Signal modulation mode identification method based on convolutional restricted Boltzmann machine

Abstract

According to a signal modulation mode identification method based on the convolutional restricted Boltzmann machine, the purpose of efficiently identifying common modulation modes can be achieved, thecomplexity and difficulty of modulation identification are reduced, and the method is applicable to identification of various common modulation modes. The method comprises the following steps: S1, preprocessing an original signal, and constructing a signal modulation category label; S2, constructing a training set and a test set for the signal data set with the label obtained after preprocessing;S3, constructing an identification network model; using two continuous convolution restricted Boltzmann machine networks as a signal feature extraction layer; finally, fully expanding and inputting the extracted signal abstract features into a classifier for modulation mode recognition; S4, training the recognition network model to obtain a trained recognition network model; and S5, detecting andacquiring a to-be-identified wireless communication signal, preprocessing the signal, inputting the signal into the trained identification network model, and identifying the signal modulation mode category.

Description

technical field [0001] The invention relates to the technical field of signal processing, in particular to an identification method of a signal modulation mode based on a convolutional restricted Boltzmann machine. Background technique [0002] In some specific scenarios, it is necessary to use signal processing technology to analyze the specific expression content of unfamiliar communication signals. To analyze the specific content of the signal, it is necessary to intercept the signal, and then first identify the modulation mode of the intercepted signal before it can be demodulated correctly; as an intermediate process between signal detection and signal demodulation, signal modulation mode identification mainly includes feature extraction and utilization. The classifier classifies in two steps. The traditional identification method of signal modulation mode is mainly based on the identification technology of expert features, which needs artificial feature extraction for...

AI Technical Summary

Problems solved by technology

[0003] In order to solve the problem that in the existing signal modulation method recognition method, the process of feature extraction is too dependent on manual extraction, resulting in unstable performance and poor scalability, the present invention provides a signal based on convolutional restricted Boltzmann machine. The identification method of modulation mode can achieve the purpose of efficiently identifying common modulation modes, which not only reduces the complexity and difficulty of modulation identification, but also has applicability to the identification of many common modulation modes

Images