May 1, 2019 –
Title: Adaptive Bitrate Streaming Over Cellular Networks: Rate Adaptation and Data Savings Strategies
Student: Yanyuan Qin
Major Advisor: Dr. Bing Wang
Associate Advisors: Dr. Song Han, Dr. Krishna R. Pattipati, Dr. Subhabrata Sen (AT&T Labs - Research)
Reviewers: Dr. Mohammad Maifi Hasan Khan, Dr. Fei Miao
Date/Time: May 1st 11am-12:30pm
Location: Homer Babbidge Library – 1947 Conference Room
Adaptive bitrate streaming (ABR) is the de facto technology in industry for dynamically adapting the video streaming quality based on varying network conditions. In this dissertation, we address two salient challenges in ABR streaming over cellular networks. The first challenge is streaming Variable Bitrate (VBR) encoded videos, which have been increasingly adopted by content providers due to their significantly higher encoding efficiency compared to traditional Constant Bitrate (CBR) encodings. However, VBR introduces new challenges for ABR streaming, whose nature and implications are little understood. The second challenge is the large amount of data required for video streaming. As an example, streaming just one-hour High Definition (HD) video on mobile Netflix can consume 3 GB data, higher than average data plan for a U.S. cellular customer. While video and network providers offer data saving options, the existing practices are suboptimal: they often lead to highly variable video quality and do not make the most effective use of the network bandwidth.
To tackle the first challenge, we identify distinguishing characteristics of VBR encodings that impact user Quality of Experience (QoE) and should be factored in any ABR adaptation decision. We develop novel best practice design principles to guide ABR rate adaptation for VBR encodings. As a proof of concept, we design a novel and practical control-theoretic rate adaptation scheme, CAVA (Control-theoretic Adaption for VBR-based ABR streaming), incorporating these concepts. Extensive evaluations show that CAVA substantially outperforms existing state-of-the-art adaptation techniques.
To address the second challenge, we analyze the underlying causes for suboptimal existing data saving practices and propose two novel approaches to achieve better tradeoffs between video quality and data usage. The first approach is Chunk-Based Filtering (CBF), which can be retrofitted to any existing ABR scheme. The second approach is QUality-Aware Data-efficient streaming (QUAD), a holistic rate adaptation algorithm that is designed ground up. Our evaluations demonstrate that compared to the state of the art, the two proposed schemes achieve consistent video quality that is much closer to the user-specified target, lead to far more efficient data usage, and incur lower stalls. As ongoing research, we are in the process of developing data saving strategies that directly take account of a user’s data budget in ABR decisions.