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
Items 101–200 of 428
Vabis: Video Adaptation Bitrate System for Time-Critical Live Streaming
Feng, T., Sun, H., Qi, Q., Wang, J., & Liao, J. (2019) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2019.2962313 · Cited by 28
“1109/TMM.2019.2962313 high quality and low rebuffering user experience [2], [3], such as buffer-based methods [4], rate-based methods [5], quality-based methods [6], and RL-based methods [7].”
BackgroundMethodInfluentialSequential Reinforced 360-Degree Video Adaptive Streaming With Cross-User Attentive Network
Fu, J., Chen, Z., Chen, X., & Li, W. (2020) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2020.3028329 · Cited by 28
“The available bandwidth estimation methods include rate-based ones [24], [25], buffer-based ones [26], [27], and target-buffer based ones [9].”
MethodZwei: A Self-Play Reinforcement Learning Framework for Video Transmission Services
Huang, T., Zhang, R.-X., & Sun, L. (2021) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2021.3063620 · Cited by 32
“This makes sense since recent work has proved that these two metrics are critical features for the ABR task [16].”
BackgroundVOXEL: cross-layer optimization for video streaming with imperfect transmission
Palmer, M., Appel, M., Spiteri, K., Chandrasekaran, B., Feldmann, A., & Sitaraman, R. K. (2021) · ACM International Conference on emerging Networking EXperiments and Technologies (CoNEXT) · doi:10.1145/3485983.3494864 · Cited by 28
“To make clients smarter, prior work also proposed numerous ways of improving ABR algorithms [23, 33, 43, 63, 64, 73], which are client-side mechanisms that dynamically choose a bitrate for each video segment based on several factors, e.g., available bandwidth estimate and playback buffer occupancy.”
BackgroundA machine learning-based framework for preventing video freezes in HTTP adaptive streaming
Petrangeli, S., Wu, T., Wauters, T., Huysegems, R., Bostoen, T., & Turck, F. D. (2017) · Journal of Network and Computer Applications · doi:10.1016/j.jnca.2017.07.009 · Cited by 28
“veloping a prediction model based on past video sessions, which is built offline in a node located in the streaming provider network [10].”
BackgroundSENSEI: Aligning Video Streaming Quality with Dynamic User SensitivityPreprint
Zhang, X., Ou, Y., Sen, S., & Jiang, J. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2008.04687 · Cited by 28
“ed algorithms (e.g., [33 ,50 51]). More recent ABR algorithms combine the two approaches to optimize for explicit QoE objectives, via control theory [61], machine learning-based throughput prediction [53,60], or deep reinforcement learning [41]. Key parameters of the ABR logic can be customized to the network conditio…”
Self-citationChorus: Coordinating Mobile Multipath Scheduling and Adaptive Video Streaming
Lv, G., Wu, Q., Liu, Y., Li, Z., Tan, Q., Yang, F., Chen, W., Ma, Y., Guo, H., Chen, Y., et al. (2024) · ACM Annual International Conference on Mobile Computing and Networking (MobiCom) · doi:10.1145/3636534.3649359 · Cited by 27
Tiyuntsong: A Self-Play Reinforcement Learning Approach for ABR Video Streaming
Huang, T., Yao, X., Wu, C., Zhang, R.-X., Pang, Z., & Sun, L. (2019) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2019.00289 · Cited by 27
“In recent work [13, 6], the lifetime of each bitrate decision has been modeled as an MDP (Markov Decision Process), meaning that action is only related to the status of the target rather than relying on the prior states.”
BackgroundBentaleb, A., Yadav, P. K., Ooi, W. T., & Zimmermann, R. (2020) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3371040 · Cited by 27
“Li et al. and Sun et al. both designed throughput-based ABR methods [36, 54], where the former provides a probe mechanism to estimate the throughput when multiple clients compete at a bottleneck link, while the latter is based on a data-driven approach to predict the available bandwidth accurately…”
BackgroundSemantic Information Marketing in the Metaverse: A Learning-Based Contract Theory Framework
Lotfi, I., Niyato, D. T., Sun, S., Kim, D. I., & Shen, X. (2023) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2023.3345402 · Cited by 26
“The Metaverse user is also required to have enough bandwidth to receive data from the VSP in addition to its internal hardware specifications, e.g., screen refresh rate [25].”
