CS2P: Improving Video Bitrate Selection and Adaptation with Data-Driven Throughput Prediction
Open accessThe DOI matches its Crossref record; 38 of the 38 references it cites were found in a registry or an index.
Abstract
Bitrate adaptation is critical in ensuring good users’ quality-of-experience (QoE) in Internet video delivery system. Several efforts have argued that accurate throughput prediction can dramatically improve (1) initial bitrate selection for low startup delay and high initial resolution; (2) midstream bitrate adaptation for high QoE. However, prior ef- forts did not systematically quantify real-world throughput predictability or develop good prediction algorithms. To bridge this gap, this paper makes three key technical contributions: First, we analyze the throughput characteristics in a dataset with 20M+ sessions. We find: (a) Sessions sharing similar key features (e.g., ISP, region) present similar initial values and dynamical patterns; (b) There is a natural “stateful” dynamical behavior within a given session. Second, building on these insights, we develop CS2P, a better throughput prediction system. CS2P leverages data-driven approach to learn (a) clusters of similar sessions, (b) an initial throughput predictor, and (c) a Hidden-Markov-Model based midstream predictor modeling the stateful evolution of throughput. Third, we develop a prototype system and show by trace-driven simulation and real-world experiments that CS2P outperforms state-of-art by 40% and 50% median pre- diction error respectively for initial and midstream through- put and improves QoE by 14% over buffer-based adaptation algorithm.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“And deep learning further empowers more effective local analysis and prediction [130] at the edge node of industry instead of the cloud.”
From Deep learning for edge computing applications: A State-of-the-Art survey (2020) · cites this paper as [130]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“It has been shown in recent work [432] that accurate throughput prediction can significantly improve the QoE for adaptive video streaming.”
From A comprehensive survey on machine learning for networking: Evolution, applications and research opportunities (2018) · cites this paper as [432]
The cited source says what the claim attributes to it.
In the citing paper
“It is also shown that CS2P achieves 3.2% improvement on overall QoE and 10.9% higher average bitrate over state-of-art Model Predictive Control (MPC) approach, which uses harmonic mean for throughput prediction.”
From A comprehensive survey on machine learning for networking: Evolution, applications and research opportunities (2018) · cites this paper as [432]
In the citing paper
“Data-driven approaches, in particular ML, have been employed mainly in two different ways: (1) to predict changes in network QoS, namely throughput, and trigger adaptation mechanism to reduce rebuffering time [432], and (2) to select appropriate adaptation action [102].”
From A comprehensive survey on machine learning for networking: Evolution, applications and research opportunities (2018) · cites this paper as [432]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Book Review: Pattern Recognition and Machine Learning
Nasrabadi, N. M. (2007) · Journal of Electronic Imaging · doi:10.1117/1.2819119 · Cited by 22,066
Pattern Recognition and Machine Learning
(2006) · doi:10.1007/978-0-387-45528-0 · Cited by 9,824
Scikit-learn: Machine Learning in Python
(2011) · Journal of Machine Learning Research · doi:10.5555/1953048.2078195 · Cited by 8,539
(2006) · Springer eBooks · Cited by 8,123
Modeling TCP throughput: a simple model and its empirical validation
Padhye, J., Firoiu, V., Towsley, D., & Kurose, J. (1998) · Proceedings of the ACM SIGCOMM '98 conference on Applications, technologies, architectures, and protocols for computer communication · doi:10.1145/285237.285291 · Cited by 2,106 · cited 1×
“Throughput prediction: Prior work either developed approximate analytical models of TCP throughput as a function of packet loss and delay [24, 38] or leveraged Time-series models (e.”
MethodVivaldi: a decentralized network coordinate system
Dabek, F., Cox, R., Kaashoek, F., & Morris, R. (2004) · Proceedings of the 2004 conference on Applications, technologies, architectures, and protocols for computer communications · doi:10.1145/1015467.1015471 · Cited by 1,154 · cited 1×
“Seen in a broader context, CS2P can be regarded as a middle ground between centralized video control platforms (e.g., C3 [23], CFA [29]) and decentralized player-based bi-trate adaptation (e.g., BB [27], FESTIVE [30], MPC [47]).”
BackgroundA Control-Theoretic Approach for Dynamic Adaptive Video Streaming over HTTP
Yin, X., Jindal, A., Sekar, V., & Sinopoli, B. (2015) · Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication · doi:10.1145/2785956.2787486 · Cited by 1,016 · cited 6×
“…adaptation: While it is possible to develop adaptation approaches without using through-put estimation (e.g., using only the playback buffer occupancy [27]), recent work [47] argues that throughput-aware bitrate adaptation can deliver a better QoE than pure buffer-occupancy based approaches.”
BackgroundMethodInfluentialSelf-citationA buffer-based approach to rate adaptation: evidence from a large video streaming service
Huang, T.-Y., Johari, R., McKeown, N., Trunnell, M., & Watson, M. (2014) · Proceedings of the 2014 ACM conference on SIGCOMM · doi:10.1145/2619239.2626296 · Cited by 1,004
Generalized Boosted Models: A guide to the gbm package
Ridgeway, G. K. (2006) · Cited by 814
Understanding the impact of video quality on user engagement
Dobrian, F., Sekar, V., Awan, A., Stoica, I., Joseph, D., Ganjam, A., Zhan, J., & Zhang, H. (2011) · Proceedings of the ACM SIGCOMM 2011 conference · doi:10.1145/2018436.2018478 · Cited by 673 · cited 2×
“Delivering good application-level video quality-of-experience (QoE) entails new metrics such as low buffering or smooth bitrate delivery [5, 22].”
