Edge computing resource scheduling optimization method for Internet of Vehicles

Abstract

With the rapid development of the Internet of Things and 5G technology, the Internet of Vehicles has become one of the iconic application scenarios of 5G technology, promoting the emergence of a large number of emerging vehicle-mounted intelligent applications. However, vehicle terminals have limited computing resources, making it difficult to meet the latency and energy consumption requirements of emerging applications such as real-time traffic conditions and intelligent identification. The emergence of Internet of Vehicles technology based on edge computing solves the above problems. In-vehicle edge computing deploys edge servers to Road Side Units (RSU) close to the vehicle to provide computing and storage services to the vehicle, relieving the computing and storage pressure on the vehicle terminal. However, due to the limited resources of the vehicle edge network, unlimited task scheduling to the edge server will cause the server to be overloaded and affect the quality of service (QoS) of vehicle users. A reasonable resource scheduling strategy can ensure the comprehensive performance of the edge network and improve user QoS. Aiming at the problem of complex road traffic and dense vehicles in the Internet of Vehicles that generate too many computing tasks that need to be processed in real time, a partial resource scheduling strategy based on an improved simulated annealing algorithm is proposed for divisible tasks. With the goal of minimizing system overhead, a partial scheduling system model including the offloading ratio factor is constructed. Decompose part of the resource scheduling problem into two factors: offloading ratio and computing resource allocation. The optimal offloading ratio factor is solved through the improved simulated annealing algorithm. During the algorithm iteration process, the optimal resource allocation is obtained through the Lagrange multiplier method. Simulation verified the convergence of the proposed strategy and its efficiency in optimizing delay, energy consumption and overhead.