PhD Defence | Modeling and Optimization of Network Assisted Video Streaming
Jan Willem Kleinrouweler’s thesis explores how collaboration between video players and network-based control elements can improve the video quality and fairness in shared networks. Kleinrouweler completed his research under the supervision of Dick Bulterman, Rob van der Mei and Pablo Cesar (all from CWI).
Computer networks are used by multiple users and devices for which the bandwidth has to be shared. Video streaming applications require a relatively large and stable share of the bandwidth to be able to deliver high quality video. The architecture and protocols used by the current dominant streaming technology, HTTP Adaptive Streaming (HAS), have enabled the massive scale of video content delivery as we know it today. Yet the protocols are not suitable for meeting the demanding bandwidth requirements in shared networks. Failure to deliver high quality video streams reflects badly on content providers and network operators, who risk the loss of revenue due to unsatisfied users. Video delivery problems distract the user, cause annoyance, and may lead to users abandoning their streams.
In his research, Kleinrouweler explores how the addition of a network-based control element that guides HAS clients improves the video quality and fairness between clients in shared networks. To this end, he developed, analyzed, and optimized mechanisms and policies for bandwidth sharing between HAS clients. By combining empirical and theoretical research methodologies, a broad understanding of network optimizations for video streaming was obtained. Using real implementations and experiments in different networks allowed to evaluate and improve the mechanisms for interaction between HAS clients and the network. In addition, Markov modeling was used to predict the impact of sharing policies on the overall streaming performance and optimize streaming efficiency in mobile networks.