MethodAlsader, M., Barakabitze, A. A., & Mkwawa, I.-H. M. (2025) · IEEE Access · doi:10.1109/access.2025.3597058 · Cited by 35
Balancing Quality of Experience and Traffic Volume in Adaptive Bitrate Streaming
Kimura, T., Kimura, T., Matsumoto, A., & Yamagishi, K. (2021) · IEEE Access · doi:10.1109/access.2021.3052552 · Cited by 25
“Model Prediction Control (MPC) [17], Cross Session Stateful Pre-dictor (CS2P) [18], and Pensieve [19] use both throughput and buffer length to select the bitrate.”
MethodXMAS: An Efficient Mobile Adaptive Streaming Scheme Based on Traffic Shaping
Kim, S., & Kim, C.-K. (2018) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2018.2856626 · Cited by 24
“world-wide trend of machine learning techniques, There are several algorithms [39], [45], [47], [48], [64] on the basis of optimization or machine-learning techniques.”
BackgroundA Survey on Client Throughput Prediction Algorithms in Wired and Wireless Networks
Schmid, J., Höß, A., & Schuller, B. W. (2021) · ACM Computing Surveys · doi:10.1145/3477204 · Cited by 29
Data analysis on video streaming QoE over mobile networks
Wang, Q., Dai, H., Wu, D., & Xiao, H. Y. (2018) · EURASIP Journal on Wireless Communications and Networking · doi:10.1186/s13638-018-1180-8 · Cited by 24
Bridging machine learning and computer network research: a survey
Yang, C. H., Geng, J., Wang, Y., Li, J., Li, D., & Wu, J. (2018) · CCF Transactions on Networking · doi:10.1007/s42045-018-0009-7 · Cited by 25
“Inspired by the similarity of throughput patterns among different video sessions, CS2P establishes a Hidden-Markov-Model (HMM) to predict the throughput and further execute adaption decisions.”
MethodA Meta-Learning Framework for Learning Multi-User Preferences in QoE Optimization of DASH
Huo, L., Wang, Z., Xu, M., Li, Y., Ding, Z., & Wang, H. (2019) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2019.2939282 · Cited by 27
“Several techniques [5], [18] were proposed to assess bandwidth more accurately for flexible adaptation.”
BackgroundMethodEnsemble Adaptive Streaming – A New Paradigm to Generate Streaming Algorithms via Specializations
Zhang, G., & Lee, J. Y. B. (2019) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2019.2909202 · Cited by 23
“When throughput predictions are incorrect, RobustMPC’s performance can degrade significantly [22].”
BackgroundReducing Traffic Wastage in Video Streaming via Bandwidth-Efficient Bitrate Adaptation
Su, H., Wang, S., Yang, S., Huang, T., & Ren, X. (2024) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2024.3373498 · Cited by 22
“A hidden Markov model (HMM) is proposed in CS2P [30] modeling the stateful evolution of intra-session throughput.”
MethodAccurate Throughput Prediction for Improving QoE in Mobile Adaptive Streaming
Lv, G., Wu, Q., Tan, Q., Wang, W., Li, Z., & Xie, G. (2023) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2023.3313592 · Cited by 22
“Existing ABR algorithms can be classified into four categories: rate-based (e.g., [7], [9], [10]), buffer-based (e.g., [22], [24]), mixed (e.g., [8]) and learning-based (e.g., [23]).”
BackgroundMethodResultInfluentialContinuous Bitrate & Latency Control with Deep Reinforcement Learning for Live Video Streaming
Hong, R., Shen, Q., Zhang, L., & Wang, J. (2019) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3343031.3356063 · Cited by 22
MPC-CSAS: Multi-Party Computation for Real-Time Privacy-Preserving Speed Advisory Systems
Liu, M., Cheng, L., Gu, Y., Wang, Y., Liu, Q., & O'Connor, N. E. (2022) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2021.3052840 · Cited by 22
On the efficiency and fairness of multiplayer HTTP-based adaptive video streaming
Yin, X., Bartulović, M., Sekar, V., & Sinopoli, B. (2017) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.23919/acc.2017.7963606 · Cited by 21
“A significant amount of research has been focused recently on understanding and designing better bitrate control algorithms [22], [21], [11], [14], [18].”
BackgroundSelf-citationLatency Aware Adaptive Video Streaming using Ensemble Deep Reinforcement Learning
Zhao, Y., Shen, Q., Li, W., Xu, T., Niu, W.-H., & Xu, S.-R. (2019) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3343031.3356071 · Cited by 21
“1) Rate-based: These approaches select the highest available bitrate based on the estimated throughput value [4, 12].”
