Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Multi-level user mobile edge computing method based on reinforcement learning

A secondary user and edge computing technology, applied in the field of wireless communication, can solve problems such as user delay energy consumption requirements, system utility, optimal strategy convergence speed, etc., to improve system resource utilization and meet delay and energy consumption requirements, and the effect of improving utility

Active Publication Date: 2019-05-24
FOSHAN SHUNDE SUN YAT SEN UNIV RES INST +2
View PDF4 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the problem that the traditional mobile edge computing method cannot take into account the user's delay energy consumption requirements, system utility, and optimal strategy convergence speed, and provides a multi-level user based on priority scanning under the Dyna structure in reinforcement learning mobile edge computing

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Multi-level user mobile edge computing method based on reinforcement learning
  • Multi-level user mobile edge computing method based on reinforcement learning
  • Multi-level user mobile edge computing method based on reinforcement learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0042] This embodiment designs a mobile edge computing method for multi-level users based on reinforcement learning. The processing flow of the exchange method in the present invention includes:

[0043] S1, system initialization parameters, determine the number of primary users N P , The number of secondary users N S , The number of edge servers and the number of control nodes N M , The transmission power P of the secondary user, the task amount of the secondary user is Task, the channel capacity C of the channel between each secondary user and the edge server, the communication urgency Em is zero; the initialization method parameters, the edge server corresponds to Channel status The initial value is unoccupied, the value is zero, the initial Q value is zero, the learning rate is α, the discount factor is δ, the priority function value is zero, the priority threshold is θ, and the priority queue is empty, start iteration;

[0044] S2, the primary user chooses to occupy the edge ...

Embodiment 2

[0054] This embodiment is attached with the description figure 1 Go to FIG. 5 and use a specific mobile edge computing embodiment with two secondary users to illustrate the mobile edge computing method based on reinforcement learning proposed in the present invention in detail.

[0055] Consider the system model as follows: In the process of mobile edge computing, 1 primary user and 2 secondary users choose to use 3 edge servers to perform offloading tasks. The task amount of secondary user 1 is Task 1 , The task amount of secondary user 2 is Task 2 , The channel capacity between the two secondary users and the three edge servers is matrix C, and the communication urgency is zero. The channel status corresponding to the edge server is all zero, that is, it is not occupied. The Q value is zero, the learning rate is 0.8, the discount factor is 0.02, the priority function value is zero, the priority threshold is 0.15, and the priority queue is empty, and the iteration starts.

[0056...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a multi-level user mobile edge calculation method based on reinforcement learning, which is based on a priority scanning method under a Dyna structure, is suitable for a mobileedge calculation wireless communication environment with shortage of spectrum resources and high requirements on time delay and energy consumption, and belongs to the field of wireless communication.The method mainly comprises the following four steps of: firstly, selecting an edge server by a primary user; Secondly, enabling a secondary user to propose an application for occupying an edge server to a control center; Then the control center processes the application and distributes an edge server, and enabling the secondary user to perform partial calculation unloading or complete local calculation; And finally, calculating the utility of each secondary user. According to the method, reinforcement learning is applied to the mobile edge computing wireless communication network, and the advantages of no model of Q learning and preferential updating of a priority scanning method are combined, so that the requirements of time delay and energy consumption of each secondary user are met while the overall utility performance of the system is ensured, and the utilization rate of resources is improved.

Description

Technical field [0001] The present invention relates to the field of wireless communication, and more specifically, to a multi-level user mobile edge computing method based on reinforcement learning, which is used for mobile edge computing wireless with tight spectrum resources, high time delay and high energy consumption requirements. The method of communication environment. Background technique [0002] The traditional priority scanning method is only suitable for deterministic environments, and it is impossible to obtain ideal results for unknown environments. However, based on the traditional model-free Q learning method, all state action values ​​are updated, and all state action values ​​are calculated multiple times, which takes a long time, and it is impossible to quickly determine the optimal selection of edge servers and determination of mobile edge offloading tasks. Strategy. In order to improve the spectrum resource utilization of the entire system, researchers have...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): H04W24/02H04W72/04H04W72/10
CPCY02D30/70
Inventor 葛颂阳肖亮龚杰陈翔
Owner FOSHAN SHUNDE SUN YAT SEN UNIV RES INST
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com