蜂窝网络下实时互动视频流性能诊断与优化

蜂窝网络作为覆盖范围最广、用户最多的无线网络，是互动视频直播应用最重要的接入网络之一。然而，由于蜂窝网络传输性能（带宽、时延等）的剧烈波动以及移动设备性能的限制，导致实时互动视频应用用户在蜂窝网络下的体验受到严重影响。本文旨在针对互动视频直播的传输需求，诊断蜂窝网络对视频直播业务的影响原因，并有针对性地设计优化机制，以提高用户的体验质量。

首先，在真实蜂窝网络下，对实时互动视频流框架WebRTC的传输性能进行诊断，发现两个问题导致用户体验下降。1) WebRTC的拥塞控制算法无法有效适应蜂窝网络的高度动态变化; 2) WebRTC的编码设置无法灵活适应网络状况和传输性能的改变。针对上述诊断发现，设计并实现了两种优化方案。

屏蔽蜂窝网络传输抖动的实时互动视频流拥塞控制机制Mustang。通过实际网络测量及对直播框架WebRTC实现进行分析，发现帧交付时间直接影响到播放器的帧率表现，从而影响视频播放的流畅性。此外，在蜂窝网络下接收速率比RTT更能反映网络拥塞情况。基于这些发现提出了Mustang，其利用帧交付时间和接收速率来判断当前网络抖动是否可以屏蔽，并据此调节拥塞控制算法的带宽估计决策。我们在WebRTC上实现了Mustang，通过实际蜂窝网络上大量实验对比，验证了其性能优势。相比于传统拥塞控制算法如GCC(Google Congestion Control), PCC(Performance-oriented Congestion Control)等，Mustang的视频比特率提升了72.1%，MOS(Mean Opinion Score)提高30%.

视频直播业务中的自适应调整编码配置机制Hughes，通过灵活调整编码配置以提高蜂窝网络下视频直播用户的观看体验。诊断发现对于高动态视频，用户更注重视频的帧率，反之，对于低动态视频，用户更注重视频的分辨率。因此，Hughes首先利用视频帧的低级图像特征判别用户对视频观看的偏好，然后将用户偏好提供给一个基于强化学习的编码决策模块，决策出最终的编码参数。我们在WebRTC上实现了Hugehes，通过实验评估验证了Hugehes在视频比特率方面较WebRTC提升17.6%,在PSNR方面较WebRTC提升了0.74dB。

[1] CISCO V. Cisco visual networking index: Forecast and methodology, 2018–2022[EB/OL].2018. https://www.cisco.com/c/en/us/solutions/%20collateral/executive-perspectives/annual-internet-report/white-paper-c11%20741490.html.
[2] CISCO. VNI Complete Forecast Highlights[EB/OL]. 2018. https://www.cisco.com/c/dam/m/en_us/solutions/service-provider/vni-forecast-highlights/pdf/Global_Business_Highlights.pdf.
[3] ALVESTRAND H. RFC 8825-Overview: Real-Time Protocols for Browser-Based Applications[J]. IETF, 2021.
[4] SCHULZRINNE H, CASNER S, FREDERICK R, et al. RFC3550: RTP: A transport protocolfor real-time applications[M]. RFC Editor, 2003.
[5] SARKER Z, PERKINS C, SINGH V, et al. RFC 8888: RTP Control Protocol (RTCP) Feedback for Congestion Control[M]. RFC Editor, 2021.
[6] HOLMER S, LUNDIN H, CARLUCCI G, et al. A Google Congestion Control Algorithm for Real-Time Communication: draft-ietf-rmcat-gcc-02[R/OL]. Internet Engineering Task Force,2016. https://datatracker.ietf.org/doc/draft-ietf-rmcat-gcc/02/.
[7] RECOMMENDATION I. E. 800: Terms and definitions related to quality of service and network performance including dependability: Vol. 1994[EB/OL]. 1994.
[8] ZADTOOTAGHAJ S, SCHMIDT S, SABET S S, et al. Quality estimation models for gaming video streaming services using perceptual video quality dimensions[C]//Proceedings of the 11th ACM Multimedia Systems Conference. 2020: 213-224.
[9] NATHAN V, SIVARAMAN V, ADDANKI R, et al. End-to-end transport for video QoE fairness [M]//Proceedings of the ACM Special Interest Group on Data Communication. 2019: 408-423.
[10] DUANMU Z, LIU W, LI Z, et al. The Waterloo Streaming Quality-of-Experience Database-IV[EB/OL]. IEEE Dataport, 2020. https://dx.doi.org/10.21227/j15a-8r35.
[11] WebRTC. WebRTC Native Code[EB/OL]. https://webrtc.github.io/webrtc-org/native-code/.
