Computational Task Offloading Method for Internet of Vehicles

Abstract

The invention discloses a computing task offloading method oriented to the Internet of Vehicles. In order to improve the revenue of computing resource providers in the computing task offloading process in the prior art and ensure the completion delay of the computing task to a certain extent, the invention adopts the base station side The computing resources of the MEC serve and the computing resources of road vehicles provide an offloading solution for the computing tasks of the roadside UE; the communication delay of the task, the computing delay, and the processing rate and queue length of the computing processing queue are included in the decision-making reference factors. The mathematical model of Markov decision-making is used to analyze the change of the state, and the deep learning algorithm is used to obtain an optimization strategy space, that is, when the corresponding state adopts the corresponding behavior, the resource provider's revenue is maximized, and the user's calculation is guaranteed to a certain extent. The latency requirement of the task.

Description

technical field [0001] The invention belongs to the field of wireless communication, and in particular relates to a technology for unloading edge computing tasks by a deep learning algorithm. Background technique [0002] With the acceleration of the growth rate of mobile smart devices and the continuous emergence of mobile applications, the traditional centralized network architecture cannot meet the computing requirements of low latency and high performance. These growing application requirements have inspired the development of mobile edge computing (MEC). As a promising technology, mobile edge computing can provide edge computing services at the edge of the cellular network because computing resources are closer to the mobile device end. By reducing network operation and service delivery delays, and improving user service quality experience, it has been widely accepted. Widely concerned, and MEC has been included in the 5G standard as a key technology. [0003] MEC aim...

AI Technical Summary

Problems solved by technology

However, less research effort has been devoted to investigating how computational tasks can be offloaded to mobile vehicles

Vehicles and base stations, as two important computing resource providers, are usually analyzed separately in existing research, and the computing resource processing capacity of separate vehicle offloading or base station offloading is usually limited

Images