페이딩 환경에서의 딥러닝 기반 고성능 자동 변조분류 기법

초록

In this paper, we propose a deep learning-based method for automatically classifying modulation formats in wireless communication systems. While existing automatic modulation schemes are mostly designed for Gaussian channels, these techniques tend not to work well in fading environments. The proposed method extracts various kinds of statistical feature values ​​from the data and classifies the modulation class using the deep neural network consisting of fully connected layers. In order to apply the proposed automatic modulation classification scheme in the fading channel, the training data is generated considering the fading environment and the deep neural network is trained by using it. As a result of applying the proposed method to the five kinds of modulation classifications of BPSK, QPSK, 8-PSK, 16-QAM and 64-QAM, we obtained better results in terms of classification accuracy than the existing methods in the fading environment.

키워드