BackgroundTrace-Driven Optimization on Bitrate Adaptation for Mobile Video Streaming
Qiao, C., Li, G., Ma, Q., Wang, J., & Liu, Y. (2020) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2020.3036707 · Cited by 20
AMIS-MU: Edge Computing Based Adaptive Video Streaming for Multiple Mobile Users
Mu, P. K., Zheng, J., Luan, T. H., Zhu, L., Su, Z., & Dong, M. (2022) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2022.3225414 · Cited by 20
“Client-side ABR algorithms have been studied for a long time, from heuristic logic [13] to stochastic or control-based algorithms [14], [15], [16], [17], [18], and then machine learning-based algorithms rise in recent years [19], [20], [21], [22], [23].”
BackgroundContent-aware QoE optimization in MEC-assisted Mobile video streaming
Rahman, W. U., & Huh, E. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-15163-w · Cited by 20
“The existing algorithms developed fixed control rules to select the video quality based on the estimated throughput [18, 42], the playback buffer level [14], or a combination of the two parameters [15, 28, 33, 34].”
BackgroundMethodVisage: enabling timely analytics for drone imagery
Jha, S., Li, Y., Noghabi, S. A., Ranganathan, V. N., Kumar, P., Nelson, A., Toelle, M., Sinha, S. N., Chandra, R., & Badam, A. (2021) · ACM Annual International Conference on Mobile Computing and Networking (MobiCom) · doi:10.1145/3447993.3483273 · Cited by 21
Toy-IoT-Oriented data-driven CDN performance evaluation model with deep learning
Zhang, W., Lu, Z., Wu, Z., Wu, J., Zou, H., & Huang, S. (2018) · Journal of Systems Architecture · doi:10.1016/j.sysarc.2018.05.005 · Cited by 19
AMIS: Edge Computing Based Adaptive Mobile Video Streaming
Mu, P. K., Zheng, J., Luan, T. H., Zhu, L., Dong, M., & Su, Z. (2021) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom42981.2021.9488426 · Cited by 19
ORL-SDN: Online Reinforcement Learning for SDN-Enabled HTTP Adaptive Streaming
Bentaleb, A., Begen, A. C., & Zimmermann, R. (2018) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3219752 · Cited by 19
“We assume that the network bandwidth variations and congestion levels for the next few steps (several seconds) vary following a Markovian model [57], which is unknown to the SDN-based HAS architecture entities.”
BackgroundMethodInfluentialEvaluation of Throughput Prediction for Adaptive Bitrate Control Using Trace-Based Emulation
Wei, B., Song, H., Wang, S., Kanai, K., & Katto, J. (2019) · IEEE Access · doi:10.1109/access.2019.2909399 · Cited by 19
“They are the hybrid model of the autoregressive model and HMM (Hybrid) [15] and throughput prediction based on LSTM (TRUST) [16].”
MethodInfluentialEdgeDASH: Exploiting Network-Assisted Adaptive Video Streaming for Edge Caching
Bayhan, S., Maghsudi, S., & Zubow, A. (2020) · IEEE Transactions on Network and Service Management · doi:10.1109/tnsm.2020.3037147 · Cited by 19
“As rate estimation approach, we use harmonic-rate estimation algorithm [38] which computes the harmonic mean of the previously downloaded five chunks.”
MethodBentaleb, A., Timmerer, C., Begen, A. C., & Zimmermann, R. (2020) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3387921 · Cited by 19
“Using data-driven prediction of the video quality, Sun et al. [31] presented a cross session stateful bandwidth predictor (CS2P) solution that improves the ABR decisions.”
MethodAdaEnlight: Energy-aware Low-light Video Stream Enhancement on Mobile Devices
Liu, S., Li, X., Zhou, Z., Guo, B., Zhang, M., Shen, H., & Yu, Z. (2022) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3569464 · Cited by 25
Achieving QoE Fairness in Bitrate Allocation of 360° Video Streaming
Li, Z., Zhong, P., Huang, J., Gao, F., & Wang, J. (2023) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2023.3277286 · Cited by 18
“CS2P [35] uses a data-driven approach to analyze the similar session, and employs the hidden Markov data model to predict future throughput to choose the suitable bitrate.”
