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

Signal classification and identification method

A recognition method and signal classification technology, applied in character and pattern recognition, advanced technology, climate sustainability, etc., can solve the problem of low classification accuracy of received signals, achieve suppression of modal confusion, reduce calculation load, and reduce weight The effect of construction error

Active Publication Date: 2022-08-09
BEIJING INSTITUTE OF TECHNOLOGYGY
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to propose a signal classification and recognition method for the radio frequency signal received by the wireless device due to noise, occlusion and various types of interference, which leads to low classification accuracy of the received signal. The method is based on the improved EEMD, radial integral bispectrum and neural network to classify and identify different types of wireless signals

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
  • Signal classification and identification method
  • Signal classification and identification method
  • Signal classification and identification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0057] The wireless communication system on which the signal classification and identification method is based includes a transmitter, a receiver and a radio frequency device. The transmitter includes a baseband modulation module, an up-conversion module and an antenna; the radio frequency equipment includes an energy collection module, a signal processing module, a logic circuit module and a transmission module;

[0058] The baseband modulation module in the transmitter is connected with the up-conversion module, and the up-conversion module is connected with the antenna; in the radio frequency equipment, the logic circuit module is connected with the energy collection module, the signal processing module and the sending module;

[0059] The baseband modulation module modulates the baseband signal to obtain a modulated symbol; the up-conversion module up-converts the modulated symbol to obtain a radio frequency signal, and the antenna sends the radio frequency signal; the ener...

Embodiment 2

[0136] Further, in order to verify the effectiveness of this method, a public dataset is used:

[0137]

[0138]The data in the 2.4GHz Indoor Channel Measurement data set in , which contains the S21 measurement values ​​of 10 frequency sweeps, each sweep contains 601 frequency points, the interval between each frequency point is 0.167MHz, covering the 2.4GHz center frequency 100MHz bandwidth . The method first uses the improved EEMD to process the signal, decomposes the original signal, and retains the relatively important signal components. Compared with the existing classification algorithms based on EMD and bispectral decomposition, the amount of data is reduced, thereby saving the time occupied by subsequent bispectral analysis. At the same time, compared with EMD, the improved EEMD suppresses the mode to a certain extent. Obfuscation reduces reconstruction error. However, the machine learning classification method based on bispectral decomposition + PCA + EMD first p...

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 data representation, classification and signal identification, and particularly relates to a radio frequency signal classification and identification method based on EMD decomposition, bispectrum features and a neural network. The method comprises the following steps: circularly adding noise into a signal to be classified, and taking a modulus value to obtain amplitude data; performing improved lumped average empirical mode decomposition on the amplitude data according to the effective data order to obtain an eigenmode function component and a residual signal; dividing residual signals to obtain a training data set and a test data set; traversing samples in the training data set and the test data set to extract bispectrum features to obtain bispectrum feature matrixes of the samples, and respectively constructing a training set and a test set; constructing a convolutional neural network; inputting the training set into a convolutional neural network for training to obtain a trained model; and inputting the test set into the trained model to obtain a classification recognition result. According to the method, high-accuracy classification of the received radio frequency signals is realized.

Description

technical field [0001] The invention belongs to the technical field of data representation, classification and signal identification, in particular to a classification and identification method based on improved overall average empirical mode decomposition (Modified Ensemble Empirical Mode Decomposition, MEEMD), radial integral bispectrum and neural network. Background technique [0002] 5G promotes the development of the Internet of Everything and the Internet of Things. An important feature of the Internet of Things is large-scale connections. A large number of devices generate different types of signals, including analog, digital, image and other signals in the form of data, causing various data classification and signal recognition problems. Data classification is also widely used, such as for Internet of Things identification, space spectrum resource management and control, and radio frequency security. Therefore, signal classification and identification have great appl...

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): G06K9/00G06N3/04
CPCG06N3/045G06F2218/06G06F2218/12Y02D30/70
Inventor 卢继华李兆军冯立辉
Owner BEIJING INSTITUTE OF TECHNOLOGYGY