BackgroundMethodSelf-citation
Research works printed but not linked: 3
- J. Jiang, V. Sekar, H. Milner, D. Shepherd, I. Stoica, and H. Zhang. CFA: A Practical Prediction System for Video QoE Optimization. In Proc. USENIX NSDI, 2016.
- H. V. Madhyastha, T. Isdal, M. Piatek, C. Dixon, T. Anderson, A. Krishnamurthy, and A. Venkataramani. iPlane: An Information Plane for Distributed Services. In Proc. USENIX OSDI, 2006.
- K. Murphy and M. Dunham. PMTK: Probabilistic Modeling Toolkit. In Proc. NIPS, 2008.
Cited by
Neural Adaptive Video Streaming with Pensieve
Mao, H., Netravali, R., & Alizadeh, M. (2017) · Proceedings of the Conference of the ACM Special Interest Group on Data Communication · doi:10.1145/3098822.3098843 · Cited by 1,357
“However, selecting the right bitrate can be very challenging due to (1) the variability of network throughput [18, 41, 48, 51, 52]; (2) the conflicting video QoE requirements (high bitrate, minimal rebuffering, smoothness, etc.”
BackgroundInfluentialResource Management with Deep Reinforcement Learning
Mao, H., Alizadeh, M., Menache, I., & Kandula, S. (2016) · Proceedings of the 15th ACM Workshop on Hot Topics in Networks · doi:10.1145/3005745.3005750 · Cited by 1,353
“[39], and operate well for different codecs, screen sizes and available bandwidths (e.g., DSL vs. T1).”
BackgroundBoutaba, R., Salahuddin, M. A., Limam, N., Ayoubi, S., Shahriar, N., Estrada-Solano, F., & Caicedo, O. M. (2018) · Journal of Internet Services and Applications · doi:10.1186/s13174-018-0087-2 · Cited by 1,036
Edge Assisted Real-time Object Detection for Mobile Augmented Reality
Liu, L., Li, H., & Gruteser, M. (2019) · The 25th Annual International Conference on Mobile Computing and Networking · doi:10.1145/3300061.3300116 · Cited by 552
A Survey on Bitrate Adaptation Schemes for Streaming Media Over HTTP
Bentaleb, A., Taani, B., Begen, A. C., Timmerer, C., & Zimmermann, R. (2019) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2018.2862938 · Cited by 494
“All theses studies as well as C3 [156], CFA [157], CS2P [158], Pytheas [159] share a common characteristic, which is that there exists a central controller to control, manage and monitor HAS traffic.”
BackgroundMachine Learning for Networking: Workflow, Advances and Opportunities
Wang, M., Cui, Y., Wang, X., Xiao, S., & Jiang, J. (2018) · IEEE Network · doi:10.1109/mnet.2017.1700200 · Cited by 448
“First, as the best known capabilities of ML, classification and prediction play basic but important roles in network problems such as intrusion detection and performance prediction [1].”
BackgroundMethodInfluentialSelf-citationOboe: auto-tuning video ABR algorithms to network conditions
Akhtar, Z., Nam, Y. S., Govindan, R., Rao, S., Chen, J., Katz-Bassett, E., Ribeiro, B., Zhan, J., & Zhang, H. (2018) · Proceedings of the 2018 Conference of the ACM Special Interest Group on Data Communication · doi:10.1145/3230543.3230558 · Cited by 327
“While the magnitude of benefits is smaller with the ideal prediction approach, in practice Oboe will likely result in larger benefits, since even more sophisticated schemes [49] cannot achieve the ideal predictions, and the errors are likely to grow with larger T.”
InfluentialA Survey of Rate Adaptation Techniques for Dynamic Adaptive Streaming Over HTTP
Kua, J., Armitage, G., & Branch, P. (2017) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2017.2685630 · Cited by 318
“[83] developed CS2P (Cross Session Stateful Predictor) to improve the bitrate selection and adapta-”
MethodA Survey on Adaptive 360° Video Streaming: Solutions, Challenges and Opportunities
Yaqoob, A., Bi, T., & Muntean, G.-M. (2020) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2020.3006999 · Cited by 254
“[71] proposed the Cross Session Stateful Predictor (CS2P), a data-driven”
BackgroundDeep learning for edge computing applications: A State-of-the-Art survey
Wang, F., Zhang, M., Wang, X., Ma, X., & Liu, J. (2020) · IEEE Access · doi:10.1109/access.2020.2982411 · Cited by 259
Bui, N., Cesana, M., Hosseini, S. A., Liao, Q., Malanchini, I., & Widmer, J. (2017) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2017.2694140 · Cited by 239
AWStream: adaptive wide-area streaming analytics
Zhang, B., Jin, X., Ratnasamy, S., Wawrzynek, J., & Lee, E. A. (2018) · Proceedings of the 2018 Conference of the ACM Special Interest Group on Data Communication · doi:10.1145/3230543.3230554 · Cited by 233
“Recent work has moved towards HTTP-based protocols and focused on designing adaptation strategy to improve QoE, as in research [47, 75, 88] and industry (HLS [58], DASH [48, 74]).”
