On Mobile Video Streaming with Statistical Performance Guarantee

• Other Authors: Liu, Yan (author.)
• Format: Others
• Language: English; Chinese
• Published: 2016
• Online Access: http://repository.lib.cuhk.edu.hk/en/item/cuhk-1292405

在3G/4G移動網絡部署的推動下，移動視頻流已經成為移動互聯網的主要應用之一。由於移動通信網絡帶寬的內在波動，業內人士以及研究人員已經開發出許多自適應流算法以應對這樣的波動來提高流性能。目前的流算法通常都是被設計和調整以期望在所有的網絡中都可以工作的很好。雖然考慮到廣泛差異的網絡類型和特性，從覆蓋不良的3G網絡到可以提供100+Mbps帶寬的LTE網絡，目前的流算法在不同網絡和系統條件下會表現出不同的性能也不足為奇，但是這種缺乏可預測性和一致性的性能表現給服務提供商帶來了極大挑戰，它使得為用戶提供始終如一體驗的視頻流服務就算是可能的，也是非常困難的。 === 本文打破了傳統的一刀切的自適應流算法設計方法，開發了一個名為後視頻流速率分析（PSRA）的新框架。該框架通過分析過去視頻流會話吞吐量跟蹤數據 – 即視頻流會話的副產品，建立統計模型來對未來的流會話進行控制，使得視頻流算法可以根據底層網絡和系統配置進行調整。相比現有的視頻流算法，PSRA可以在寬範圍的網絡和系統配置下實現可預測的，一致的，可控的流性能。此外，PSRA也為服務供應商提供了一個能夠精確地控制視頻質量和視頻流性能之間平衡的新工具。大量跟蹤數據驅動的仿真實驗以及一個基於原型機實現的實驗結果驗證了PSRA在真實世界移動網絡和系統配置下的性能。這種視頻流算法設計方式開闢了流媒體算法設計和優化的新途徑並進一步由後視頻流質量分析（PSQA）框架的提出而驗證。PSQA是PSRA的一般化，它不僅可以用於優化目前流算法的性能而且還為探索新的流算法和用戶體驗度量開辟了新的道路。基於真實網絡跟蹤數據的仿真結果表明，PSQA不僅可以優化現有的和新的數據流算法而且可以達到非常接近事後採用窮舉法達到的最優用戶體驗。 === Fueled by the deployment of 3G/4G mobile network, mobile video streaming has become one of the main applications in mobile Internet. Due to mobile networks’ inherent bandwidth fluctuations, the industry as well as researchers have developed various new streaming algorithms to tackle such fluctuations to improve streaming performance. Common to all existing streaming algorithms they are designed and tuned to work across all networks. Given the wide range of networks, ranging from 3G networks under poor coverage to LTE networks offering 100Mbps-plus bandwidth, it is not surprising that existing algorithms can and do perform differently under different networks and system conditions. This lack of performance predictability and consistency poses significant challenges to service providers, rendering it difficult, if not impossible, to provision streaming services with a consistent user experience. === This thesis breaks away from the conventional one-size-fits-all approach to designing adaptive streaming by developing a new framework called Post-Streaming Rate Analysis (PSRA) where past throughput trace data - captured as a by-product of streaming sessions, are analyzed to construct a statistical model for the control of future streaming sessions such that the streaming algorithm can be tuned to the underlying network and system configurations Compared to existing streaming algorithms, PSRA can achieve predictable, consistent, and controllable streaming performance across a wide-range of networks and system configurations. Moreover, PSRA offers service provider a new tool to precisely control the tradeoff between video quality and streaming performance. Results from extensive trace-driven simulations as well as experimental results from a prototype implementation verified PSRA’s performance under real-world mobile network and system configurations. This thesis opens up a new way to the design and optimization of streaming algorithms. This is further demonstrated by a generalization of the PSRA framework into a Post-Streaming Quality Analysis (PSQA) framework which not only can be applied to optimize the performance of existing streaming algorithms, but also opens a new way for the exploration of new adaptive video streaming protocols and QoE metrics. Simulation results based on real network throughput traces show that PSQA can optimize existing and new streaming algorithms to achieve QoE that is remarkably close to the optimal achieved using brute-force method ex post facto. === Liu, Yan. === Thesis Ph.D. Chinese University of Hong Kong 2016. === Includes bibliographical references (leaves ). === Abstracts also in Chinese. === Title from PDF title page (viewed on …). === Detailed summary in vernacular field only. === Detailed summary in vernacular field only.