Resource allocation and task unloading optimization method based on multiple agents

By constructing a Stackelberg game model in a heterogeneous network environment and using a multi-agent deep reinforcement learning algorithm to optimize edge cloud resource allocation and task offloading, the efficiency issues of resource allocation and task offloading in heterogeneous networks are solved. Improved calculation success rate and resource utilization.

CN115175217AActive Publication Date: 2022-10-11GUANGZHOU FATONG NETWORK TECH CO LTD
5 Cites 38 Cited by

Patent Information

Application Number
CN202210758663.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2022-06-30
Publication Date
2022-10-11
Estimated Expiration
2042-06-30

AI Technical Summary

Technical Problem

In a heterogeneous network environment, it is difficult for existing technologies to effectively optimize resource allocation and task offloading between edge cloud servers and mobile devices, resulting in insufficient and uneven resource utilization and affecting user service experience.

Method used

Treat edge cloud servers and energy harvesting-enabled mobile devices as independent decision-making agents, construct a Stackelberg game model, and optimize edge cloud resource collaborative allocation and tasks through multi-agent deep reinforcement learning algorithms. Offloading strategy adapts to stochastic time-varying network environments and incomplete status observations.

Benefits of technology

It improves the task calculation success rate of edge cloud servers, reduces the task discarding rate of mobile devices, and improves resource utilization efficiency and user experience.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention belongs to the technical field of mobile communication, and relates to a resource allocation and task unloading optimization method based on multiple agents. The method comprises the following steps: taking an edge cloud server and an energy collection enabled mobile device in a heterogeneous network environment as intelligent agent units capable of independently deciding, combining an actual calculation unloading scene, and constructing a Stackelberg game model through the earnings of edge cloud resource allocation and mobile device task unloading; in consideration of a random time-varying edge environment and incomplete state observation, modeling the Stackelberg game model into a partially observable Markov decision process again, and establishing a multi-agent deep reinforcement learning algorithm to solve a partially observable Markov decision process game model; and the optimal edge cloud resource collaborative allocation strategy and task unloading strategy are learned. The task calculation success rate of the edge cloud server can be effectively improved, and the task discarding rate of the mobile equipment is reduced.
Need to check novelty before this filing date? Find Prior Art