MethodLumos5G: Mapping and Predicting Commercial mmWave 5G Throughput
Narayanan, A., Ramadan, E., Mehta, R., Hu, X., Liu, Q., Fezeu, R. A. K., Dayalan, U. K., Verma, S., Ji, P., Li, T., et al. (2020) · Proceedings of the ACM Internet Measurement Conference · doi:10.1145/3419394.3423629 · Cited by 199
“, adaptive video bitrate streaming [27, 58, 62, 70].”
BackgroundInfluentialTang, F., Mao, B., Kawamoto, Y., & Kato, N. (2021) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2021.3073009 · Cited by 199
“The work in [107] proposes a throughput prediction based ABR algorithm.”
MethodSpiteri, K., Sitaraman, R. K., & Sparacio, D. (2019) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3336497 · Cited by 139
Rubiks: Practical 360-Degree Streaming for Smartphones
He, J., Qureshi, M. A., Qiu, L., Li, J., Li, F., & Han, L. (2018) · Proceedings of the 16th Annual International Conference on Mobile Systems, Applications, and Services · doi:10.1145/3210240.3210323 · Cited by 130
“It is widely recognized that the user perceived QoE for a video chunk is determined by the following factors: video quality, quality changes, and rebuffering time [36, 43, 47].”
BackgroundLearning in situ: a randomized experiment in video streamingPreprint
Yan, F. Y., Ayers, H. R., Zhu, C., Fouladi, S., Hong, J. W., Zhang, K., Levis, P. A., & Winstein, K. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1906.01113 · Cited by 317
“nacknowledged packets in flight, the current RTT, the minimum RTT, and the estimated throughput (tcpi_delivery_rate). Prior approaches have used Harmonic Mean (HM) [33] or a Hidden Markov Model (HMM) [29] to predict a single throughput for the entire lookahead horizon. In contrast, the transmission-time predictor outp…”
BackgroundMethodInfluentialDigital Twin for Networking: A Data-Driven Performance Modeling Perspective
Hui, L., Wang, M., Zhang, L., Lu, L., & Cui, Y. (2022) · IEEE Network · doi:10.1109/mnet.119.2200080 · Cited by 133
“Mirza et al. [10] and CS2P [9] both predicted throughput under specific background for file transfer and video bitrate adaptation separately.”
BackgroundMethodInfluentialEdge Computing Assisted Adaptive Mobile Video Streaming
Mehrabi, A., Siekkinen, M., & Ylä-Jääski, A. (2018) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2018.2850026 · Cited by 122
“Since purely buffer-based adaptation mechanisms may fail specially under high throughput fluctuation, some techniques combine the buffer occupancy level with some extent of throughput prediction to determine the quality for the next video chunk to be downloaded [12], [19], [20].”
MethodSDNHAS: An SDN-Enabled Architecture to Optimize QoE in HTTP Adaptive Streaming
Bentaleb, A., Begen, A. C., Zimmermann, R., & Harous, S. (2017) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2017.2733344 · Cited by 99
“and integrate new bitrate decision theories, bandwidth and QoE estimators [19], [44] that use machine learning algorithms such as advanced RL-based techniques.”
BackgroundJiang, J., Sun, Y. S., Sekar, V., & Zhang, H. (2017) · Networked Systems Design and Implementation · Cited by 140
Self-citationLearning to Coordinate Video Codec with Transport Protocol for Mobile Video Telephony
Zhou, A., Zhang, H., Su, G., Wu, L., Ma, R., Meng, Z., Zhang, X., Xie, X., Ma, H., & Chen, X. (2019) · ACM Annual International Conference on Mobile Computing and Networking (MobiCom) · doi:10.1145/3300061.3345430 · Cited by 109
“For instance, CS2P [49] discovers the performance correlation among VoD flows and exploits the correlation to decide an optimal initial bitrate for a new flow.”
BackgroundTowards Perceptually Optimized End-to-end Adaptive Video StreamingPreprint
Bampis, C. G., Li, Z., Katsavounidis, I. X., Huang, T.-Y., Ekanadham, C., & Bovik, A. C. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1808.03898 · Cited by 106
“Client-based ABR strategies can be broadly classified as: throughput-based [37]–[39], buffer-based [2], [40]–[42] and hybrid/control-theoretical approaches [43]–[48].”
BackgroundImproving TCP Congestion Control with Machine Intelligence
Kong, Y., Zang, H., & Ma, X. (2018) · ACM SIGCOMM Workshop on Network Meets AI & ML (NetAI) · doi:10.1145/3229543.3229550 · Cited by 98
Spiteri, K., Sitaraman, R. K., & Sparacio, D. (2018) · ACM Multimedia Systems Conference (MMSys) · doi:10.1145/3204949.3204953 · Cited by 87
Lu, J., Yang, J., Li, S., Li, Y., Wu, J., Dai, J., & Hu, J. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3366252 · Cited by 87
Realtime mobile bandwidth and handoff predictions in 4G/5G networks
Mei, L., Gou, J., Cai, Y., Cao, H., & Liu, Y. (2022) · Computer Networks · doi:10.1016/j.comnet.2021.108736 · Cited by 102
The development of leak detection model in subsea gas pipeline using machine learning
Kim, J., Chae, M., Han, J., Park, S. J., & Lee, Y. (2021) · Journal of Natural Gas Science and Engineering · doi:10.1016/j.jngse.2021.104134 · Cited by 83
Realtime mobile bandwidth prediction using LSTM neural network and Bayesian fusion
Mei, L., Hu, R., Cao, H., Liu, Y., Han, Z., Li, F., & Li, J. (2020) · Computer Networks · doi:10.1016/j.comnet.2020.107515 · Cited by 72
Sensor-Augmented Neural Adaptive Bitrate Video Streaming on UAVs
Xiao, X., Wang, W., Chen, T., Cao, Y., Jiang, T., & Zhang, Q. (2019) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2019.2945167 · Cited by 73
“The majority of existing fixed control rules generate ABR algorithms based on the available bandwidth estimates (rate-based algorithms [9], [10]), playback buffer occupancy (buffer-based algorithms [11], [12], [18]) or the hybrid methods of combining these two approaches ( [13]–[16]).”
BackgroundAdaptive Bitrate with User-level QoE Preference for Video Streaming
Zuo, X., Yang, J., Wang, M., & Cui, Y. (2022) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom48880.2022.9796953 · Cited by 80
ABR streaming of VBR-encoded videos
Qin, Y., Hao, S., Pattipati, K. R., Qian, F., Sen, S. B., Wang, B., & Yue, C. (2018) · ACM International Conference on emerging Networking EXperiments and Technologies (CoNEXT) · doi:10.1145/3281411.3281439 · Cited by 71
“There is a large body of existing work on ABR streaming [1, 6, 8, 16, 20, 23, 28, 33, 38, 39, 41, 44, 47, 49, 51].”
BackgroundCatching the Moment With LoL$^+$ in Twitch-Like Low-Latency Live Streaming Platforms
Bentaleb, A., Akcay, M. N., Lim, M. L., Begen, A. C., & Zimmermann, R. (2021) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2021.3079288 · Cited by 76
Quality-Aware Neural Adaptive Video Streaming With Lifelong Imitation Learning
Huang, T., Zhou, C., Yao, X., Zhang, R.-X., Wu, C., Yu, B., & Sun, L. (2020) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2020.3000363 · Cited by 73
“For example, CS2P [5] increased the QoE by 3.2% compared with MPC [10], and ABRL [65] improved the video quality by 1.6% compared with Pensieve.”
BackgroundMethodResultInfluentialHuang, T., Zhou, C., Zhang, R.-X., Wu, C., & Sun, L. (2022) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2022.3180804 · Cited by 72
“By contrast, learning-based schemes employ several learning technologies, such as reinforcement learning [11], [12], supervised learning [13], [2] and imitation learning [14], [15] to train a neural network (NN) w.r.t the given network traffic distributions, and make a zero-shot inference for…”
BackgroundTowards Lightweight and Robust Machine Learning for CDN Caching
Berger, D. S. (2018) · ACM Workshop on Hot Topics in Networks (HotNets) · doi:10.1145/3286062.3286082 · Cited by 77
“Finally, there is an emerging literature on successful machine learning applications in networking research such as traffic classification [52], video rate adaptation [48], resource allocation [36], and congestion control [47].”
BackgroundHotDASH: Hotspot Aware Adaptive Video Streaming Using Deep Reinforcement Learning
Sengupta, S., Ganguly, N., Chakraborty, S., & De, P. (2018) · IEEE International Conference on Network Protocols (ICNP) · doi:10.1109/icnp.2018.00026 · Cited by 66
“proposed CS2P, a system which optimizes bitrate adaptation based on large-scale data-driven throughput prediction [35].”
MethodStick: A Harmonious Fusion of Buffer-based and Learning-based Approach for Adaptive Streaming
Huang, T., Zhou, C., Zhang, R.-X., Wu, C., Yao, X., & Sun, L. (2020) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom41043.2020.9155411 · Cited by 71
“CS2P [42] assume that throughput factors can be efficiently captured by Hidden-Markov-Model (HMM), then they optimize the model on the cloud with huge amounts of data.”
BackgroundMehrabi, A., Siekkinen, M., Kämäräinen, T., & YlJski, A. (2021) · ACM Trans. Multim. Comput. Commun. Appl. · doi:10.1145/3429441 · Cited by 33
Lee, R., Venieris, S. I., & Lane, N. D. (2021) · ACM Computing Surveys · doi:10.1145/3469094 · Cited by 73
“Formany years, bitrate adaptation has been the dominant approach for counteracting the networking dynamicity and the main driver behind adaptive video streaming [11], with already deployed solutions by industry content providers such as iQiyi [140], Netflix [62] and YouTube [103].”
LBP: Robust Rate Adaptation Algorithm for SVC Video Streaming
Elgabli, A., Aggarwal, V., Hao, S., Qian, F., & Sen, S. B. (2018) · IEEE/ACM Transactions on Networking · doi:10.1109/tnet.2018.2844123 · Cited by 55
“[19] proposed crowd-source based bandwidth prediction and used the streaming algorithm proposed in [6] to make the bit rate decision per video’s chunk.”
BackgroundMethodData-Driven Evaluation of Anticipatory Networking in LTE Networks
Bui, N., & Widmer, J. (2018) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2018.2809750 · Cited by 54
“[25], [26] propose a throughput prediction solution based on clustering and hidden Markov models.”
BackgroundCase for 5G-aware video streaming applications
Ramadan, E., Narayanan, A., Dayalan, U. K., Fezeu, R., Qian, F., & Zhang, Z.-L. (2021) · 5G-MeMZ@SIGCOMM · doi:10.1145/3472771.3474036 · Cited by 55
“Such high accuracy is shown to be adequate for video ABR adaptation [28].”
BackgroundImproving Generalization for Neural Adaptive Video Streaming via Meta Reinforcement Learning
Kan, N., Jiang, Y., Li, C., Dai, W., Zou, J., & Xiong, H. (2022) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3503161.3548331 · Cited by 53
“, CS2P[22], BayesMPC[10], Fugu[23]) have been proposed to seek for a higher QoE by improving the throughput prediction accuracy.”
BackgroundInfluentialTowards Perceptually Optimized Adaptive Video Streaming-A Realistic Quality of Experience Database
Bampis, C. G., Li, Z., Katsavounidis, I. X., Huang, T.-Y., Ekanadham, C., & Bovik, A. C. (2021) · IEEE Transactions on Image Processing · doi:10.1109/tip.2021.3073294 · Cited by 62
“Client-based ABR strategies can be broadly classified as: throughput-based [43]–[45], buffer-based [2], [46]–[48] and hybrid/control-theoretical approaches [49]–[55].”
BackgroundThe QUIC Fix for Optimal Video Streaming
Palmer, M., Krüger, T., Chandrasekaran, B., & Feldmann, A. (2018) · doi:10.1145/3284850.3284857 · Cited by 59
“frastructures [21, 29]. In this section we briefly discuss only those most relevant to our work. Adaptive bitrate schemes. Buffer-based and rate-based schemes that dynamically adapt the video bitrate [19, 22, 30, 43, 50] suffer invariably from the limitations of the underlying transport: these schemes simply operate o…”
BackgroundZhang, L., Suo, Y., Wu, X., Wang, F., Chen, Y., Cui, L., Liu, J., & Ming, Z. (2021) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3474085.3475590 · Cited by 50
“Adaptive streaming has long been the research interest for conventional videos [4, 17, 24, 35].”
BackgroundLumos: towards Better Video Streaming QoE through Accurate Throughput Prediction
Lv, G., Wu, Q., Wang, W., Li, Z., & Xie, G. (2022) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom48880.2022.9796948 · Cited by 50
ARBITER+: Adaptive Rate-Based InTElligent HTTP StReaming Algorithm for Mobile Networks
Zahran, A. H., Raca, D., & Sreenan, C. J. (2018) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2018.2825384 · Cited by 49
“[24] presented a data-drive throughput prediction model that improved the performance of different streaming strategies after integrating the prediction model as a rate estimator.”
TRUST: A TCP Throughput Prediction Method in Mobile Networks
Wei, B., Kawakami, W., Kanai, K., Katto, J., & Wang, S. (2018) · Globecom · doi:10.1109/glocom.2018.8647390 · Cited by 49
“Although the experimental results indicated that HMM method is effective, the method did not consider scenarios of moving user, and it only used historical throughput data for the prediction.”
MethodInfluentialA Supervised Learning Based QoS Assurance Architecture for 5G Networks
Zhu, G., Zan, J., Yang, Y., & Qi, X. (2019) · IEEE Access · doi:10.1109/access.2019.2907142 · Cited by 61
T-Gaming: A Cost-Efficient Cloud Gaming System at Scale
Chen, H., Zhang, X., Xu, Y., Ren, J., Fan, J., Ma, Z., & Zhang, W. (2019) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2019.2922205 · Cited by 57
“Existing video streaming strategies are mainly based on the HTTP protocol [6], [31], [32], [33], [34], [35], [36].”
MethodTCLiVi: Transmission Control in Live Video Streaming Based on Deep Reinforcement Learning
Cui, L., Su, D., Yang, S., Wang, Z., & Ming, Z. (2020) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2020.2985631 · Cited by 50
Optimized Adaptive Streaming of Multi-video Stream Bundles
Carlsson, N. F., Eager, D. L., Krishnamoorthi, V., & Polishchuk, T. (2017) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2017.2673412 · Cited by 47
“In the context of linear video (without viewpoint selection), much work has been done on HAS rate adaptation [16], [38], [39].”
BackgroundMethodSuffering from buffering? Detecting QoE impairments in live video streams
Ahmed, A., Shafiq, Z., Bedi, H. S., & Khakpour, A. R. (2017) · IEEE International Conference on Network Protocols (ICNP) · doi:10.1109/icnp.2017.8117561 · Cited by 47
“Since end-to-end network bandwidth fluctuates over time, adaptive bitrate controllers [6], [19], [26], [34], [35] are implemented at the client side to estimate network bandwidth, request video at appropriate bitrates and optimize tradeoffs associated with QoE metrics and streaming bitrate.”
BackgroundMosaic: Advancing User Quality of Experience in 360-Degree Video Streaming With Machine Learning
Park, S., Bhattacharya, A., Yang, Z., Das, S. R., & Samaras, D. (2021) · IEEE Transactions on Network and Service Management · doi:10.1109/tnsm.2021.3053183 · Cited by 47
“Pensieve and CS2P use neural networks and data-driven rate adaptation scheme respectively, while Festive and Pytheas are based on throughput and network-capacity.”
BackgroundFleet: Improving Quality of Experience for Low-Latency Live Video Streaming
Li, Y., Zhang, X., Cui, C., Wang, S., & Ma, S. (2023) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2023.3243901 · Cited by 46
“CS2P uses a Hidden Markov Model (HMM) to capture and predict bandwidth variability [52].”
MethodUnleashing the Potential of Data-Driven Networking
Jiang, J., Sekar, V., Stoica, I. G., & Zhang, H. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-67235-9_9 · Cited by 54
“, each client runs a local control logic that can be updated based on a multi-session view on a coarse timescale [54]).”