[12] CARLUCCI G, DE CICCO L, HOLMER S, et al. Congestion control for web real-time communication[J]. IEEE/ACM Transactions on Networking, 2017, 25(5): 2629-2642.
[13] WEI X, DING P, ZHOU L, et al. QoE oriented chunk scheduling in P2P-VoD streaming system[J]. IEEE Transactions on Vehicular Technology, 2019, 68(8): 8012-8025.
[14] ZHANG B, TEIXEIRA T, REZNIK Y. Performance of low-latency HTTP-based streaming players[C]//Proceedings of the 12th ACM Multimedia Systems Conference. 2021: 356-362.
[15] ZHANG Y, WU W, BANERJEE S, et al. SLA-verifier: Stateful and quantitative verification for service chaining[C]//IEEE INFOCOM 2017-IEEE Conference on Computer Communications. IEEE, 2017: 1-9.
[16] RECOMMENDATION I. Vocabulary for Performance, Quality of Service and Quality of Experience[J]. International Telecommunication Union, 2017: 10.
[17] BALACHANDRAN A, SEKAR V, AKELLA A, et al. Developing a predictive model of quality of experience for internet video[J]. ACM SIGCOMM Computer Communication Review, 2013,43(4): 339-350.
[18] KELEş O, YıLMAZ M A, TEKALP A M, et al. On the Computation of PSNR for a Set of Images or Video[C]//2021 Picture Coding Symposium (PCS). IEEE, 2021: 1-5.
[19] CHEN S, ZHANG Y, LI Y, et al. Spherical structural similarity index for objective omnidirectional video quality assessment[C]//2018 IEEE international conference on multimedia and expo (ICME). IEEE, 2018: 1-6.
[20] HU S, XU M, ZHANG H, et al. Affective content-aware adaptation scheme on QoE optimization of adaptive streaming over HTTP[J]. ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM), 2019, 15(3s): 1-18.
[21] CARDWELL N, CHENG Y, GUNN C S, et al. BBR: Congestion-based congestion control:Measuring bottleneck bandwidth and round-trip propagation time[J]. Queue, 2016, 14(5): 20-53.
[22] HA S, RHEE I, XU L. CUBIC: a new TCP-friendly high-speed TCP variant[J]. ACM SIGOPS operating systems review, 2008, 42(5): 64-74.
[23] DONG M, LI Q, ZARCHY D, et al. {PCC}: Re-architecting congestion control for consistent high performance[C]//12th USENIX Symposium on Networked Systems Design and Implementation (NSDI 15). 2015: 395-408.
[24] DONG M, MENG T, ZARCHY D, et al. {PCC} Vivace:{Online-Learning} Congestion Control[C]//15th USENIX Symposium on Networked Systems Design and Implementation (NSDI 18).2018: 343-356.
[25] WU L, ZHOU A, CHEN X, et al. Gcc-beta: Improving interactive live video streaming via an adaptive low-latency congestion control[C]//ICC 2019-2019 IEEE International Conference on Communications (ICC). IEEE, 2019: 1-6.
[26] ZHAO Y, ZHOU A, CHEN X. Reducing Latency in Interactive Live Video Chat Using Dynamic Reduction Factor[C]//2020 IEEE Wireless Communications and Networking Conference (WCNC). IEEE, 2020: 1-6.
[27] ZHOU A, ZHANG H, SU G, et al. Learning to coordinate video codec with transport protocol for mobile video telephony[C]//The 25th Annual International Conference on Mobile Computing and Networking. 2019: 1-16.
[28] ZHANG H, ZHOU A, LU J, et al. OnRL: improving mobile video telephony via online reinforcement learning[C]//Proceedings of the 26th Annual International Conference on Mobile Computing and Networking. 2020: 1-14.
[29] ZHANG H, ZHOU A, HU Y, et al. Loki: improving long tail performance of learning-based real-time video adaptation by fusing rule-based models[C]//Proceedings of the 27th Annual International Conference on Mobile Computing and Networking. 2021: 775-788.
[30] ZHANG S, LEI W, ZHANG W, et al. An online learning based path selection for multipath realtime video transmission in overlay network[J]. Transactions on Emerging Telecommunications Technologies, 2020, 31(11): e4131.
[31] LEE I, KIM S, SATHYANARAYANA S, et al. R-FEC: RL-based FEC Adjustment for Better QoE in WebRTC[C]//Proceedings of the 30th ACM International Conference on Multimedia. 2022: 2948-2956.
[32] CHEN K, WANG H, FANG S, et al. RL-AFEC: adaptive forward error correction for realtime video communication based on reinforcement learning[C]//Proceedings of the 13th ACM Multimedia Systems Conference. 2022: 96-108.
[33] YEO H, JUNG Y, KIM J, et al. Neural adaptive content-aware internet video delivery[C]//13th USENIX Symposium on Operating Systems Design and Implementation (OSDI 18). 2018:645-661.
