Energy-efficient User-centric Dynamic Adaptive Multimedia Streaming in 5G Cellular Networks

5G multimedia mobile wireless network is designed to support on-demand encoding of rich mobile multimedia content for heterogeneous users. Due to the heterogeneity of the users, adaptive multimedia services are essential to provide a satisfactory Quality of Experience (QoE). In this paper, we propose a utility-based dynamic adaptive multimedia streaming scheme, named UDAS, for heterogeneous users that helps in extending the battery life of the low-battery users and also uses the bandwidth of the wireless channel efficiently. At each scheduling interval, the adaptation algorithm considers four utility functions of the user devices, namely, quality utility, power consumption utility, packet error ratio utility, and remaining battery utility to adapt the data rate dynamically. We formulate an optimization problem to maximize a joint utility function of these four utilities. The solution of the problem provides the best adaptive multimedia content that is selected for transmission to the end-users in every scheduling interval. The mobile edge computing (MEC) server situated at the base station performs an on-demand HEVC (high efficiency video coding) encoding of videos and select the best suitable videos for different users. Simulation results verify the improved performance of UDAS in terms of saved battery energy and the number of unserved low-battery users in comparison with state-of-the-art non-adaptive multimedia streaming schemes and a popular scheme ESDOAS from the literature.