Joint optimization strategy of heterogeneous resources in multi-MEC-server vehicular network

As an important application scenario of Internet of Things in 5G, the vehicular network will produce a large number of computing tasks and data, which will bring huge pressures to the limited on-board resource, so shorter task processing delay is required. Mobile edge computing (MEC) is a promising paradigm to achieve low-latency and low-energy consumption by allowing Vehicle Users (VUs) to offload tasks to the MEC server. However, a single MEC server serves multiple VUs which is prone to resource congestion. In this paper, a scenario of multi-vehicle users and multi-MEC servers in vehicular networks composed of heterogeneous resources is built. In order to make full use of the resources and maximize the average system utility, the joint optimization problem of tasks offloading and heterogeneous resource allocation is formulated as a mixed integer nonlinear problem, where the transmission power allocation scheme, computing resource allocation scheme and optimal offloading policy are given. Then, the three-stage Multi-round combined offloading scheduling mechanism and joint resource allocation strategy is proposed, which decomposes the joint optimization problem of tasks offloading and heterogeneous resource allocation into three stages. Due to the coupling relationship between resource allocation and task offloading, a stable convergent solution can be obtained after several iterations. Finally, the simulation results show that with the increase of workloads and vehicle numbers, compared with other algorithms, the proposed algorithm has better performance on system utility.