San Diego, 1628 USA, November 27-30, 2012. 1630 [9] DASH Research at the Institute of Information Technology, 1631 Multimedia Communication Group, Alpen-Adria Universitaet 1632 Klagenfurt, URL: http://dash.itec.aau.at 1634 [10] A. Detti, N. Blefari-Melazzi, S. Salsano, and M. Pomposini 1635 CONET: A content centric inter-networking architecture," in ACM 1636 Workshop on Information-Centric Networking (ICN), 2011. 1638 [11] W. K. Chai, N. Wang, I. Psaras, G. Pavlou, C. Wang, G. C. de 1639 Blas, F. Ramon-Salguero, L. Liang, S. Spirou, A. Beben, and E. 1640 Hadjioannou, "CURLING: Content-ubiquitous resolution and 1641 delivery infrastructure for next-generation services," IEEE 1642 Communications Magazine, vol. 49, no. 3, pp. 112-120, March 1643 2011 1645 [12] NetInf project Website http://www.netinf.org 1647 [13] N. Magharei, R. Rejaie, Yang Guo, "Mesh or Multiple-Tree: A 1648 Comparative Study of Live P2P Streaming Approaches," INFOCOM 1649 2007. 26th IEEE International Conference on Computer 1650 Communications. IEEE , vol., no., pp.1424,1432, 6-12 May 2007 1652 [14] PPSP WG Website https://datatracker.ietf.org/wg/ppsp/ 1654 [15] A. Bakker, R. Petrocco, V. Grishchenko, "Peer-to-Peer Streaming 1655 Peer Protocol (PPSPP)", draft-ietf-ppsp-peer-protocol-08 1657 [16] Rui S. Cruz, Mario S. Nunes, Yingjie Gu, Jinwei Xia, Joao P. 1658 Taveira, Deng Lingli, "PPSP Tracker Protocol-Base Protocol 1659 (PPSP-TP/1.0)", draft-ietf-ppsp-base-tracker-protocol-02 1661 [17] A.Detti, B. Ricci, N. Blefari-Melazzi,"Peer-To-Peer Live 1662 Adaptive Video Streaming for Information Centric Cellular 1663 Networks", IEEE PIMRC 2013,London, UK, 8-11 September 2013 1665 [18] http://www.ict-medieval.eu 1667 [19] B. Fu, G. Kunzmann, M. Wetterwald, D. Corujo, R. Costa, "QoE- 1668 aware Traffic Management for Mobile Video Delivery", Proc. 2013 1669 IEEE ICC, Workshop on Immersive & Interactive Multimedia 1670 Communications over the Future Internet (IIMC), Budapest, 1671 Hungary, Jun 2013. 1673 [20] Corujo D., Vidal I., Garcia-Reinoso J., Aguiar R., "A Named 1674 Data Networking Flexible Framework for Management 1675 Communications", IEEE Communications Magazine, Vol. 50, no. 12, 1676 pp. 36-43, Dec 2012 1678 [21] Crabtree B., Stevens T., Allan B., Lederer S., Posch D., 1679 Mueller C., Timmerer C., Video Adaptation in Limited or Zero 1680 Network Coverage, CCNxConn 2013,PARC, Palo Alto, pp. 1-2, 2013 1682 [22] Fiat A., Naor M., "Broadcast Encryption", in Advances in 1683 Cryptology (Crypto'93), volume 773 of Lecture Notes in Computer 1684 Science, pages 480-491. Springer Berlin / Heidelberg, 1994. 1686 [23] Posch D., Hellwagner H., Schartner P., "On-Demand Video 1687 Streaming based on Dynamic Adaptive Encrypted Content Chunks", th in Proceedings of the 8 International Workshop on Secure 1688 Network Protocols (NPSec'13), Los Alamitos, IEEE Computer 1689 Society Press, October, 2013. 1691 [24] Montpetit M.J., Westphal C., Trossen D., "Network Coding Meets 1692 Information Centric Networks," in Proceedings of the workshop 1693 on Name-Oriented Mobility (NOM), jointly with ACM MobiHoc 2013, 1694 Hilton Head, SC, June 2013. 1696 [25] Le Callet P., Moeller S. and Perkis A. (eds.), Qualinet 1697 White Paper on Definitions of Quality of Experience 1698 (2012). European Network on Quality of Experience in Multimedia 1699 Systems and Services (COST Action IC 1003), Lausanne, 1700 Switzerland, Version 1.2, March 2013. 