CN-116896492-B - Modulation and coding joint identification method and system based on multichannel attention network

CN116896492BCN 116896492 BCN116896492 BCN 116896492BCN-116896492-B

Abstract

The invention relates to the technical field of signal modulation coding joint identification, in particular to a modulation coding joint identification method and a modulation coding joint identification system based on a multichannel attention network, which utilize a randomly generated radio information sequence to simulate a communication environment coding modulation information sequence and generate a digital modulation signal and a simulation modulation signal, and construct signal sample data; the method comprises the steps of constructing a multichannel convolution gating depth attention network model for modulation and coding joint identification, training and optimizing the multichannel convolution gating depth attention network model by utilizing signal sample data, inputting a signal to be identified into the multichannel convolution gating depth attention network model after training and optimizing, and obtaining modulation and coding types of the signal to be identified by utilizing the multichannel convolution gating depth attention network model after training and optimizing. The invention can improve the modulation and coding joint identification efficiency, can realize the signal modulation and coding joint automatic identification task, and is convenient for deployment and implementation.

Inventors

WU DI
YI DONG
WANG SHIJU
LIU KAIYUE
ZHANG JINGZHI
WANG SHU
LU WANJIE
HU TAO

Assignees

中国人民解放军战略支援部队信息工程大学

Dates

Publication Date: 20260508
Application Date: 20230614

Claims (5)

