Adaptive offloading in mobile-edge computing for ultra-dense cellular networks based on genetic algorithm

Abstract With the combination of Mobile Edge Computing (MEC) and next generation cellular networks, computation requests from end devices can be offloaded promptly accurately by edge servers equipped on Base Stations (BSs). However, due to densified heterogeneous deployment BSs, device may covered more than one BS, which brings new challenges for offloading decision, that is whether where offload computing tasks low latency energy cost. This paper formulates a multi-user-to-multi-servers (MUMS) problem in ultra-dense networks. The MUMS divided conquered two phases, are server selection decision. For mobile users grouped BS considering both physical distance workload. After grouping, original into parallel multi-user-to-one-server decision subproblems. To get fast near-optimal solutions these subproblems, distributed strategy based binary-coded genetic algorithm designed an adaptive Convergence analysis given extensive simulations show proposed significantly reduces average consumption devices. Compared with state-of-the-art researches, our delay 56% total 14%