BackgroundSelf-citationAdvancing User Quality of Experience in 360-degree Video Streaming
Park, S., Bhattacharya, A., Yang, Z., Dasari, M., Das, S. R., & Samaras, D. (2019) · IFIP Networking Conference (IFIP Networking) · doi:10.23919/ifipnetworking46909.2019.8999460 · Cited by 49
“Pensieve and CS2P use neural networks and data-driven rate adaptation scheme respectively, while Festive and Pytheas are based on throughput and network-capacity.”
BackgroundTurkkan, B. O., Dai, T., Raman, A., Kosar, T., Chen, C., Bulut, M. F., Żola, J., & Sow, D. (2022) · ACM Multimedia Systems Conference (MMSys) · doi:10.1145/3524273.3528188 · Cited by 45
MetaABR: A Meta-Learning Approach on Adaptative Bitrate Selection for Video Streaming
Li, W., Li, X., Xu, Y., Yi, Y., & Lu, S. (2023) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2023.3260086 · Cited by 52
“Selecting the right bitrate in dynamic network is challenging due to the variability of network conditions and the trade-off of conflicting video QoE requirements [3], [4], [5].”
BackgroundMethodInfluentialSODA: An Adaptive Bitrate Controller for Consistent High-Quality Video Streaming
Chen, T., Lin, Y., Christianson, N., Akhtar, Z., Dharmaji, S., Hajiesmaili, M., Wierman, A., & Sitaraman, R. K. (2024) · ACM SIGCOMM Conference (SIGCOMM) · doi:10.1145/3651890.3672260 · Cited by 43
Assessing the Quality-of-Experience of Adaptive Bitrate Video StreamingPreprint
Duanmu, Z., Liu, W., Li, Z., Chen, D., Wang, Z., Wang, Y., & Gao, W. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2008.08804 · Cited by 43
Optimizing Quality of Experience for Adaptive Bitrate Streaming via Viewer Interest Inference
Gao, G., Zhang, H., Hu, H., Wen, Y., Cai, J., Luo, C., & Zeng, W. (2018) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2018.2838330 · Cited by 46
“[34] advocated that the precise bandwidth prediction can improve the performance of rate adaptation.”
BackgroundBentaleb, A., Begen, A. C., Harous, S., & Zimmermann, R. (2018) · ACM Multimedia Systems Conference (MMSys) · doi:10.1145/3204949.3204961 · Cited by 42
Realtime Mobile Bandwidth Prediction Using LSTM Neural Network
Mei, L., Hu, R., Cao, H., Liu, Y., Han, Z., Li, F., & Li, J. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-15986-3_3 · Cited by 41
Towards Optimal Low-Latency Live Video Streaming
Sun, L., Zong, T., Wang, S., Liu, Y., & Wang, Y. (2021) · IEEE/ACM Transactions on Networking · doi:10.1109/tnet.2021.3087625 · Cited by 40
R-FEC: RL-based FEC Adjustment for Better QoE in WebRTC
Lee, I.-S., Kim, S., Sathyanarayana, S. D., Bin, K., Chong, S., Lee, K., Grunwald, D. C., & Ha, S. (2022) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3503161.3548370 · Cited by 40
“Many studies on ABR video streaming have appeared in the past few years to improve the performance [1, 23, 25].”
BackgroundOptimized Preference-Aware Multi-Path Video Streaming with Scalable Video Coding
Elgabli, A., Liu, K., & Aggarwal, V. (2018) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2018.2889039 · Cited by 39
“In recent studies, various approaches for making ABR streaming decisions have been investigated, for example, by using control theory [19], [20], Markov Decision Process [21], machine learning [22], client buffer information [13], and data-driven techniques [23], [24], [25].”
MethodData-Driven Bandwidth Prediction Models and Automated Model Selection for Low Latency
Bentaleb, A., Begen, A. C., Harous, S., & Zimmermann, R. (2020) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2020.3013387 · Cited by 43
Quality of Experience-Centric Management of Adaptive Video Streaming Services: Status and Challenges
Petrangeli, S., Hooft, J. V. D., Wauters, T., & Turck, F. D. (2018) · ACM Transactions on Multimedia Computing, Communications, and Applications · doi:10.1145/3165266 · Cited by 38
“The CS2P algorithm by Sun et al. falls in this category [93], for example.”
BackgroundMethodOptimizing Adaptive Video Streaming with Human Feedback
Huang, T., Zhang, R.-X., Wu, C., & Sun, L. (2023) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3581783.3611771 · Cited by 42
Reinforcement learning-based QoE-oriented dynamic adaptive streaming framework
Wei, X., Zhou, M., Kwong, S., Yuan, H., Wang, S., Zhu, G., & Cao, J. (2021) · Information Sciences · doi:10.1016/j.ins.2021.05.012 · Cited by 37
Hu, H., Wen, Y., & Niyato, D. T. (2017) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2017.2676598 · Cited by 36
“In the traditional video streaming service, the QoS/QoE function is measured as the summation of four terms, including average bitrate, bitrate variation, buffer outage probability, and startup delay [40], [41].”
MethodImpact of Device Performance on Mobile Internet QoE
Dasari, M., Vargas, S., Bhattacharya, A., Balasubramanian, A., Das, S. R., & Ferdman, M. (2018) · ACM Internet Measurement Conference (IMC) · doi:10.1145/3278532.3278533 · Cited by 36
“Pytheas [18] and CS2P [32] propose data-driven approaches to study the impact of different parameters that impact QoE.”
