ABSTRACT

A lot of research has been done in the area of TCP long-lived (TLL) and short-lived flows and well as UDP but there is limited research done on how they can affect adaptive video streaming at a single household bottleneck router. The goal of Adaptive streaming is to deliver videos to the user in the most efficient way possible and in the best quality possible. Several factors can affect the quality of videos including internet connection and other applications using the same router. It is important to note which applications can negatively affect streaming and cause issues such as poor-quality videos and buffering. This paper investigates the impact that various applications that use TCP long-lived flows have on streaming and its effect is compared to applications that use UDP flows. The aim is to determine which type of flow affects adaptive streaming the most. This investigation is carried out with the use of the TAPAS (a Tool for rApid Prototyping of Adaptive Streaming) player. Live experiments were conducted at a household bottleneck link with a client player, a TCP or UDP application. We show that UDP flows disrupt adaptive video streaming to a greater extent when compared to TCP TLL flows

Keywords: - Adaptive streaming, bottleneck, live experiments, metrics, TAPAS, TCP long-lived flows, TCP shortlived flows, UDP flows