Unlock instant, AI-driven research and patent intelligence for your innovation.

Machine learning precoding method of single-cell multi-user MIMO system

A machine learning, multi-user technology, applied in the field of wireless communication MIMO systems, can solve the problems of limited learning and feature extraction capabilities, few research and application of multi-user MIMO, and achieve the effect of reducing computational complexity

Active Publication Date: 2021-09-03
SOUTHEAST UNIV
View PDF2 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Because of the limited ability of the network to learn and extract features, most of the literature deals with single-user MIMO or multi-cell MISO scenarios.
However, there are few researches and applications on the precoding design of multi-user MIMO based on the combination of convolutional neural network and unsupervised supervision method.

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
  • Machine learning precoding method of single-cell multi-user MIMO system
  • Machine learning precoding method of single-cell multi-user MIMO system
  • Machine learning precoding method of single-cell multi-user MIMO system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0064] See figure 1 This embodiment provides a single cell multi-user MIMO system machine learning precoding method, which first establishes weight and rate maximization model, and then constructs a convolution with feature extraction capabilities and consolidation nuclear parameters. Neural network, and use the method of monitoring learning and non-supervised learning to train neural networks, the supervisory learning uses the WMMSE algorithm to generate training data to prepare the convolutional neural network, and then use non-supervised learning to train. The method of the present invention is obtained, and the rate performance comparable to the WMMSE algorithm is obtained, and the calculation complexity of the precoding design can be reduced.

[0065] Specifically, in the present embodiment, a single cell multi-user MIMO communication system for serving K users is considered. Each user has N antenna, and the base station side has a M transmit antenna, and also transmits data ...

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 machine learning precoding method for a single-cell multi-user MIMO (Multiple Input Multiple Output) system, which comprises the following steps of: firstly, establishing a weight and rate maximization model, and then constructing a convolutional neural network with feature extraction capability and convolution kernel parameter sharing; training the neural network by adopting a method of combining supervised learning and unsupervised learning, wherein the supervised learning uses a WMMSE algorithm to generate training data to pre-train the convolutional neural network, and then using the unsupervised learning to carry out retraining. By adopting the method provided by the invention, the sum rate performance equivalent to that of the WMMSE algorithm is obtained, and the calculation complexity of precoding design can be reduced.

Description

Technical field [0001] The present invention relates to wireless communications technologies MIMO systems, more particularly to a method of machine learning precoding a single-cell multiuser MIMO system. Background technique [0002] Pre-coding single-cell designs currently used on multi-user MIMO pre-coding system, based on the previous feedback has been fully connected neural network training methods are mostly based on supervised learning or unsupervised learning. Because learning and extraction features limited capacity networks, problems are most single-user MIMO or multi-document processing cell MISO scenario. Pre-coding design and less on research and application of multi-user MIMO-based convolution neural network and unsupervised supervised combination of methods. Inventive content [0003] In view of this, object of the present invention is to provide a machine for single-cell multiuser MIMO system learning precoding method, to solve the problem mentioned in the backgro...

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
IPC IPC(8): G06N3/04G06N3/08H04B7/0452
CPCG06N3/08G06N3/088H04B7/0452G06N3/045
Inventor 陈明张明辉
Owner SOUTHEAST UNIV