1. A modulation and coding joint identification method based on a multichannel attention network, comprising: Simulating a communication environment using the randomly generated radio information sequences to code the modulated information sequences and generating digital modulated signals and analog modulated signals; The method comprises the steps of utilizing a digital modulation signal and an analog modulation signal to construct signal sample data, and dividing the signal sample data into a training sample set, a verification sample set and a test sample set according to a preset proportion; Constructing a multichannel convolution gating depth attention network model for modulation and coding joint identification, and training and optimizing the multichannel convolution gating depth attention network model by using a training sample set, a verification sample set and a test sample set; the constructed multichannel convolution gating depth attention network model comprises a multichannel convolution module for extracting a feature map from input data through multichannel convolution operation, a dimension-increasing convolution module for carrying out dimension-increasing convolution on the extracted feature map to extract a target shape and detail in the feature map, a gating cyclic classification module for capturing deep connection dimensional information and classifying output modulation categories, and an extrusion excitation module for utilizing interdependence among channel weight vector explicit modeling channels to self-adaptively recalibrate channel weights, wherein the multichannel convolution module is used for extracting the feature map from the input data through multichannel convolution operation and comprises a step-by-step up sampling three convolution units for gradually outputting a high resolution feature map from the input low resolution feature map, wherein the convolution units comprise one-dimensional convolution layers, random inactivation layers, a reverse channel activation layer and a hidden node, the three convolution layers are used for respectively convolving an I channel signal, a Q channel signal and an I/Q signal in the input data, and splicing the output of the three convolution layers according to the input dimension to obtain the feature map of the input data, the dimension-increasing convolution module is used for carrying out dimension-increasing convolution on the extracted feature map and comprises a step-increasing sampling step by using three convolution units connected in series to output a high resolution feature map, the convolution unit comprises a one-dimensional convolution layer, a random inactivation layer, a reverse activation layer and a hidden node is used for capturing hidden node output hidden node which is connected with the current state hidden node output and hidden node output hidden node transmission hidden state, capturing dimension information of a feature map of deep connection through gating state updating, classifying the feature map by utilizing a full-connection output layer to obtain a corresponding signal modulation type, and enabling the extrusion excitation module to utilize the interdependence among channel weight vectors to explicitly model the process of self-adaptively recalibrating channel weights, wherein the process is expressed as follows: , for new feature vectors Is the first of (2) The number of elements to be added to the composition, For inputting feature vectors Channel weight vector of (2) Middle (f) Individual elements And (2) channel weight of , Channel statistics vectors generated by global average pooling operations for a layer of output feature vectors on the network, As a function of the ReLU, The function is activated for Sigmoid, To squeeze the weight parameters of the first fully connected layer in the excitation module, Weight parameters of a second full connection layer in the extrusion excitation module; Inputting the signal to be identified into a multi-channel convolution gating depth attention network model after training optimization, and obtaining the modulation coding type of the signal to be identified by using the multi-channel convolution gating depth attention network model after training optimization.
2. The joint recognition method of modulation and coding based on a multichannel attention network according to claim 1, wherein the analog communication environment coding the modulation information sequence using the randomly generated radio information sequence and generating the digital modulation signal and the analog modulation signal comprises: firstly, a 01 bit radio information sequence is randomly generated by utilizing a random function; Then, simulating a communication environment, and adding noise interference in the code modulation and sampling process of the radio information sequence to obtain an IQ sampling sequence of a modulation signal, wherein the noise interference comprises additive Gaussian noise and Rayleigh fading channel model influence; Then, for the IQ sample sequence, a plurality of digital modulation signals combined by each channel coding and modulation scheme are generated according to the digital modulation signal code rate, and an analog modulation signal without channel coding is generated.
3. The joint recognition method of modulation and coding based on multichannel attention network according to claim 1 or 2, wherein the digital modulation signals comprise 28 digital modulation signals formed by combining four channel codes of LDPC codes, RS codes, TCM codes and TPC codes and seven modulation modes of 2FSK, GMSK, BPSK, QPSK, 8PSK, 8APSK and 16APSK, and the analog modulation signals comprise AM and FM signals without channel codes.
4. The modulation and coding joint recognition method based on the multichannel attention network according to claim 1, wherein the method is characterized in that signal sample data are built by utilizing digital modulation signals and analog modulation signals, the signal sample data are divided into a training sample set, a verification sample set and a test sample set according to preset proportions, the method comprises the steps of continuously collecting M information points to form signal samples by taking n information points as sampling intervals for each modulation signal, collecting M signal samples by each modulation signal, building the signal sample data set according to the signal samples of all modulation signals, and extracting sample data in each type of modulation and coding signals from the signal sample data set according to preset proportions to form the training sample set, the verification sample set and the test sample set, wherein M, n and M are preset thresholds.
5. A modulation and coding joint recognition system based on a multichannel attention network is characterized by comprising a data simulation module, a data processing module, a model construction module and a target recognition module, wherein, A data analog module for simulating a communication environment using the randomly generated radio information sequences to code the modulated information sequences and generating digital modulation signals and analog modulation signals; The data processing module is used for utilizing the digital modulation signals and the analog modulation signals to construct signal sample data and dividing the signal sample data into a training sample set, a verification sample set and a test sample set according to a preset proportion; The model construction module is used for constructing a multi-channel convolution gating depth attention network model for modulation coding joint identification and utilizing a training sample set, a verification sample set and a test sample set to train and optimize the multi-channel convolution gating depth attention network model, the constructed multi-channel convolution gating depth attention network model comprises a multi-channel convolution module used for extracting a feature map from input data through multi-channel convolution operation, an up-dimension convolution module used for carrying out up-dimension convolution on the extracted feature map to extract target shapes and details in the feature map, a gating circulation classification module used for capturing deep connection dimension information and classifying output modulation categories, and an extrusion excitation module used for utilizing interdependence among channel weight vector explicit modeling channels to self-adapt to recalibration channel weights, wherein the multi-channel convolution module is used for extracting the feature map from the input data through multi-channel convolution operation and comprises the steps of respectively carrying out convolution on I channel signals, Q channel signals and I/Q signals in the input data through three convolution layers according to input dimensions so as to acquire the feature map of the input data, the up-dimension convolution module carries out up-dimension convolution on the extracted feature map to capture deep connection dimension information and classify and output modulation category information, the up-dimension convolution module comprises a one-dimension conversion unit, and a one-dimensional convolution unit, the down-conversion unit is sequentially connected with the down-conversion unit, the LU classification unit is sequentially and the down-conversion unit is activated, the down-conversion unit is normalized, and the down-conversion unit is sequentially normalized, and the down-conversion unit is activated, the method comprises the steps of outputting current hidden node output by combining a current input and a transmission hidden state of a previous node by using a gating circulation unit GRU, and transmitting a hidden state of a next node to capture dimension information of a feature map of deep connection by using gating state update, classifying the feature map by using a fully connected output layer to obtain a corresponding signal modulation class, wherein the process of adaptively recalibrating channel weights by using interdependence between channel weight vector explicit modeling channels by using an extrusion excitation module is expressed as follows: , for new feature vectors Is the first of (2) The number of elements to be added to the composition, For inputting feature vectors Channel weight vector of (2) Middle (f) Individual elements And (2) channel weight of , Channel statistics vectors generated by global average pooling operations for a layer of output feature vectors on the network, As a function of the ReLU, The function is activated for Sigmoid, To squeeze the weight parameters of the first fully connected layer in the excitation module, Weight parameters of a second full connection layer in the extrusion excitation module; the target recognition module is used for inputting the signals to be recognized into the multi-channel convolution gating depth attention network model after training optimization, and obtaining the modulation coding types of the signals to be recognized by utilizing the multi-channel convolution gating depth attention network model after training optimization.

