Efficient Task and Data Scheduling Policy for Vehicular Fog Computing Based on Link Weight

Abstract

Vehicular network has several applications in the smart city and IoT. Recently with the advancement in the computing technology such as fog computing and its application in the vehicular network and its services, a new paradigm known as vehicular fog computing has evolved as a hot topic of investigation in the research community because of the next generation computing and communication requirements. Vehicular fog computing can be used to solve the issues of next generation computing and communication scenario. There are several issues in vehicular fog computing. Efficient task computing and data dissemination is an important issue. Several approaches are proposed by different authors to solve the issues, but none of them has addressed the service completion and failure rate which is very important in the vehicular scenario as the vehicles move very fast and its contact time with the RSU controller is limited. The task has to be completed by the vehicular server within that time period, otherwise computation will fail. Once the computation and communication fails, the RSU controller will reinitiate to form the vehicular fog resulting high overhead. In this paper we address this issue and proposed an efficient scheduling algorithm based on multiple parameters namely queue length, response time and link weight. We simulated the algorithm using java and compared with the existing algorithm showing better performance.