Joint user association and power allocation for minimizing multi-bitrate video transmission delay in mobile-edge computing networks

Abstract

Fast-growing video services place higher demands on network performance especially in terms of latency, but the traditional networks architecture with congested backhaul link can no longer meet the requirement. Recently, mobile edge computing (MEC) has become a promising paradigm to achieve low latency performance and can provide multi-bitrate video streaming at the edge of radio access networks (RAN) with the ability of caching and transcoding. In this paper, we consider the scenario of multi-cell MEC networks, where each BS deployed with one MEC server is connected to the core network through the limited-capacity backhaul link. Our goal is to minimize the system delay which includes backhaul transmission delay and wireless side transmission delay. To this end, we propose a collaborative optimization of user-BS association and power allocation strategy with the given cache status. This is a mixed-integer nonlinear programming (MINLP) problem which is NP-hard. Thus we propose an improved genetic algorithm to solve this problem based on the traditional genetic algorithm. Simulation results demonstrate that our proposed algorithm performs better in terms of convergence and can get better solution as compared with traditional genetic algorithm.