Large-Scale User-Assisted Multi-Task Online Offloading for Latency Reduction in D2D-Enabled Heterogeneous Networks
Mengying Sun, Xiaodong Xu, Xiaofeng Tao, Ping Zhang
Abstract
Currently, the computing capability of smart mobile devices has been extremely improved. Exploiting computing resources of mobile devices to assist the network through offloading computation tasks will promisingly boost the fifth-generation (5G) and beyond heterogeneous networks. We investigate the user-assisted multi-task offloading scheme based on the mobile edge computing (MEC)-Cloud architecture to reduce the end-to-end computing latency. The offloading strategy, computing resource allocation and spectrum allocation are jointly optimized to minimize the computation latency while guaranteeing the energy available to the users. The formulated optimizing problem is a large-scale mixed-integer nonlinear optimizing problem which is hard to solve within a rational time. To overcome this problem, a low-complexity distributed framework based on the alternating direction method of multipliers algorithm is proposed to minimize the computing latency for all tasks. Compared with the existing schemes, the proposed scheme can reduce the computing latency and improve the performance efficiently. Simulation results illustrate the effectiveness of the proposed scheme in respect of latency reduction with different parameters.