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

Power distribution method based on machine learning

An allocation method and machine learning technology, applied in the field of communication anti-interference, to achieve the effect of reducing complexity and reducing repetitive calculations

Inactive Publication Date: 2018-09-21
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF3 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The traditional power allocation algorithm requires repeated calculations. Although researchers have successively proposed algorithms with lower complexity to avoid repeated calculations under a given channel matrix, the whole process is still repetitive. For example, The channel matrix of two times is the same or similar, and the traditional power allocation algorithm still needs to perform two corresponding operations

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
  • Power distribution method based on machine learning
  • Power distribution method based on machine learning
  • Power distribution method based on machine learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] The parameters in the examples do not affect the generality of the invention.

[0031] a. Construction of training set

[0032] The construction of the training set is mainly divided into three aspects: (1) designing training samples from the channel matrix; (2) designing key performance indicators (Key Performance Indicator, KPI); (3) labeling samples based on KPI.

[0033] (1) Generate a training set: training samples are input into the learning system as known variables, assuming there are M N r ×N t dimensional channel matrix as training samples. Because the training sample needs to be a real-valued vector, the channel sample H needs to be m , processed as an N-dimensional real-valued feature vector, and features can be angle, magnitude, and real and imaginary parts of matrix elements, etc. In addition, the extracted feature vectors also need to be normalized to avoid major deviations during training.

[0034] Step 1: From the channel matrix H m generate a rea...

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 belongs to the technical field of communication anti-jamming and particularly relates to a power distribution method based on machine learning. The invention mainly aims to lower the complexity of a power distribution algorithm. The method comprises the following specific steps: processing channel samples into N-dimensional real-valued feature vectors, designing a preset power distribution matrix, labeling, and repeating the step until each sample has a corresponding label; classifying newly input samples by adopting an SVM (Support Vector Machine) algorithm by utilizing a samplefeature set and a corresponding label set, and taking a power distribution matrix corresponding to the output labels as an optimal power distribution matrix corresponding to the samples. The method disclosed by the invention has the beneficial effects that the whole calculation process of the traditional power distribution manner is repetitive, and in order to alleviate the problem, the newly input samples are classified by adopting the SVM algorithm, and the algorithm complexity can be effectively reduced.

Description

technical field [0001] The invention belongs to the technical field of communication anti-jamming, and relates to spatial modulation (Spatial Modulation, SM) technology, multiple input multiple output (Multiple Input Multiple Output, MIMO) technology, and Support Vector Machines (Support Vector Machines, SVM) algorithm. Background technique [0002] As a new MIMO technology, spatial modulation system has been paid attention recently. The basic idea of ​​spatial modulation is: In spatial modulation, only one transmitting antenna is activated to transmit data in each time slot. The transmitting antenna is not only a medium for forming a wireless radio frequency link, but also carries the information bits themselves. Since only one transmit antenna works in each transmission time slot, the interference between sub-antenna channels can be completely eliminated, and precise synchronization timing of the transmit antenna is not required, and at the receiving end, even when the num...

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): G06N99/00G06K9/62
CPCG06F18/2411G06F18/214
Inventor 李泳洋游龙飞杨平肖悦
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA