An Efficient FLI-KDMSSA Framework for Computing Resource Allocation of IoV in Edge Computing

The combination of Mobile Edge Computing (MEC) and Internet of Vehicles (IoV) can effectively improve the network performance. However, the mobility of vehicles and the diversity of tasks make the allocation of computing resources more complex. When the vehicle is in motion, its position can change at any time. This can result in overload of the edge servers. Meanwhile, vehicle tasks are sensitive to latency. It makes resource allocation within edge servers more difficult. In order to solve the above problems, this article proposes a FLI-KDMSSA framework for rational allocation of computing resources in the Internet of Vehicles. First, Fuzzy Logic Inference (FLI) algorithm is used in this framework to determine the computing nodes of IoV tasks in edge computing scenarios. This algorithm uses task length, edge server virtual machine utilization, and cloud bandwidth as parameters to establish fuzzy rules. Then, with the objective function of minimizing latency and load balancing values, this paper proposes a Discrete Multi-objective Sparrow Search Algorithm based on K-means (KDMSSA) to solve the virtual machine resource allocation scheme. The experiment is simulated on the iFogSim platform. To compare with PSO algorithm, the performance of KDMSSA is improved by 12.7%. To compare with SSA, the performance of KDMSSA is improved by 7.7%.