MethodPerceptual QoE-Optimal Resource Allocation for Adaptive Video Streaming
Eswara, N., Chakraborty, S., Sethuram, H. P., Kuchi, K., Kumar, A., & Channappayya, S. S. (2019) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2019.2954064 · Cited by 19
Understanding quality of experience of heuristic-based HTTP adaptive bitrate algorithms
Taraghi, B., Bentaleb, A., Timmerer, C., Zimmermann, R., & Hellwagner, H. (2021) · International Workshop on Network and Operating Systems Support for Digital Audio and Video (NOSSDAV) · doi:10.1145/3458306.3458875 · Cited by 18
DAVE: Dynamic Adaptive Video Encoding for Real-time Video Streaming Applications
Huang, S., & Xie, J. L. (2021) · IEEE International Conference on Sensing, Communication, and Networking (SECON) · doi:10.1109/secon52354.2021.9491588 · Cited by 21
Wang, C., Guan, J., Feng, T., Zhang, N., & Cao, T. (2019) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3343031.3356069 · Cited by 18
PLVER: Joint Stable Allocation and Content Replication for Edge-Assisted Live Video Delivery
Wang, H., Tang, G., Wu, K., & Wang, J. (2021) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2021.3090784 · Cited by 18
“, subnet, ISP, location) and thus are likely to experience similar QoE when dispatched to the same server [25], [26].”
BackgroundNaresh, M., Das, V., Saxena, P., & Gupta, M. (2022) · Computing · doi:10.1007/s00607-021-01046-1 · Cited by 17
“RB [37] uses prior chunk downloads to estimate available network throughput and then request chunks at a bit rate that maximizes the throughput without congesting the network.”
BackgroundMethodInfluentialPrediction‐Based Resource Deployment and Task Scheduling in Edge‐Cloud Collaborative Computing
Su, M., Wang, G., & Choo, K. R. (2022) · Wireless Communications and Mobile Computing · doi:10.1155/2022/2568503 · Cited by 17
Fang, H., Zhao, H., Shi, J., Zhang, M., Wu, G., Chou, Y. C., Wang, F., & Liu, J. (2024) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3664647.3680712 · Cited by 17
Wang, B., Ren, F., Yang, J., & Zhou, C. (2019) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2019.2939124 · Cited by 17
“Recent studies [19], [20] show that the observed throughput fluctuates around the link capacity following Gaussian.”
BackgroundMethodInfluentialA Cloud Computing Based Deep Compression Framework for UHD Video Delivery
Huang, S., Xie, J. L., & Muslam, M. M. A. (2022) · IEEE Transactions on Cloud Computing · doi:10.1109/tcc.2022.3149420 · Cited by 17
“Rate-based ABR algorithms [7], [42] make the bitrate selection based on the predicted network bandwidth.”
MethodJoint optimization of QoE and fairness through network assisted adaptive mobile video streaming
Mehrabi, A., Siekkinen, M., & Ylä-Jääski, A. (2017) · IEEE International Conference on Wireless and Mobile Computing, Networking and Communications (WiMob) · doi:10.1109/wimob.2017.8115778 · Cited by 20
“Since purely buffer-based adaptation mechanisms may be sub-optimal especially under high throughput fluctuation, some techniques combine the buffer occupancy level with throughput prediction [11], [18], [19].”
BackgroundFastScan: Robust Low-Complexity Rate Adaptation Algorithm for Video Streaming over HTTP
Elgabli, A., & Aggarwal, V. (2019) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2019.2914388 · Cited by 17
“In recent studies, researchers have investigated various approaches for making streaming decisions, for example, by using control theory [7], [8], Markov Decision Process [9], machine learning [10], client buffer information [11], and data-driven techniques [12], [13].”
BackgroundGroupCast: Preference-Aware Cooperative Video Streaming With Scalable Video Coding
Elgabli, A., Felemban, M., & Aggarwal, V. (2018) · IEEE/ACM Transactions on Networking · doi:10.1109/infcomw.2018.8406852 · Cited by 17
Zhang, Y., Li, G., Xiong, C., Lei, Y., Huang, W., Han, Y., Walid, A., Yang, Y., & Zhang, Z. (2020) · doi:10.1145/3405672.3409489 · Cited by 17
“Such information exposure is helpful for adaptive bit rate (ABR) applications, such as DASH-based video streaming ABR algorithms [15, 22, 23] for video quality selection.”
MethodDeepmpc: A Mixture Abr Approach Via Deep Learning And Mpc
Huang, T., & Sun, L. (2020) · Proceedings - International Conference on Image Processing · doi:10.1109/icip40778.2020.9191198 · Cited by 17
“Recent work [9, 8] assumed that TCP connection throughput can be modeled as piece-wise stationary process , where the network status is defined as a tuple { µ, σ } .”
