Joint Task Proportion and Edge Node Selection for Latency Minimization Based on MEC Collaboration

By migrating computation-intensive tasks to edge nodes(ENs) deployed at the edge of the network in Internet of Vehicles(IoV),Mobile Edge Computing (MEC) significantly improves the performance of mobile applications. In order to fur-ther improve the resource utilization efficiency of MEC system, task proportion strategy and EN selection strategy based on MEC collaboration are proposed in this paper. To minimize the total latency of all vehicle users(VUs), we consider a mixed-integer nonlinear programming(MINLP) problem. This problem can be divided into two sub-problems: task proportion sub-problem and EN selection sub-problem. To solve the task proportion sub-problem, the optimal solution can be obtained by the time relation between cooperative ENs. For the sub-problem of EN selection, we use the Lagrange dual method and the gradient descent method to get the optimal solution. Simulation results show that the combination of two strategies can effectively reduce the total latency of VUs, which improves the experience of VUs and the resource utilization efficiency of MEC system.