The invention discloses an Internet of Vehicles task unloading method and device. The method comprises the following steps: acquiring a first task unloading request sent by a first vehicle in a coverage range of a first road side unit through the first road side unit, and acquiring first mobile edge computing (MEC) server state information of a first MEC server in communication connection with the first road side unit; analyzing the first task unloading request and the first MEC server state information by using a target unloading decision model to obtain a first task unloading proportion of the first task unloading request, the target unloading decision model is a task unloading proportion constructed based on a deep convolutional neural network and used for determining the minimum processing overhead of the task unloading request; and determining a target task unloading strategy of the first task unloading request based on the first task unloading proportion. The technical problem that the actual unloading proportion of the vehicle task cannot be quickly known by a related task unloading decision algorithm is solved.
本申请公开了一种车联网任务卸载方法、装置。其中,该方法包括:获取第一路侧单元覆盖范围内的第一车辆通过第一路侧单元发送的第一任务卸载请求,并获取与第一路侧单元通信连接的第一移动边缘计算MEC服务器的第一MEC服务器状态信息;利用目标卸载决策模型对第一任务卸载请求和第一MEC服务器状态信息进行分析,得到第一任务卸载请求的第一任务卸载比例,其中,目标卸载决策模型是基于深度卷积神经网络所构建的用于确定任务卸载请求的处理开销最小的任务卸载比例;基于第一任务卸载比例确定第一任务卸载请求的目标任务卸载策略。本申请解决了相关任务卸载决策算法无法快速得知车辆任务的实际卸载比例的技术问题。
Internet of vehicles task unloading method and device
车联网任务卸载方法、装置
25.10.2024
Patent
Elektronische Ressource
Chinesisch