Partial Offloading and Resource Allocation for MEC-Assisted Vehicular Networks

Due to the long-distance communication and the limited resources, traditional full offloading algorithms cannot fully exploit radio and computation resources, which may cause more delays and higher energy consumption in mobile edge computing (MEC)-based vehicle-to-everything (V2X) networks. To solve this problem, a novel resource allocation (RA) algorithm with partial offloading for uplink multiuser MEC-based V2X networks is proposed. The optimization objective is to minimize the weighted sum of delays and energy consumption subject to the constraints on the delay requirements of tasks, communication distance, and the computation capability of roadside units. The formulated problem is non-convex and challenging to solve. To this end, an alternating optimization approach is designed to convert the original problem into three subproblems. A tabu search-based matching algorithm (TSM) is designed to solve the offloading object-selection subproblem. Then, a graph coloring algorithm is used to achieve channel allocation. Moreover, the computation RA subproblem is transformed into a convex optimization problem through variable substitution, and the suboptimal offloading ratios and computation RA coefficients are obtained. Simulation results show that the proposed algorithm achieves a lower total cost than the full offloading algorithm and the shortest distance matching algorithm.