The ubiquity of high quality video and proliferation of mobile devices has contributed to an unprecedented rise in video consumption. HTTP, in conjunction with adaptive streaming, has become the de facto mechanism for delivering the vast majority of video as it readily caters to heterogeneous networks and devices. This dissertation presents adaptive streaming methods using the Flexible Dual Streaming TCP-UDP Protocol (FDSP) as well as queueing theory within an Markov Decision Process (qMDP).
FDSP is an improved alternative to traditional HTTP/TCP-based streaming that combines TCP's reliability with the low latency provided by UDP, resulting in an improved quality of experience (QoE) for the viewer. On the other hand, qMDP builds on HTTP/TCP-based streaming via a queueing-theory-based state space for applying reinforcement learning towards improved QoE.