Framework of Cloud Computing Resource Scheduling for Vehicle Fault Diagnosis

Abstract

Internet of Vehicles (IoVs) provides communication and computing resources, which makes the on-board diagnosis of vehicle faults possible. However, those resources need to be expanded to support the accurate analysis of the on-board diagnosis. Vehicular Cloud Computing (VCC) can solve the pressure of local vehicle processing but will cause an unavoidable delay. Thus, the accuracy and timeliness of on-board diagnosis cannot be guaranteed. To address the issue, we propose a Mobile Edge Caching based Resource Scheduling (MECRS) mechanism for the on-board diagnosis of vehicle faults. According to the urgency of vehicle fault diagnosis, we first design a cloud scheduling algorithm to meet the computation requirements of both the essential business of IoVs and the fault diagnosis. Subsequently, the priority allocation strategy is made for all four types of requests. Then, the urgent requests can be processed timely. Specifically, a multi-objective optimization method is proposed to allocate communication and computing resources for the above requests. In addition, we present a mobile edge caching algorithm in which the large-scale file with high popularity is offloaded to alleviate the pressure of the cloud. Finally, we carry out comprehensive simulations. The results reveal that the developed mechanism provides a high service rate for on-board diagnosis with limited network resources, while the performances of the other three essential services are not compromised.

Keywords

• Mobile edge caching
• vehicle fault diagnosis
• resource scheduling
• vehicular cloud computing