Description

Modulation and coding joint identification method and system based on multichannel attention network Technical Field The invention relates to the technical field of signal modulation and coding joint identification, in particular to a modulation and coding joint identification method and system based on a multichannel attention network. Background With the development of communication technology, in order to improve communication capacity, variable code modulation (AVM) and Adaptive Code Modulation (ACM) technologies are being generated and widely used in communication. The self-adaptive code modulation technology can flexibly select a code modulation mode according to the change of channel transmission conditions, and the frequency spectrum utilization rate is greatly improved. When the method is applied in an interactive point-to-point mode, the ACM technology is adopted, so that the satellite communication capacity can be increased by 100% -200%. With the continuous enhancement of modern informatization and digitalization construction, a satellite communication network is easily threatened by detection, interference and attack of a third party, and for improving the communication safety of the satellite communication network, a modulation mode and a coding mode of a communication signal can be identified through unintended interference so as to obtain the modulation mode and the coding mode of the communication signal to further obtain relevant signal parameters, and decision guidance can be provided for perfecting satellite communication management, enhancing inter-system coordination and optimizing anti-interference performance. The modulation and coding joint recognition of the current signals can be divided into a traditional hierarchical recognition algorithm and a modulation and coding joint recognition algorithm based on deep learning. The traditional algorithm is mainly realized in a layering identification mode, namely, firstly, the signals are modulated and identified, then the signals are demodulated, and finally, the demodulated information flow is subjected to coding type identification and parameter estimation, so that the modulation and coding combined identification is realized. By adopting layered identification, the code identification performance depends on the modulation identification performance and the demodulation algorithm error to a certain extent, and when the modulation identification performance is reduced or the demodulation algorithm error is larger, the code identification performance is necessarily reduced. Modulation and coding joint recognition based on deep learning mostly adopts intermediate frequency signal waveforms as deep network inputs, and intermediate frequency signals are unfavorable for deep network extraction of features compared with baseband quadrature in-phase (I/Q) signals. This is because the modulation and coding joint constraint relationship in the signal is difficult to mine, the carrier frequency in the intermediate frequency signal occupies most of the energy of the signal, so that the difficulty of mining the modulation and coding joint constraint relationship in the signal is increased to a certain extent, and the baseband I/Q signal does not have the worry. Meanwhile, the current depth network structure for modulation and coding joint identification is not strong in pertinence, and the extraction capability of modulation and coding joint features is insufficient, so that the identification rate and the robustness of an algorithm are not high. . Disclosure of Invention Therefore, the invention provides a modulation and coding joint identification method and a modulation and coding joint identification system based on a multichannel attention network, which solve the problems of unreasonable signal input form, unsatisfactory identification rate and robustness caused by poor structural design of a depth network in the existing modulation and coding joint identification. According to the design scheme provided by the invention, a modulation and coding joint identification method based on a multichannel attention network is provided, which comprises the following steps: Simulating a communication environment using the randomly generated radio information sequences to code the modulated information sequences and generating digital modulated signals and analog modulated signals; The method comprises the steps of utilizing a digital modulation signal and an analog modulation signal to construct signal sample data, and dividing the signal sample data into a training sample set, a verification sample set and a test sample set according to a preset proportion; Constructing a multichannel convolution gating depth attention network model for modulation and coding joint identification, and training and optimizing the multichannel convolution gating depth attention network model by using a training sample set, a verification sample set and a test sample set; I