[34] YEO H, CHONG C J, JUNG Y, et al. Nemo: enabling neural-enhanced video streaming on commodity mobile devices[C]//Proceedings of the 26th Annual International Conference on Mobile Computing and Networking. 2020: 1-14.
[35] FOULADI S, EMMONS J, ORBAY E, et al. Salsify:{Low-Latency} Network Video through Tighter Integration between a Video Codec and a Transport Protocol[C]//15th USENIX Symposium on Networked Systems Design and Implementation (NSDI 18). 2018: 267-282.
[36] RUDOW M, YAN F Y, KUMAR A, et al. Tambur: Efficient loss recovery for videoconferencing via streaming codes[C]//20th USENIX Symposium on Networked Systems Design and Implementation (NSDI 23). 2023: 953-971.
[37] HU P, MISRA R, KATTI S. Dejavu: Enhancing videoconferencing with prior knowledge[C]//Proceedings of the 20th International Workshop on Mobile Computing Systems and Applications. 2019: 63-68.
[38] KIM J, JUNG Y, YEO H, et al. Neural-enhanced live streaming: Improving live video ingest via online learning[C]//Proceedings of the Annual conference of the ACM Special Interest Group on Data Communication on the applications, technologies, architectures, and protocols for computer communication. 2020: 107-125.
[39] LUO Z, WANG Z, CHEN J, et al. Crowdsr: enabling high-quality video ingest in crowdsourced livecast via super-resolution[C]//Proceedings of the 31st ACM Workshop on Network and Operating Systems Support for Digital Audio and Video. 2021: 90-97.
[40] SARKAR A, MURRAY J, DASARI M, et al. L3BOU: Low Latency, Low Bandwidth, Optimized Super-Resolution Backhaul for 360-Degree Video Streaming[C]//2021 IEEE International Symposium on Multimedia (ISM). IEEE, 2021: 138-147.
[41] MENG Z, WANG T, SHEN Y, et al. Enabling High Quality {Real-Time} Communications with Adaptive {Frame-Rate}[C]//20th USENIX Symposium on Networked Systems Design and Implementation (NSDI 23). 2023: 1429-1450.
[42] WANG F, LI Q, SHI W, et al. Reparo: QoE-Aware Live Video Streaming in Low-Rate Networks by Intelligent Frame Recovery[C]//Proceedings of the 31st ACM International Conference on Multimedia. 2023: 9194-9204.
[43] ZHU X, SEN S, MAO Z M. Livelyzer: analyzing the first-Mile ingest performance of live video streaming[C]//Proceedings of the 12th ACM Multimedia Systems Conference. 2021: 36-50.
[44] RAY D, KOSAIAN J, RASHMI K, et al. Vantage: optimizing video upload for time-shifted viewing of social live streams[M]//Proceedings of the ACM Special Interest Group on Data Communication. 2019: 380-393.
[45] TAN Z, ZHAO J, LI Y, et al. {Device-Based}{LTE} Latency Reduction at the Application Layer [C]//18th USENIX Symposium on Networked Systems Design and Implementation (NSDI 21).2021: 471-486.
[46] LEE J, LEE S, LEE J, et al. PERCEIVE: deep learning-based cellular uplink prediction using real-time scheduling patterns[C]//Proceedings of the 18th International Conference on Mobile Systems, Applications, and Services. 2020: 377-390.
[47] LIN Y, GAO Y, DONG W. Bandwidth Prediction for 5G Cellular Networks[C]//2022IEEE/ACM 30th International Symposium on Quality of Service (IWQoS). IEEE, 2022: 1-10.
[48] JIANG J, SEKAR V, ZHANG H. Improving fairness, efficiency, and stability in http-based adaptive video streaming with festive[C]//Proceedings of the 8th international conference on Emerging networking experiments and technologies. 2012: 97-108.
[49] HUANG T Y, JOHARI R, MCKEOWN N, et al. A buffer-based approach to rate adaptation:Evidence from a large video streaming service[C]//Proceedings of the 2014 ACM conference on SIGCOMM. 2014: 187-198.
[50] MAO H, NETRAVALI R, ALIZADEH M. Neural adaptive video streaming with pensieve[C]//Proceedings of the conference of the ACM special interest group on data communication. 2017:197-210.
[51] YAN F Y, AYERS H, ZHU C, et al. Learning in situ: a randomized experiment in video streaming[C]//17th USENIX Symposium on Networked Systems Design and Implementation (NSDI20). 2020: 495-511.
[52] ZUO X, YANG J, WANG M, et al. Adaptive bitrate with user-level QOE preference for video streaming[C]//IEEE INFOCOM 2022-IEEE Conference on Computer Communications. IEEE,2022: 1279-1288.