BackgroundGreenABR+: Generalized Energy-Aware Adaptive Bitrate Streaming
Turkkan, B. O., Dai, T., Raman, A., Kosar, T., Chen, C., Bulut, M. F., Żola, J., & Sow, D. (2024) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3649898 · Cited by 16
“For instance, CS2P [43] employs a Hidden-Markov Model to learn network characteristics from historical data, while Pen-sieve [30] uses a policy-gradient Reinforcement Learning method to optimize ABR decisions under various network conditions.”
On Uploading Behavior and Optimizations of a Mobile Live Streaming Service
Li, J., Li, Z., Wu, Q., & Gareth Tyson, A. (2022) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom48880.2022.9796824 · Cited by 16
Biases in Data-Driven Networking, and What to Do About Them
Bartulović, M., Jiang, J., Balakrishnan, S., Sekar, V., & Sinopoli, B. (2017) · ACM Workshop on Hot Topics in Networks (HotNets) · doi:10.1145/3152434.3152448 · Cited by 19
“• To compare multiple adaptive bitrate (ABR) algorithms under the same network conditions, video content providers often use traces of throughput observed by real clients to predict the quality if a new ABR algorithm were to run on the same clients [31, 37, 42].”
BackgroundSelf-citationSeer: Learning-Based $360^{\circ }$ Video Streaming for MEC-Equipped Cellular Networks
Kumar, S., Franklin, A. A., Jin, J., & Dong, Y. (2023) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2023.3257403 · Cited by 15
Fuzzy-Based Quality Adaptation Algorithm for Improving QoE from MPEG-DASH Video
Rahman, W. U., Hossain, M. D., & Huh, E. (2021) · Applied Sciences · doi:10.3390/app11115270 · Cited by 15
“The majority of existing algorithms employ fixed control rules to pick the video rate based on available throughput [4,5], the amount of video in the playback buffer [6], or a combination of the two [7–11].”
BackgroundMedia Production Using Cloud and Edge Computing: Recent Progress and NBMP-Based Implementation
Xu, Y., Yin, J., Yang, Q., & Yang, L. (2022) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2022.3140704 · Cited by 15
Elastically Reliable Video Transport Protocol Over Lossy Satellite Links
Li, X., & Wang, J. (2018) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2018.2832858 · Cited by 15
“The rules of the PNSR conversion are: 1: Bad ∈ [0, 20], 2: Poor ∈ [20, 25), 3: Fair ∈ [25, 31), 4: Good ∈ [31, 37), and 5: Excellent ∈ [37, +∞).”
MethodZhang, L., Wang, S., & Chang, R. N. (2018) · 2018 IEEE International Conference on Web Services (ICWS) · doi:10.1109/icws.2018.00025 · Cited by 14
“[5] focused on bitrate selection and adaption in video clients by using data-driven throughput prediction algorithms.”
MethodNaresh, M., Saxena, P., & Gupta, M. (2023) · International Wireless Communications and Mobile Computing Conference (IWCMC) · doi:10.1109/iwcmc58020.2023.10182379 · Cited by 14
“the following state-of-the-art DRL-based and non-DRL-based ABR algorithms: SAC-ABR [12], Pensieve [6], BB [3], RB [2], BOLA [4], and Robust-MPC [5].”
MethodMMVS: Enabling Robust Adaptive Video Streaming for Wildly Fluctuating and Heterogeneous Networks
Wang, S., Lin, J., & Dai, Y. (2024) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2024.3443609 · Cited by 14
Leveraging QoE Heterogenity for Large-Scale Livecaset Scheduling
Zhang, R.-X., Ma, M., Huang, T., Li, H., Liu, J., & Sun, L. (2020) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3394171.3413918 · Cited by 18
Harmonizing Energy Efficiency and QoE for Brightness Scaling-based Mobile Video Streaming
Qian, C., Liu, D., & Jiang, H. (2022) · International Workshop on Quality of Service · doi:10.1109/iwqos54832.2022.9812899 · Cited by 14
Restoring Application Traffic of Latency-Sensitive Networked Systems Using Adversarial Autoencoders
Sacco, A., Esposito, F., & Marchetto, G. (2022) · IEEE Transactions on Network and Service Management · doi:10.1109/tnsm.2022.3192305 · Cited by 14
“Because of their ability to capture important traffic statistical properties with a relatively small number of states, HMMs have gained popularity as the traffic model of choice in years states [21], [39].”