BackgroundExploring the interplay between CDN caching and video streaming performance
Ghabashneh, E., & Rao, S. (2020) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom41043.2020.9155338 · Cited by 45
“Researchers [35] have developed better throughput prediction approaches for video streaming based on Hidden Markov Models.”
BackgroundMultiLive: Adaptive Bitrate Control for Low-Delay Multi-Party Interactive Live Streaming
Wang, Z., Cui, Y., Hu, X., Wang, X., Ooi, W. T., Cao, Z. Z., & Li, Y. (2021) · IEEE/ACM Transactions on Networking · doi:10.1109/tnet.2021.3129481 · Cited by 35
Dai, P., Wu, M., Li, K., Wu, X. Y., & Ding, Y. F. (2024) · IEEE Transactions on Services Computing · doi:10.1109/tsc.2023.3349051 · Cited by 34
“Theauthors in [12] developed CS2, a throughput prediction system that learns clusters of similar sessions, throughput predictors, and Hidden-Markov-Model midstream predictors.”
BackgroundRldish: Edge-Assisted QoE Optimization of HTTP Live Streaming with Reinforcement Learning
Wang, H., Wu, K., Wang, J., & Tang, G. (2020) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom41043.2020.9155467 · Cited by 34
“The existing research mainly uses adaptive bitrate algorithms to improve the QoE of viewers via dynamically choosing a bitrate for each video segment [7], [8], [9].”
MethodDetecting if LTE is the Bottleneck with BurstTracker
Balasingam, A., Bansal, M., Misra, R., Nagaraj, K., Tandra, R., Katti, S., & Schulman, A. (2019) · ACM Annual International Conference on Mobile Computing and Networking (MobiCom) · doi:10.1145/3300061.3300140 · Cited by 34
“Prior work has used application-layer throughput measurements to estimate future network performance [20, 30, 34].”
MethodOptimizing Adaptive Coding and Modulation for Satellite Network with ML-based CSI Prediction
Wang, X., Li, H., & Wu, Q. (2019) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc.2019.8885616 · Cited by 33
“adaptive video bitrate based on network throughput prediction [17], [18].”
Cuttlefish: Neural Configuration Adaptation for Video Analysis in Live Augmented Reality
Chen, N., Quan, S., Zhang, S., Qian, Z., Jin, Y., Wu, J., Li, W., & Lu, S. (2020) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2020.3035044 · Cited by 35
“Bandwidth Based Adaptation [31].”
BackgroundEmpowering video players in cellular: throughput prediction from radio network measurements
Raca, D., Zahran, A. H., Sreenan, C. J., Sinha, R. K., Halepovic, E., Jana, R., Gopalakrishnan, V., Bathula, B. G., & Varvello, M. (2019) · ACM Multimedia Systems Conference (MMSys) · doi:10.1145/3304109.3306233 · Cited by 39
An Ensemble Rate Adaptation Framework for Dynamic Adaptive Streaming Over HTTP
Yuan, H., Hu, X., Hou, J., Wei, X., & Kwong, S. (2019) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2019.2954074 · Cited by 33
“, methods for a single server [6], [17]–[35] and methods for multiservers [18], [36], [37].”
MethodAdaptive Streaming of 360-Degree Videos with Reinforcement Learning
Park, S., Hoai, M., Bhattacharya, A., & Das, S. R. (2021) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv48630.2021.00188 · Cited by 32
“ML techniques have been used to improve video streaming but most prior methods are based on supervised or unsupervised learning [40, 33, 6, 38, 15, 26].”
MethodOsmoticGate: Adaptive Edge-Based Real-Time Video Analytics for the Internet of Things
Qian, B., Wen, Z., Tang, J., Yuan, Y., Zomaya, A. Y., & Ranjan, R. K. (2022) · IEEE Transactions on Computers · doi:10.1109/tc.2022.3193630 · Cited by 36
“The existing work usually considers two factors: i) Rate-based algorithms that decide the bitrate based on network bandwidth assumption [40], [41], [42]. ii) Buffer-based algorithms that consider the client’s play-back buffer [43], [44].”
BackgroundEdge-Assisted Real-Time Video Analytics With Spatial–Temporal Redundancy Suppression
Wang, Z., He, X., Zhang, Z., Zhang, Y., Cao, Z. Z., Cheng, W., Wang, W., & Cui, Y. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3224750 · Cited by 34
“Recent work has moved toward HTTP-based protocols and focused on designing adaptation strategy to improve the performance, as in research (Pensieve [31], CS2P [32]) and industry (HLS [33], DASH [34]).”
MethodEAVS: Edge-assisted Adaptive Video Streaming with Fine-grained Serverless Pipelines
Hou, B., Yang, S., Kuipers, F. A., Jiao, L., & Fu, X. (2023) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom53939.2023.10229102 · Cited by 35
Sammy: smoothing video traffic to be a friendly internet neighbor
Spang, B., Kunamalla, S., Teixeira, R., Huang, T.-Y., Armitage, G. J., Johari, R., & McKeown, N. (2023) · ACM SIGCOMM Conference (SIGCOMM) · doi:10.1145/3603269.3604839 · Cited by 32
“For separate systems that predict initial throughput like CS2P [66], this can be done by supplying this system only with initial throughput measurements.”
