Abstract
Loki is a state-of-the-art adaptive bitrate algorithm for the transmission of real-time-communication (RTC) video. It fuses traditional heuristic methods with a learning-based model to maximize the quality of experience (QoE) under diverse network conditions. However, a recurring rebound pattern is observed in Loki’s decision-making process where the decision frequently oscillates between the two boundaries of the action space, making Loki fail to adapt to the fluctuating network bandwidth. To address this issue, we propose Loki+, which improves both the fusion mechanism and the design of the learning-based actor. Specifically, we replace the element-wise multiplication with a simple but effective trend fusion and further optimize the design of reward and loss functions for training Loki+. Extensive simulation results show that Loki+ significantly improves the QoE in the aspects of reducing the stall rate by 20%\(\sim\)60% and the frame delay by 3.5%\(\sim\)30.5% while maintaining a similar sending bitrate or video quality, compared with Loki.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00530-023-01197-5/MediaObjects/530_2023_1197_Fig1_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00530-023-01197-5/MediaObjects/530_2023_1197_Fig2_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00530-023-01197-5/MediaObjects/530_2023_1197_Figa_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00530-023-01197-5/MediaObjects/530_2023_1197_Fig3_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00530-023-01197-5/MediaObjects/530_2023_1197_Fig4_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00530-023-01197-5/MediaObjects/530_2023_1197_Fig5_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00530-023-01197-5/MediaObjects/530_2023_1197_Fig6_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00530-023-01197-5/MediaObjects/530_2023_1197_Fig7_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00530-023-01197-5/MediaObjects/530_2023_1197_Fig8_HTML.png)
Similar content being viewed by others
Data availability
The data that support the findings of this study are available on request from the corresponding author HC, upon reasonable request.
Code availability
Not applicable.
References
Trade, U.N.C., Development: Estimates of global e-commerce 2019 and preliminary assessment of COVID-19 impact on online retail 2020. https://news.un.org/zh/story/2021/05/1083402 (2021)
Information, C.A., Technology, C.: Real-time interactive industry development research report. http://www.caict.ac.cn/kxyj/qwfb/ztbg/202206/t20220614_ 404308.htm (2022)
Carlucci, G., De Cicco, L., Holmer, S., Mascolo, S.: Congestion control for web real-time communication. IEEE/ACM Trans. Netw. 25(5), 2629–2642 (2017). https://doi.org/10.1109/TNET.2017.2703615
Zhang, H., Zhou, A., Hu, Y., Li, C., Wang, G., Zhang, X., Ma, H., Wu, L., Chen, A., Wu, C.: Loki: Improving long tail performance of learning-based real-time video adaptation by fusing rule-based models. Association for Computing Machinery, New York, NY, USA (2021). https://doi.org/10.1145/3447993.3483259
Mao, H., Chen, S., Dimmery, D., Singh, S., Blaisdell, D., Tian, Y., Alizadeh, M., Bakshy, E.: Real-world Video Adaptation with Reinforcement Learning (2020)
Zhou, A., Zhang, H., Su, G., Wu, L., Ma, R., Meng, Z., Zhang, X., **e, X., Ma, H., Chen, X.: Learning to coordinate video codec with transport protocol for mobile video telephony. Association for Computing Machinery, New York, NY, USA (2019)
Schulman, J., Wolski, F., Dhariwal, P., Radford, A., Klimov, O.: Proximal Policy Optimization Algorithms (2017)
Zhang, H., Zhou, A., Ma, H.: Improving mobile interactive video qoe via two-level online cooperative learning. IEEE Trans. Mobile Comput. 22(10), 5900–5917 (2023). https://doi.org/10.1109/TMC.2022.3179782
Kingma, D.P., Ba, J.: Adam: A method for stochastic optimization. CoRR ar**v:abs/1412.6980 (2014)
Pereyra, G., Tucker, G., Chorowski, J., Kaiser, L., Hinton, G.: Regularizing Neural Networks by Penalizing Confident Output Distributions (2017). https://openreview.net/forum?id=HkCjNI5ex
Sutton, R.S., Barto, A.G.: Reinforcement Learning: An Introduction. A Bradford Book, Cambridge, MA, USA (2018)
Schulman, J., Moritz, P., Levine, S., Jordan, M.I., Abbeel, P.: High-dimensional continuous control using generalized advantage estimation. CoRR ar**v:abs/1506.02438 (2015)
Commission, F.C.: Raw Data-Measuring Broad-band America 2016. https://www.fcc.gov/reports-research/reports/measuring-broadband-america/raw-data-measuring-broadband-ameri-2016 (2016)
Riiser, H., Vigmostad, P., Griwodz, C., Halvorsen, P.: Commute path bandwidth traces from 3g networks: Analysis and applications. Association for Computing Machinery, New York, NY, USA (2013)
Yi, G., Yang, D., Bentaleb, A., Li, W., Li, Y., Zheng, K., Liu, J., Ooi, W.T., Cui, Y.: The acm multimedia 2019 live video streaming grand challenge. Association for Computing Machinery, New York, NY, USA (2019)
Akhtar, Z., Nam, Y.S., Govindan, R., Rao, S., Chen, J., Katz-Bassett, E., Ribeiro, B., Zhan, J., Zhang, H.: Oboe: Auto-tuning video abr algorithms to network conditions. Association for Computing Machinery, New York, NY, USA (2018)
Acknowledgements
This work was partially supported by the National Natural Science Foundation of China (62101241), Jiangsu Provincial Double-Innovation Doctor Program (JSSCBS20210001), and Changzhou Power Supply Branch of State Grid Jiangsu Electric Power Co., Ltd.(SGJSCZ00KJJS2311209).
Funding
Research grants from the National Natural Science Foundation of China (62101241), Jiangsu Provincial Double-Innovation Doctor Program (JSSCBS20210001), and Changzhou Power Supply Branch of State Grid Jiangsu Electric Power Co., Ltd. ( SGJSCZ00KJJS2311209).
Author information
Authors and Affiliations
Contributions
Material preparation, data collection, and analysis were performed by WZ, WS, and KY. Conceptualization and methodology were performed by WS and HC. The first draft of the manuscript was written by WS and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Conflict of interest
The authors have no relevant financial or non-financial interests to disclose.
Ethical approval
Not applicable.
Consent to participate
All authors agreed to participate.
Consent for publication
All authors gave permission for publication.
Additional information
Communicated by B. Bao.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Zhu, W., Su, W., Yang, K. et al. Improving the application performance of Loki via algorithm optimization. Multimedia Systems 30, 2 (2024). https://doi.org/10.1007/s00530-023-01197-5
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s00530-023-01197-5