Reverse Offloading for Latency Minimization in Vehicular Edge Computing

The safety of autonomous driving can be improved with the support of Cooperative Vehicle-Infrastructure System (CVIS) and Vehicular Edge Computing (VEC), which benefit greatly from crowdsensing of CVIS and accurate decision in a short deadline of VEC. In the CVIS, the vehicles will upload the crowdsensing data to the VEC server for data fusion and tasks generating. However, with the ever-increasing number of vehicles, the VEC server cannot undertake massive computation-intensive tasks due to the limited edge computing capabilities. In this paper, we propose a reverse offloading framework to fully utilize the computation resource of vehicles to relieve the burden of the VEC server in a multi-vehicle mobile edge network. First, the system latency minimization problem is formulated as a mixed integer nonlinear programming problem by optimizing reverse offloading decisions and the communication and computation resources allocation. Next, the original problem is transformed into an equivalent weighted-sum optimization problem, which can be decoupled as two subproblems, i.e., resource allocation and decision selection subproblems. The closed-form expressions for the optimal resource allocation are derived by the dual decomposition method in a distributed fashion. Moreover, a low complexity greedy based efficient searching (GES) algorithm is proposed to obtain the reverse offloading decision strategies. Simulation results show that the proposed algorithm can significantly improve the performance compared with other baseline schemes.