BackgroundMethodA Global Perspective on the Past, Present, and Future of Video Streaming over Starlink
Izhikevich, L., Enghardt, T., Huang, T.-Y., & Teixeira, R. (2024) · Proceedings of the ACM on Measurement and Analysis of Computing Systems · doi:10.1145/3700412 · Cited by 31
“At a high level, 5 this algorithm follows a simple Model Predictive Control Style ABR model [47] that picks the highest sustainable bitrate for chunk 𝑖 (denoted as 𝑟 𝑖 ) such that 𝑟 𝑖 ≤ 𝑓 ( 𝑇 𝑖 , 𝐵 𝑖 − 1 ) , where 𝑇 𝑖 represents the estimated throughput for downloading chunk 𝑖 and 𝐵…”
Proactive Video Push for Optimizing Bandwidth Consumption in Hybrid CDN-P2P VoD Systems
Zhang, Y., Gao, C., Guo, Y., Bian, K., Jin, X., Yang, Z. C., Song, L., Cheng, J., Tuo, H., & Li, X. (2018) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom.2018.8485962 · Cited by 31
“Markovian models learn a stationary distribution over a predefined or automatically deciphered state space, and Hidden markov models (HMM) [17] for the hidden states.”
BackgroundRoPE: An Architecture for Adaptive Data-Driven Routing Prediction at the Edge
Sacco, A., Esposito, F., & Marchetto, G. (2020) · IEEE Transactions on Network and Service Management · doi:10.1109/tnsm.2020.2980899 · Cited by 36
“In our implementation, we use a collection period of 5-seconds, as in [5].”
BackgroundMethodMachine Learning for Computer Systems and Networking: A Survey
Kanakis, M. E., Khalili, R., & Wang, L. (2022) · ACM Computing Surveys · doi:10.1145/3523057 · Cited by 40
“In the context of adaptive video streaming, CS2P aims to improve bitrate adaptation by adopting data-driven approaches in throughput prediction [154].”
MethodUtility-Oriented Optimization for Video Streaming in UAV-Aided MEC Network: A DRL Approach
Miao, J., Bai, S., Mumtaz, S., Zhang, Q., & Mu, J. (2024) · IEEE Transactions on Green Communications and Networking · doi:10.1109/tgcn.2024.3352173 · Cited by 30
“Therefore, in the MDP, the distance is directly adopted as state.”
BackgroundA Control Theoretic Approach to ABR Video Streaming: A Fresh Look at PID-Based Rate Adaptation
Qin, Y., Jin, R., Hao, S., Pattipati, K. R., Qian, F., Sen, S. B., Yue, C., & Wang, B. (2019) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2019.2929125 · Cited by 30
“CS2P [36] uses a data-driven approach to improve the bandwidth prediction for ABR streaming.”
MethodPiTree: Practical Implementation of ABR Algorithms Using Decision Trees
Meng, Z., Chen, J., Guo, Y., Sun, C., Hu, H., & Xu, M. (2019) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3343031.3350866 · Cited by 37
“Hybrid (ML) CS2P [58] Pensieve [47] HotDASH [53]”
BackgroundQoE-Aware Bandwidth Broker for HTTP Adaptive Streaming Flows in an SDN-Enabled HFC Network
Bentaleb, A., Begen, A. C., & Zimmermann, R. (2018) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2018.2816789 · Cited by 29
“cluster identification algorithm from [24]: video sessions of the HAS clients that match based on defined characteristics such as IP, ISP, region, city and MSO are aggregated.”
Izima, O., de Fréin, R., & Malik, A. (2021) · Electronics · doi:10.3390/electronics10222851 · Cited by 31
$\mathsf{HxL3}$: Optimized Delivery Architecture for HTTP Low-Latency Live Streaming
Tashtarian, F., Bentaleb, A., Erfanian, A., Hellwagner, H., Timmerer, C., & Zimmermann, R. (2022) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2022.3148587 · Cited by 29
“Similarly, Sun et al. [28] proposed CS2P, a data-driven based throughput prediction framework.”
BackgroundSmartStreamer: Preference-Aware Multipath Video Streaming Over MPTCP
Elgabli, A., & Aggarwal, V. (2019) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2019.2915355 · Cited by 28
“cess [16], machine learning [17], client buffer information [18], and data-driven techniques [19]–[21].”
Metadata
- Authors
- Sun, Y., Yin, X., Jiang, J., Sekar, V., Lin, F., Wang, N., Liu, T., & Sinopoli, B.
- Affiliations
- ICT/CAS
- Capital Medical University (CN)
- iQIYI (China) (CN)
- Venue
- Proceedings of the 2016 ACM SIGCOMM Conference
- Publisher
- ACM
- Published
- 2016-08-22
- Pages
- 272-285
- OpenAlex
- W2482797934
- Cited by
- 407
- References
- 38
- Citations by year
- 2026: 20 · 2025: 25 · 2024: 40 · 2023: 43 · 2022: 40 · 2021: 57 · 2020: 55 · 2019: 50 · 2018: 54 · 2017: 21 · 2016: 2
- FWCI
- 36.18
- Type
- conference paper
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- ACM SIGCOMM Conference (SIGCOMM) (publishedVersion)
- Funders
- National Science Foundation (1345305)
- National Natural Science Foundation of China (2012CB315802, 61379133, 61133015)
- National Key Research and Development Program of China (2012CB315802)
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › Network Traffic and Congestion Control
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Image and Video Quality Assessment
- Physical Sciences › Computer Science › Computer Networks and Communications › Caching and Content Delivery
- Keywords
- throughput prediction, bitrate adaptation, video streaming, quality of experience, hidden Markov models, session clustering
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.