A Traffic Flow Prediction Based Task Offloading Method in Vehicular Edge Computing

Abstract

Vehicular Edge Computing (VEC) emerges as a promising paradigm by deploying computation and storage resources on edge servers located in close proximity to vehicles, such as roadside units or base stations. This proximity enables VEC to provide abundant resources and low-latency services, catering to the computational needs of vehicles. However, the dynamic nature of traffic flow presents new challenges in terms of task offloading decisions and resource allocation within VEC environments. This paper addresses the task offloading problem in VEC systems, considering the impact of dynamic traffic flow. To address this problem, we formulate an integer programming model that captures the essence of the studied scenario. To devise an efficient solution, we propose a novel traffic flow prediction-based heuristic algorithm (TFPVTO). TFPVTO incorporates different rules and strategies to generate an optimal offloading task sequence, make informed offloading decisions, and allocate resources effectively. To assess the performance of the proposed algorithm, we utilize a real-world traffic flow dataset. In order to fine-tune and optimize the algorithm’s components, we employ a multi-factor analysis of variance (ANOVA) technique. The proposed TFPVTO algorithm is then rigorously compared against other state-of-the-art algorithms, namely TAVF, MONSA, and RANDOM. Through extensive experiments and statistical analysis, we demonstrate the effectiveness and efficiency of the proposed algorithm in terms of task offloading performance.