Distributed communication signal modulation identification method

Abstract

The invention discloses a distributed communication signal modulation identification method, comprising the steps that a plurality of sensing nodes are set in a communication area; partial sensing nodes receive signals transmitted by a transmitting source and generate sample data; a receiving source receives the sample data transmitted by the partial sensing nodes and the receiving source extractsfeature parameters from the sample data; the receiving source carries out information fusion processing on all sample data and generates to-be-processed data; the receiving source establishes a neural network classifier through training and learning; and the to-be-processed data is input into the neural network classifier and the neural network classifier outputs modulation identification resultsof the signals transmitted by the transmitting source. According to the method, the receiving source receives wireless communication signals transmitted by the transmitting source from the pluralityof sensing nodes, and the receiving source can comprehensively judge the feature parameters of the sample data of various sensing nodes in the communication signal receiving process, so the accuracy of identifying communication signal modulation modes by the receiving source is improved. The method is used for identifying the modulation modes of the communication signals.

Description

technical field [0001] The invention relates to the communication field, and more specifically relates to a signal modulation identification method. Background technique [0002] Signal modulation recognition technology is widely used in civilian and military communications, such as signal monitoring, signal confirmation, interference identification, electronic countermeasures, electronic rescue, and military threat analysis. It is the basis for research in software radio, cognitive radio, and spectrum sensing. [0003] The existing communication signal modulation recognition method adopts the pattern recognition method, which mainly includes two parts: the extraction of eigenvalues ​​and the selection of classifiers. , high-order cumulant, wavelet transform, spectral correlation, cyclic spectral correlation, power spectrum, constellation diagram, kurtosis of the signal, etc., and the neural network classifier is often used in the step of selecting the classifier. The premi...

AI Technical Summary

Problems solved by technology

[0003] The existing communication signal modulation recognition method adopts the pattern recognition method, which mainly includes two parts: the extraction of eigenvalues ​​and the selection of classifiers. , high-order cumulant, wavelet transform, spectral correlation, cyclic spectral correlation, power spectrum, constellation diagram, kurtosis of the signal, etc., and the neural network classifier is often used in the step of selecting the classifier. The premise of the neural network classifier in the modulation recognition method is to identify the modulation of the wireless communication signal with a single-node signal. Unable to obtain high signal modulation recognition accuracy

Images