Trace-Driven Optimization on Bitrate Adaptation for Mobile Video Streaming

Qiao, Chunyu; Li, Gen; Ma, Qiang; Wang, Jiliang

doi:10.1109/tmc.2020.3036707

Cited by 13 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…4. Among four ABR optimization algorithms, namely Rate-based (RB) [4], Buffer-based (BB) [5], Model Predictive Control (MPC) [10], and RL [6], we observe a explicit increasing trend in video stall rate by 47.6% on average for RL when running over 5G. In most cases, ABR algorithms under 5G achieve almost similar normalized bitrates in 4G settings at only 2.8% degradation, as shown in Fig.…”

Section: B Adaptive Video Streamingmentioning

confidence: 79%

“…ABR algorithms are widely used to select bitrates adaptively according to network conditions. There are two traditional optimization categories: (i) parametric methods that leverage the video information and throughput estimation to make bitrate decisions, such as rate-based algorithms [4], [8], buffer-based algorithms [5], [7], [40], and control theoretic approaches [2], [9], [10]. However, their reliance on accurate throughput estimations limits their performance enhancement.…”

Section: E Serverless Pipelines Performancementioning

confidence: 99%

“…In order to prevent significant losses in revenue caused by QoE degradation, video service providers have to optimize critical components spanning from the cloud servers to the clients, such as bitrate adaptation strategy [4], [5], buffer management [6], [7], and QoE enhancement scheme [8], [9]. In particular, most existing ABR algorithms for bitrate adaptation try to balance the tradeoff between video bitrate and network bandwidth under fluctuating network conditions [10], but only the size and download speed of video chunks are considered. Such coarse-grained bitrate adaptation cannot adapt efficiently to dynamic network environments [11], which may incur frequent video bitrate switching and break playback smoothness.…”

Section: Introductionmentioning

confidence: 99%

“…• NEIVA[10]: It improves network throughput prediction accuracy by training the predictor with the Hidden Markov Model (HMM), then incorporates Model Predictive Control (MPC) for mobile video streaming.• QFlow[37]: It develops a model-free Double Deep Q Network (DDQN) that enables bitrate decision adaptation to video feedback control loop to maximize QoE.…”

mentioning

confidence: 99%

See 3 more Smart Citations

EAVS: Edge-assisted Adaptive Video Streaming with Fine-grained Serverless Pipelines

Hou,

Yang,

Kuipers

et al. 2023

IEEE INFOCOM 2023 - IEEE Conference on Computer Communications

View full text Add to dashboard Cite

Recent years have witnessed video streaming gradually evolve into one of the most popular Internet applications. With the rapidly growing personalized demand for realtime video streaming services, maximizing their Quality of Experience (QoE) is a long-standing challenge. The emergence of the serverless computing paradigm has potential to meet this challenge through its fine-grained management and highly parallel computing structures. However, it is still ambiguous how to implement and configure serverless components to optimize video streaming services. In this paper, we propose EAVS, an Edge-assisted Adaptive Video streaming system with Serverless pipelines, which facilitates fine-grained management for multiple concurrent video transmission pipelines. Then, we design a chunk-level optimization scheme to solve the video bitrate adaptation issue. We propose a Deep Reinforcement Learning (DRL) algorithm based on Proximal Policy Optimization (PPO) with a trinal-clip mechanism to make bitrate decisions efficiently for better user-perceived QoE. Finally, we implement the serverless video streaming system prototype and evaluate the performance of EAVS on various real-world network traces. Extensive results show that our proposed EAVS significantly improves userperceived QoE and reduces the video stall rate, achieving over 9.1% QoE improvement and 60.2% latency reduction compared to state-of-the-art solutions.

show abstract

Section: B Adaptive Video Streamingmentioning

confidence: 79%