Utilizing Reinforcement Learning for Adaptive Sensor Data Sharing Over C-V2X Communications
Flowers, B., Ku, Y.-J., Baidya, S., & Dey, S. (2023) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2023.3322068 · Cited by 14
“Further, ABR algorithms typically use estimates of 1) throughput [28], [29], 2) buffer status [30], or 3) both [31], [32], which are not available in this work due to the blind broadcast nature of C-V2X Sidelinks and real time nature of the sensor data sharing task.”
BackgroundZhang, D., Zhu, H., Shen, K., Wang, D., & Wang, F. (2023) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2023.3339671 · Cited by 14
Understanding Video Streaming Algorithms in the Wild
Licciardello, M., Grüner, M., & Singla, A. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-44081-7_18 · Cited by 14
Karma: Adaptive Video Streaming via Causal Sequence Modeling
Xu, B., Chen, H., & Ma, Z. (2023) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3581783.3612177 · Cited by 13
“In this paper, we use the mean 𝜇 𝑡 and the variance 𝜎 𝑡 of throughput values to characterize the network condition at time 𝑡 by following prior works [2, 42].”
BackgroundMethodThroughput Prediction Using Recurrent Neural Network Model
Wei, B., Okano, M., Kanai, K., Kawakami, W., & Katto, J. (2018) · IEEE Global Conference on Consumer Electronics (GCCE) · doi:10.1109/gcce.2018.8574877 · Cited by 15
“We compare the prediction result with a traditional last sample (LS) method and HMM in [1].”
MethodResultDeeplive: QoE Optimization for Live Video Streaming through Deep Reinforcement Learning
Tian, Z., Zhao, L., Nie, L., Chen, P., & Chen, S. (2019) · IEEE International Conference on Parallel and Distributed Systems (ICPADS) · doi:10.1109/icpads47876.2019.00122 · Cited by 13
“Among them, FESTIVE [2], RBA [3], BBA [4], BOLA [5] and DYNAMIC [6] all have limitations.”
BackgroundMethodInfluentialQuality Adaptation and Resource Allocation for Scalable Video in D2D Communication Networks
Ullah, S., Kim, K. T., Manzoor, A., Khan, L. U., Kazmi, S. M. A., & Hong, C. S. (2020) · IEEE Access · doi:10.1109/access.2020.2978544 · Cited by 13
“To cope with such variations, many researches presented quality adaption proposals for the next video segment based on download buffer as well as the throughput prediction for the subsequent time frame [9], [11], [12], [16], [18].”
BackgroundPnP-DRL: A Plug-and-Play Deep Reinforcement Learning Approach for Experience-Driven Networking
Xu, Z., Wu, K., Zhang, W., Tang, J., Wang, Y., & Xue, G. (2021) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2021.3087270 · Cited by 13
Learning-based Fuzzy Bitrate Matching at the Edge for Adaptive Video Streaming
Shi, W., Li, Q., Wang, C., Zou, L., Shen, G., Zhang, P., & Jiang, Y. (2022) · International World Wide Web Conference (WWW) · doi:10.1145/3485447.3511944 · Cited by 13
“, [31]) that choose the most appropriate bitrate according to the predicted available bandwidth.”
BackgroundRaca, D., Zahran, A. H., Sreenan, C. J., Sinha, R. K., Halepovic, E., Jana, R., & Gopalakrishnan, V. (2017) · ACM Workshop on Hot Topics in Wireless (HotWireless) · doi:10.1145/3127882.3127892 · Cited by 13
“Other studies rely only on a device’s instantaneous radio channel quality indicator (CQI) and base its future decision on that estimate [9].”
Duque‐Torres, A., Amezquita-Suárez, F., Rendón, O. M. C., Ordóñez, A., & Muñoz, W. Y. C. (2019) · Applied Sciences · doi:10.3390/app9224808 · Cited by 15
“In the networking context, some possibilities arise from using ML: (i) forecast behaviour in the network [37,38], (ii) anomaly detection [39,40], (iii) traffic identification and flow classification [17,41]; and (iv) adaptive resource allocation [42,43].”
