System-Wide Energy Efficient Computation Offloading in Vehicular Edge Computing With Speed Adjustment

Abstract

Vehicle-to-everything (V2X) communications in future 6G intelligent transportation systems are expected to enable various convenience applications which consume amount of computation and storage resources in vehicular networks to deliver high-quality, low-latency immersive experiences via vehicular edge computing (VEC). However, as the number of intensive tasks increases, the trade-off problem between task latency requirements and energy consumption becomes more prominent. In this paper, we study the problem of system-wide energy efficient computation offloading in speed-adjustable vehicular edge computing. We firstly consider a novel task offloading environment that considers vehicle speed adjustment to provide latency-constrained computation services for resource-limited vehicles, which fully stimulates the collaborative ability of the transportation system. We formulate the problem as a mixed-integer nonlinear programming problem to minimize the weighted energy consumption of multiple tasks. To solve this problem, we decouple it into two sub-problems, namely the task offloading decision and resource allocation problem, and the vehicle speed adjustment problem. We propose a low-complexity algorithm based on dynamic programming and a speed adjustment algorithm using a direction operator. Simulation results demonstrate the effectiveness of the proposed algorithms in optimizing the weighted energy consumption of the whole system.