Modulation pattern recognition method and system based on CNN and combined high-order spectral images

Abstract

The invention discloses a modulation pattern recognition method based on CNN and combining multiple high-order spectral feature images, including: receiving a radio frequency signal from a signal source, performing analog-to-digital conversion on the radio frequency signal to obtain a digital signal, and converting the digital signal to The signal is digitally down-converted and filtered successively to obtain I / Q data, and the I / Q data is preprocessed to obtain a combined high-order spectrum image, and the signal-to-noise ratio of the obtained I / Q data is obtained, and the signal-to-noise ratio is judged. Whether the noise ratio is greater than or equal to the preset threshold, and if so, input the obtained combined high-order spectral image into the trained first convolutional neural network model to obtain a modulation pattern recognition result. In view of the large difference in the high-order spectral characteristics of the signal under different signal-to-noise ratios, the present invention trains two convolutional neural network models, which improves the robustness and can be used for BPSK, QPSK, 8PSK, 16APSK, 32APSK, Modulation pattern identification for 16QAM, 32QAM and other signals.

Description

technical field [0001] The invention belongs to the technical field of wireless communication and machine learning, and more particularly, relates to a modulation pattern recognition method and system based on CNN (Convolutional neutral network, CNN for short) and combined high-order spectral images. Background technique [0002] In recent years, with the advancement of communication technology and signal processing technology, there are more and more types of radio signals, and the modulation pattern has become more and more complex. It is difficult to deal with complex radio signals using traditional modulation pattern recognition methods, and the accuracy of traditional modulation pattern recognition relies heavily on manual extraction of signal features. And with the wide application of mobile communication technology, the communication environment is full of various radio signals, and at the same time there are various noises and interferences, which poses many technica...

AI Technical Summary

Problems solved by technology

[0010] Aiming at the above defects or improvement needs of the prior art, the present invention provides a modulation pattern recognition method and system based on CNN and combined high-order spectral images. Prior knowledge, and the technical problems of high delay in the classification process, as well as the technical problems of considerable algorithm changes due to the need to consider cumbersome judgment conditions and threshold values ​​when new signals appear, and the existing The recognition method based on deep learning requires high image pixels, which leads to technical problems such as large amount of calculation, slow recognition speed, and high system delay, and because partly depends on the constellation diagram of the signal, when the signal has a frequency offset, the modulation The pattern recognition rate is low, which further leads to the technical problem that the method cannot be applied in the field of non-cooperative communication

Images