Network and user centric performance analysis of scheduling strategies for video streaming over LTE

Diverse scheduling strategies have been designed for video streaming traffic ranging from Quality of Service (QoS) aware scheduling rules to more complex video quality based scheduling strategies. In this work, we analyze and compare some of the well known scheduling rules for video streaming traffic. Our main goal is to compare and analyze different classes of scheduling strategies (QoS and video quality aware rules) in terms of network centric performance metrics as well as user centric metrics. QoS evaluation involves the evaluation of network performance parameters, e.g., packet loss rate, average system throughput and end-to-end packet delay. On the other hand, video quality evaluation involves the computation of objective and subjective video quality metrics. According to simulation results, the proxy based video quality aware scheduling strategy performs best in terms of number of satisfied users and should be used in an Long-Term Evolution (LTE) downlink to offer high quality video streaming services.