MethodMPTCP+: Enhancing Adaptive HTTP Video Streaming over Multipath
Zhao, J. M., Liu, J., Zhang, C., Cui, Y., Jiang, Y., & Gong, W. (2020) · International Workshop on Quality of Service · doi:10.1109/iwqos49365.2020.9213038 · Cited by 13
“While there have been many application-layer approaches that improve DASH ABR logic [5–7], recent work has started learning transport protocol’s throughput pattern during DASH video delivery [8].”
BackgroundChen, K., Wang, B., Wang, W., Li, X., & Ren, F. (2023) · International World Wide Web Conference (WWW) · doi:10.1145/3543507.3583364 · Cited by 12
“For example, CS2P uses the long-term average bandwidth to select bitrate and ignores the fuctuation of instantaneous bandwidth, which is not suitable for LLLS scenarios that require short-term bandwidth prediction.”
BackgroundMethodInfluentialReinforcement Learning-Based Adaptive Bitrate Caching at MEC Server
Lekharu, A., Chauhan, A. P. S., Sur, A., & Patra, M. (2024) · IEEE Transactions on Network and Service Management · doi:10.1109/tnsm.2024.3367333 · Cited by 12
“The performance of the proposed ABRCache model was compared against various state-of-the-art caching strategies such as QoECache [11], RLCache [20], AViC [22], PoPCache [12], AdaptSize [13], Rate Based (RB) [28], Buffer Based (BB) [29], LRU, and LFU.”
BackgroundMethodInfluentialJing, W., Liu, C., Cai, H., Wen, X., Lu, Z., Wang, Z., & Zhang, H. (2024) · IEEE Transactions on Network and Service Management · doi:10.1109/tnsm.2024.3377693 · Cited by 12
“As the ABR algorithm has a great impact on video streaming performance, especially users’ quality of experience (QoE), it has been investigated intensively [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22].”
BackgroundOnline Safety Assurance for Learning-Augmented Systems
Rotman, N. H., Schapira, M., & Tamar, A. (2020) · ACM Workshop on Hot Topics in Networks (HotNets) · doi:10.1145/3422604.3425940 · Cited by 17
SAC-ABR: Soft Actor-Critic based deep reinforcement learning for Adaptive BitRate streaming
Naresh, M., Gireesh, N., Saxena, P., & Gupta, M. (2022) · International Conference on Communication Systems & Networks (COMSNETS) · doi:10.1109/comsnets53615.2022.9668424 · Cited by 12
Sacco, A., Esposito, F., & Marchetto, G. (2023) · IEEE Transactions on Network and Service Management · doi:10.1109/tnsm.2023.3270166 · Cited by 12
“Our estimator consists of an HMM model that dictates the evolution over time of traffic values and an autoencoder used to improve the performance of the HMM.”
MethodEfficient Real-time Video Conferencing with Adaptive Frame Delivery
Liu, L., Li, J., Xu, H. L., Xue, K., & Xue, C. J. (2023) · Computer Networks · doi:10.1016/j.comnet.2023.109918 · Cited by 12
FedABR: A Personalized Federated Reinforcement Learning Approach for Adaptive Video Streaming
Xu, Y., Li, X., Yi, Y., Lin, Z., Wang, L., & Li, W. (2023) · IFIP Networking Conference (IFIP Networking) · doi:10.23919/ifipnetworking57963.2023.10186404 · Cited by 18
“The learning-based methods [11]– [13], [15], [16] were proposed to learn personalized ABR strategies for various conditions.”
MethodGiantClient: Video HotSpot for Multi-User Streaming
Elgabli, A., Felemban, M., & Aggarwal, V. (2018) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2018.2870715 · Cited by 12
“For example, using control theory [9], [10], Markov Decision Process [11], machine learning [12], client buffer information [13], and data-driven techniques [14]–[16].”
MethodSATE: Providing Stable and Agile Adaptation in HTTP-Based Video Streaming
Choi, W., & Yoon, J. (2019) · IEEE Access · doi:10.1109/access.2019.2901279 · Cited by 12
“proposed in [20] to address this problem, but it is still difficult to predict network bandwidth accurately.”
