Internet of vehicles task unloading method based on service priority

Abstract

The invention relates to Internet of Vehicles task unloading based on service priorities. The method comprises the following steps: receiving a first task unloading request sent by a vehicle terminal in the coverage range of the RSU through the RSU in real time, wherein the first task unloading request is a task unloading request sent for the first time when the vehicle terminal enters the coverage range of the RSU; according to the first task unloading request, distributing a weight for the first task unloading request, and determining a task priority of the first task unloading request; according to the task priority of the first task unloading request, determining a currently accepted unloading task, processing the currently accepted unloading task, and feeding back to the corresponding vehicle terminal; according to the task priority of the first task unloading request, generating an unloading strategy according to a deep learning network, and generating a reward value; and sending the unloading strategy to the vehicle terminals in the coverage range of the RSU, so that each vehicle terminal performs data transmission of task unloading through the RSU according to the unloading strategy, and the calculation task execution success probability is improved.

Description

technical field [0001] The present application relates to the technical field of Internet of Vehicles communication, and in particular to a task offloading method for Internet of Vehicles based on service priority. Background technique [0002] In recent years, with the development of Internet of Vehicles (IoV) and artificial intelligence (AI) technologies, it has promoted the realization of a series of computing-intensive applications, such as autonomous driving, virtual reality, etc., vehicles will generate a large amount of related content, including time, space information and user application information, in order to meet the increasing computing demands, on-board computers will face great challenges in providing high-quality services. [0003] In the early research stage of the Internet of Vehicles, the first consideration is to use the powerful computing power of the cloud to provide service support. Although cloud computing greatly improves resource utilization and c...

AI Technical Summary

Problems solved by technology

but failed to take into account the priority of tasks and the availability of services from neighboring vehicles

As a result, all tasks have the same probability to be offloaded to the server and obtain the corresponding computing resources, some tasks with strict delay may not be completed within the maximum delay, meanwhile, due to the short duration of the V2V link or insufficient computing resources of the vehicle, May cause uninstallation to fail

[0006] Therefore, the current task offloading method for the Internet of Vehicles has a low offloading success rate.

Images