Joint offloading decision and resource allocation for mobile edge computing enabled networks

Mobile edge computing (MEC) based solutions are essential and of great significance for a wide range of promising 5G wireless big data services such as remote healthcare systems and AR/VR games. Present research in this area focuses on the downlink resource allocation scenarios from MEC servers to user equipments (UEs). This paper considers a mull-user MEC-enabled wireless communication system, where UEs suffer limited communication and computation resources. To achieve higher energy efficiency and the better experience for UEs, we aim to maximize the number of offloaded tasks for all UEs in uplink communication while maintaining the computation resources of MEC at an acceptable level. The formulated problem is an NP-hard mixed-integer nonlinear programming problem and it is a challenge to solve it efficiently. As such, an efficient low-complexity heuristic algorithm is proposed, which provides a near-optimal solution with a low time cost. The results show that the proposed scheme achieves the higher number of successful offloaded tasks than the existing centralized resource allocation algorithm (CRAA) and centralized decision and resource allocation algorithm that UEs with the largest saved energy consumption accepted first (CAR-E) under different scenarios. Moreover, the relationship between the optimal transmission power and the computation resource of MEC is investigated. The results obtained in this paper can be extended to design a novel framework of communication, computation and smart coded caching MEC networks.