A Control-Theoretic Approach for Dynamic Adaptive Video Streaming over HTTP

By Xiaoqi Yin, Abhishek Jindal, Vyas Sekar, Bruno Sinopoli. Computer communication review, 2015

1,183 citations recorded by the source corpus.

User-perceived quality-of-experience (QoE) is critical in Internet video applications as it impacts revenues for content providers and delivery systems. Given that there is little support in the network for optimizing such measures, bottlenecks could occur anywhere in the delivery system. Consequently, a robust bitrate adaptation algorithm in client-side players is critical to ensure good user experience. Previous studies have shown key limitations of state-of-art commercial solutions and proposed a range of heuristic fixes. Despite the emergence of several proposals, there is still a distinct lack of consensus on: (1) How best to design this client-side bitrate adaptation logic (e.g., use rate estimates vs. buffer occupancy); (2) How well specific classes of approaches will perform under diverse operating regimes (e.g., high throughput variability); or (3) How do they actually balance different QoE objectives (e.g., startup delay vs. rebuffering). To this end, this paper makes three key technical contributions. First, to bring some rigor to this space, we develop a principled control-theoretic model to reason about a broad spectrum of strategies. Second, we propose a novel model predictive control algorithm that can optimally combine throughput and buffer occupancy information to outperform traditional approaches. Third, we present a practical implementation in a reference video player to validate our approach using realistic trace-driven emulations.

Subfields: Computer Networks and Communications, Computer Vision and Pattern Recognition, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/ce5ce72111b0950b3b5fdec0fd9d4a8c06544b02

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Computer Vision and Pattern Recognition

In Electrical and Electronic Engineering

Citations in the corpus

Resource Management with Deep Reinforcement Learning (2016, background): “Examples include job scheduling in compute clusters [17], bitrate adaptation in video streaming [23, 39], relay selection in Internet telephony [40], virtual machine placement in cloud computing [20, 6], congestion control [38, 37, 13], and so on.”
Applications of Deep Reinforcement Learning in Communications and Networking: A Survey (2018, methodology): “The simulation results based on the mobile dataset from [68] show that the proposed DQL can improve the average QoE up to 25% compared with the bitrate control scheme [69].”
DeepDecision: A Mobile Deep Learning Framework for Edge Video Analytics (2018, background): “We assume there is a throughput predictor that can estimate the future bandwidth and latency over a short period of time [13].”
Learning in situ: a randomized experiment in video streaming (2019, methodology): “Fugu is based on MPC (model predictive control) [46], a classical control policy, but replaces its throughput predictor with a deep neural network trained using supervised learning on data recorded in situ (in place), meaning from Fugu’s actual deployment environment, Puffer.”
Oboe: auto-tuning video ABR algorithms to network conditions (2018, methodology): “Despite widespread deployment, ABR algorithms continue to be an active area of research [32, 34, 39, 48, 51, 59].”
A Survey on Bitrate Adaptation Schemes for Streaming Media Over HTTP (2019, result): “The proposed system significantly improves the bitrate decision of clientbased adaptation schemes like BOLA [67] and FastMPC [81], and it offers a 24% on average better viewer QoE compared to Pensieve [102].”
Review on model predictive control: an engineering perspective (2021, background): “From balancing walking robots [134], hanging crane loads [110], and cruise control for heavy duty trucks [62, 140], to optimizing buffering and quality in video streaming [138].”
Optimizing 360 video delivery over cellular networks (2016, methodology): “Extensive research has been conducted on improving the QoE of DASH video [18, 21, 13, 23, 17].”
A variegated look at 5G in the wild: performance, power, and QoE implications (2021, methodology): “For the following sections, we focus on mmWave 5G which are considered key to mainstream 5G and have not been studied before in the context of mobile applications. has proposed a plethora of ABR algorithms for video streaming in recent years [33, 38, 61, 62].”
BOLA: Near-optimal bitrate adaptation for online videos (2016, methodology): “When the user plays a video, the video player can download each chunk at a bitrate that is appropriate for the available bandwidth of the network connection.”
Neural Adaptive Video Streaming with Pensieve (2017, methodology): “(4) MPC [50]: uses buffer occupancy observations and throughput predictions (computed in the same way as RB) to select the”
CS2P: Improving Video Bitrate Selection and Adaptation with Data-Driven Throughput Prediction (2016, methodology): “…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.”
Flare: Practical Viewport-Adaptive 360-Degree Video Streaming for Mobile Devices (2018, methodology): “In non-360° video streaming, numerous rate adaptation schemes have been proposed [34, 36, 44, 59].”