Effect of Delivery Latency, Feedback Frequency and Network Load on Adaptive Multimedia Streaming

As video on demand systems gain popularity, it seems likely that the desire to serve a high number of customers from limited network resources could lead to a degradation of the end- users´ perceived quality. Quality-oriented adaptation scheme (QOAS) balances the need for high quality with increased network utilization when streaming multimedia. QOAS requires client-side monitoring of some transmission-related parameters, grading of the end-user´s quality and feedback that informs the server about the received quality. In response to this feedback, the server adjusts the streaming process in order to maximize the end-user perceived quality in the current conditions. This paper studies the effect of delivery latency and feedback frequency on quality-oriented adaptive multimedia streaming. It also shows how high end-user perceived quality is maintained in the presence of different types of background traffic while recording a significant increase in link utilization and a very low loss rate.