Rate adaptation scheme for HTTP-based streaming to achieve fairness with competing TCP traffic

HTTP Adaptive Streaming (HAS) is an adaptive bitrate streaming technique which is able to adapt to the network conditions using conventional HTTP web servers. An HAS player requests pre-encoded video chunks by sending an HTTP GET message. When the playout buffer of the HAS player exceeds a pre-defined threshold, the player sends HTTP GET messages periodically to the HTTP web server and switches to the steady state. The HAS player generates an ON/OFF traffic pattern during the steady state. We show that the HAS player estimates the network bandwidth incorrectly when the OFF duration is too long. Moreover, an HAS flow cannot fairly share the bottleneck with TCP flows due to the OFF period and TCP dynamics. In this paper, we propose a client-side rate adaptation scheme that eliminates OFF periods through two buffer management states in order to fairly share the bottleneck. We evaluate the performance of the HAS player implemented in the ns-3 network simulator. Our simulation results show that the proposed scheme significantly improves fairness by 45% without the rebuffering problem compared to the conventional HAS player.