A low-latency vehicle edge computing network distributed task offloading solution
Wei Wei Hua, Yisheng An · 2023
With the continuous development of the Internet of Vehicles (IoV), the computational capabilities of vehicle nodes have been gradually enhanced, allowing them to handle numerous computationally intensive and latency-sensitive applications. However, these applications generate complex data that cannot be processed entirely by individual vehicle nodes. To address this issue effectively, task offloading using vehicle edge computing networks can be employed. This paper introduces the concept of vehicle offloading reputation as one of the criteria for selecting service vehicles. It evaluates the overall performance of nearby vehicles based on factors such as available computational resources and vehicle link stability, and identifies vehicles with higher overall performance as service providers. Subsequently, a heuristic algorithm is used to decompose tasks into subtasks equal to the number of selected service vehicles, which are then distributed to their corresponding service vehicles. Experimental results demonstrate that this approach achieves significant performance improvements in terms of latency and offloading success rate.