[53] ZHENG Z, MA Y, LIU Y, et al. Xlink: Qoe-driven multi-path quic transport in large-scale video services[C]//Proceedings of the 2021 ACM SIGCOMM 2021 Conference. 2021: 418-432.
[54] MENG Z, GUO Y, SUN C, et al. Achieving consistent low latency for wireless real-time communications with the shortest control loop[C]//Proceedings of the ACM SIGCOMM 2022 Conference. 2022: 193-206.
[55] DHAWASKAR SATHYANARAYANA S, LEE K, GRUNWALD D, et al. Converge: QoEdriven Multipath Video Conferencing over WebRTC[C]//Proceedings of the ACM SIGCOMM 2023 Conference. 2023: 637-653.
[56] WU B, QIAN K, LI B, et al. XRON: A Hybrid Elastic Cloud Overlay Network for Video Conferencing at Planetary Scale[C]//Proceedings of the ACM SIGCOMM 2023 Conference.2023: 696-709.
[57] BRADNER S. Benchmarking terminology for network interconnection devices[R]. 1991.
[58] AGARWAL S, KRISHNAMURTHY A, AGARWAL R. Host Congestion Control[C]//Proceedings of the ACM SIGCOMM 2023 Conference. 2023: 275-287.
[59] NARAYANAN A, RAMADAN E, CARPENTER J, et al. A first look at commercial 5G performance on smartphones[C]//Proceedings of The Web Conference 2020. 2020: 894-905.
[60] NARAYANAN A, RAMADAN E, CARPENTER J, et al. A First Look at Commercial 5G Performance on Smartphones[EB/OL]. 2020. https://fivegophers.umn.edu/www20/.
[61] MUSTANGCCA. MustangCCA[EB/OL]. 2023. https://github.com/MustangCCA/MustangCCA.
[62] SACHA G, ANDY G, ENRICO V, et al. Big Buck Bunny[EB/OL]. https://peach.blender.org/media-gallery/.
[63] BANKOSKI J, KOLESZAR J, QUILLIO L, et al. VP8 data format and decoding guide[R].2011.
[64] VAN DER HOOFT J, PETRANGELI S, WAUTERS T, et al. HTTP/2-based adaptive streaming of HEVC video over 4G/LTE networks[J]. IEEE Communications Letters, 2016, 20(11): 2177-2180.
[65] VAN DER HOOFT J, PETRANGELI S, WAUTERS T, et al. 4G/LTE Bandwidth Logs[EB/OL].2016. https://users.ugent.be/~jvdrhoof/dataset-4g/.
[66] ZHU S S, DONG Y N, XU C. A statistical QoE-QoS model of video streaming services[C]//Proceedings of 2020 the 6th International Conference on Computing and Data Engineering.2020: 195-199.
[67] MIRANDA G, MACEDO D F, MARQUEZ-BARJA J M. Estimating video on demand QoE from network QoS through ICMP probes[J]. IEEE Transactions on Network and Service Management, 2021, 19(2): 1890-1902.
[68] ZHANG X, OU Y, SEN S, et al. {SENSEI}: Aligning Video Streaming Quality with Dynamic User Sensitivity[C]//18th USENIX Symposium on Networked Systems Design and Implementation (NSDI 21). 2021: 303-320.
[69] HUANG S, XIE J. DAVE: Dynamic adaptive video encoding for real-time video streaming applications[C]//2021 18th Annual IEEE International Conference on Sensing, Communication,and Networking (SECON). IEEE, 2021: 1-9.
[70] FFmpeg. FFmpeg[EB/OL]. https://ffmpeg.org/.
[71] LI Z, AARON A, KATSAVOUNIDIS I, et al. Toward a practical perceptual video quality metric[EB/OL]. 2016.
[72] RASSOOL R. VMAF reproducibility: Validating a perceptual practical video quality metric[C]//2017 IEEE international symposium on broadband multimedia systems and broadcasting(BMSB). IEEE, 2017: 1-2.
[73] KEYS R. Cubic convolution interpolation for digital image processing[J]. IEEE transactionson acoustics, speech, and signal processing, 1981, 29(6): 1153-1160.
[74] LI Y, PADMANABHAN A, ZHAO P, et al. Reducto: On-camera filtering for resource-efficient real-time video analytics[C]//Proceedings of the Annual conference of the ACM Special Interest Group on Data Communication on the applications, technologies, architectures, and protocols for computer communication. 2020: 359-376.
[75] SCHULMAN J, WOLSKI F, DHARIWAL P, et al. Proximal policy optimization algorithms[A]. 2017.
[76] PAN Z, ZHOU J, QIU X, et al. Marten: A built-in security drl-based congestion control framework by polishing the expert[C]//IEEE INFOCOM 2023-IEEE Conference on Computer Communications. IEEE, 2023: 1-10.