BackgroundVibra: neural adaptive streaming of VBR-encoded videos
Zhou, G., Wu, R., Hu, M., Zhou, Y., Fu, T. Z. J., & Wu, D. (2021) · International Workshop on Network and Operating Systems Support for Digital Audio and Video (NOSSDAV) · doi:10.1145/3458306.3460993 · Cited by 12
eff-HAS: Achieve higher efficiency in data and energy usage on dynamic adaptive streaming
Kim, S., Oh, H., & Kim, C.-K. (2018) · Journal of Communications and Networks · doi:10.1109/jcn.2018.000045 · Cited by 12
Improving Robustness of DASH Against Unpredictable Network Variations
Wang, B., Xu, M., Ren, F., & Wu, J. (2021) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2021.3050086 · Cited by 12
ANT: Learning Accurate Network Throughput for Better Adaptive Video StreamingPreprint
Yin, J., Xu, Y., Chen, H., Zhang, Y., Appleby, S., & Ma, Z. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2104.12507 · Cited by 12
Vivisecting Starlink Throughput: Measurement and Prediction
Liu, Z., Reidys, F.-X. G., Tanveer, S., & Vasisht, D. (2025) · Proceedings of the ACM on Networking · doi:10.1145/3768971 · Cited by 25
“[46, 70].”
Preference-Aware Dynamic Bitrate Adaptation for Mobile Short-Form Video Feed Streaming
Ran, D., Hong, H., Chen, Y., Ma, B., Zhang, Y., Zhao, P., & Bian, K. (2020) · IEEE Access · doi:10.1109/access.2020.3042619 · Cited by 11
“Beyond that, much efforts have been made on implementing collaborative information to facilitate video streaming, e.g., CS2P [12] which focuses on improving the wireless network bandwidth estimation for bitrate adaptation, SDNDASH [13] which introduces central control to coordinate the usage of the…”
BackgroundOpportunistic Transmission for Video Streaming over Wild Internet
Huang, J., Su, Q., Li, W., Liu, Z., Zhang, T., Liu, S., Zhong, P., Jiang, W., & Wang, J. (2022) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3488722 · Cited by 11
“CS2P [7] uses a data-driven method to analyze the characteristics of the video session window, and establishes the hidden Markov data model to predict future throughput, so as to help the client choose the appropriate bitrate.”
MethodPreSR: Neural-Enhanced Adaptive Streaming of VBR-Encoded Videos With Selective Prefetching
Zhou, G., Luo, Z., Hu, M., & Wu, D. (2022) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2022.3227419 · Cited by 11
“(2) Rate-based algorithms (e.g., FESTIVE [14], CS2P [15]), which adapt the bitrate according to the prediction of network throughput.”
MethodSong, T., Garza, P., Meo, M., & Munafò, M. M. (2024) · Computer Networks · doi:10.1016/j.comnet.2024.110507 · Cited by 13
“The work of [66] aimed to improve the selection of bitrate by predicting throughput based on data-driven approach.”
BackgroundIntelligent Video Ingestion for Real-time Traffic Monitoring
Zhang, X., Zhao, Y., Min, G., Miao, W., Huang, H., & Ma, Z. (2022) · ACM Transactions on Sensor Networks · doi:10.1145/3529511 · Cited by 11
“BBA [15] decides the bitrate of video chunks based on the bufer occupancy information while [29] made decisions according to the estimated available bandwidth.”
Enhancing Low Latency Adaptive Live Streaming Through Precise Bandwidth Prediction
Wang, B., Su, M., Wang, W., Chen, K., Liu, B., Ren, F., Xu, M., Liu, J., & Wu, J. (2024) · IEEE/ACM Transactions on Networking · doi:10.1109/tnet.2024.3426607 · Cited by 10
“Although network state evolutions exhibit complexity, they have cross-session similarity: video sessions with similar features share similar dynamic patterns of network state [8], [9], [10].”
InfluentialPractically High Performant Neural Adaptive Video Streaming
Patel, S., Zhang, J.-Y., Narodystka, N., & Jyothi, S. A. (2024) · PACMNET · doi:10.1145/3696401 · Cited by 10
A Distributed Approach for Bitrate Selection in HTTP Adaptive Streaming
Bentaleb, A., Begen, A. C., Harous, S., & Zimmermann, R. (2018) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3240508.3240589 · Cited by 10
“Our design also needs to obtain accurate available bandwidth estimations, and we use two algorithms, namely PANDA [18] and CS2P [26] for this purpose.”
MethodNancy: Neural Adaptive Network Coding Methodology for Video Distribution Over Wireless Networks
Saxena, P., Naresh, M., Gupta, M., Achanta, A., Kota, S. L., & Gupta, S. (2020) · Globecom · doi:10.1109/globecom42002.2020.9322332 · Cited by 10
“In order to cope up with varying wireless network conditions, one of the traditional approaches to improve video quality is to make use of adaptive bit rate algorithms [3], [4].”
Background
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.