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

Multi-base-station cooperative wireless network resource allocation method based on graph attention mechanism reinforcement learning

A technology of wireless network resources and reinforcement learning, applied in the field of multi-base station cooperative wireless network resource allocation based on graph attention mechanism reinforcement learning, can solve problems such as inability to cope well, lack of flexibility and scalability, etc.

Active Publication Date: 2021-03-16
ZHEJIANG LAB +1
View PDF3 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Traditional dedicated resource allocation schemes and resource allocation strategies based on optimization algorithms and heuristic algorithms often have strict restrictions and complex derivations to form specific optimization problems. Such methods lack flexibility and scalability. When user characteristics and The proportion of users with various performances changes, and these algorithms cannot cope well

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-base-station cooperative wireless network resource allocation method based on graph attention mechanism reinforcement learning
  • Multi-base-station cooperative wireless network resource allocation method based on graph attention mechanism reinforcement learning
  • Multi-base-station cooperative wireless network resource allocation method based on graph attention mechanism reinforcement learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0041] In order to describe in detail the technical content, structural features, achieved goals and effects of the technical solution, the following will be described in detail in conjunction with the accompanying drawings.

[0042] refer to figure 1 , is a flow chart of the multi-base station cooperative wireless network resource allocation method based on graph attention mechanism reinforcement learning of the present invention, specifically including the following steps:

[0043] S1. Algorithm network structure G and target network Build and initialize, including the following sub-steps:

[0044] S11. The algorithm network structure G of this method includes three parts: a state vector encoding network (Embed), a graph attention mechanism network (GAT) and a deep Q network (DQN).

[0045] S12. The state vector encoding network is composed of two layers of fully connected networks, denoted as

[0046] , (1)

[0047] in , is the weight matrix of the layer, is th...

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-base-station cooperative wireless network resource allocation method based on graph attention mechanism reinforcement learning. The method comprises the following steps: constructing and initializing an algorithm network structure G and a target network; executing resource allocation; repeating the resource allocation in the step 2, and training an algorithm networkstructure G; assigning a weight parameter of the algorithm network structure G to the target network every time the algorithm network structure G in the step 3 is trained for X times, thereby realizing updating of the target network; and after the step 3 is executed, completing the training process of the algorithm network structure G after executing for several times. The internal relation between the subjects is obtained through the graph attention mechanism, the fluctuation situation of each slice data packet in space-time is analyzed, and compared with a resource allocation strategy basedon an optimization algorithm and a genetic algorithm and a resource allocation strategy based on traditional reinforcement learning, higher system return can be obtained, namely higher spectral efficiency and better user experience are obtained, and the method can be adapted to a dynamically changing environment, and is more flexible and robust.

Description

technical field [0001] The present invention relates to a multi-base station cooperative network resource allocation method and the field of reinforcement learning, more specifically, to a multi-base station cooperative wireless network resource allocation method based on graph attention mechanism reinforcement learning, and belongs to the technical field of wireless communication. Background technique [0002] Faced with the rapid growth of mobile data traffic, the fifth generation (5G) mobile communication network needs to provide network services with different performances for diverse business scenarios from different subscribers. The three core application scenarios are: (a) enhanced Enhanced mobile broadband (eMBB) is used to provide users with stable and high-peak data transmission rates to meet typical services such as 4k / 8k HD, AR / VR, and holographic images; (b) massive machine communication (massivemachine -type communications (mMTC), used to provide services for l...

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): H04W24/02H04W72/04G06K9/62G06N3/04G06N3/08
CPCH04W24/02G06N3/08G06N3/045G06F18/214H04W72/53
Inventor 李荣鹏邵燕郭荣斌赵志峰张宏纲
Owner ZHEJIANG LAB
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