1702 [26] Medard M., Kim M., ParandehGheibi A., Zeng W., Montpetit M.J., 1703 Quality of Experience for Multimedia Communications: Network 1704 Coding Strategies, Technical Report, MIT, 2012/3 1706 [27] Montpetit M.J., Holtzman H., Chakrabarti K., Matijasevic M., 1707 Social video consumption: Synchronized viewing experiences 1708 across devices and networks, IEEE International Conference on 1709 Communications Workshops (ICC), 2013, 286-290 1711 [28] Su K., Westphal C., On the Benefit of Information Centric 1712 Networks for Traffic Engineering, IEEE ICC Conference, June 1713 2014 1715 [29] Chanda A., Westphal C., Raychaudhuri D., Content Based Traffic 1716 Engineering in Software Defined Information Centric Networks, 1717 in IEEE INFOCOM Workshop NOMEN'13, April, 2013 1719 [30] Castro M., Druschel P., Kermarrec A.-M., Nandi A., Rowstron A., 1720 Singh A. SplitStream: Highbandwidth multicast in cooperative 1721 environments. In Proceedings of ACM SOSP (2003). 1723 [31] ATIS IPTV Interoperability Forum, ATIS IFF, 1724 http://www.atis.org/iif/deliv.asp 1726 14. Authors' Addresses 1728 Stefan Lederer, Christian Timmerer, Daniel Posch 1729 Alpen-Adria University Klagenfurt 1730 Universitaetsstrasse 65-67, 9020 Klagenfurt, Austria 1732 Email: {firstname.lastname}@itec.aau.at 1734 Cedric Westphal, Aytac Azgin. Shucheng (Will) Liu 1735 Huawei 1736 2330 Central Expressway, Santa Clara, CA95050, USA 1738 Email: {cedric.westphal,aytac.azgin,liushucheng}@huawei.com 1739 Christopher Mueller 1740 bitmovin GmbH 1741 Lakeside B01, 9020 Klagenfurt, Austria 1743 Email: christopher.mueller@bitmovin.net 1745 Andrea Detti 1746 Electronic Engineering Dept. 1747 University of Rome Tor Vergata 1748 Via del Politecnico 1, Rome, Italy 1750 Email: andrea.detti@uniroma2.it 1752 Daniel Corujo, 1753 Advanced Telecommunications and Networks Group 1754 Instituto de Telecomunicacoes 1755 Campus Universitario de Santiago 1756 P-3810-193 Aveiro, Portugal 1758 Email: dcorujo@av.it.pt 1760 Jianping Wang 1761 City University of Hong Kong 1762 Hong Kong, China 1764 Email: jianwang@cityu.edu.hk 1766 Marie-Jose Montpetit 1768 Email: marie@mjmontpetit.com 1770 Niall Murray 1771 Dept. of Electronic, Computer and Software Engineering 1772 Athlone Institute of Technology 1773 Dublin Rd., Athlone, Ireland 1775 Email: nmurray@research.ait.ie 1777 15. Acknowledgements 1779 This work was supported in part by the EC in the context of the 1780 SocialSensor (FP7-ICT-287975) project and partly performed in the 1781 Lakeside Labs research cluster at AAU. SocialSensor receives 1782 research funding from the European Community's Seventh Framework 1783 Programme. The work for this document was also partially performed 1784 in the context of the FP7/NICT EU-JAPAN GreenICN project, 1785 http://www.greenicn.org. Apart from this, the European Commission 1786 has no responsibility for the content of this draft. The information 1787 in this document is provided as is and no guarantee or warranty is 1788 given that the information is fit for any particular purpose. The 1789 user thereof uses the information at its sole risk and liability. 1790 The authors would like to thank Shucheng (Will) Liu from Huawei for 1791 supporting Dr. Wang's research and Marie-Jose Montpetit of MIT for 1792 her help in writing the AAA for DRM section.