딥러닝 기술을 이용한 디지털 변조타입 자동 인식 기술

초록

In this paper, we propose a new automatic modulation classification method based on deep neural networks (DNN). The proposed method uses nineteen statistical features extracted from the received signal samples as an input to the fully connected neural networks with the four layers. The deep neural network is trained with the number of 30,000 training data generated by computer simulations. Various signal to noise ratios and fading channel conditions are considered for generation of the training data. The experimental results show that the proposed modulation classification technique outperforms the existing methods both in additive white Gaussian noise(AWGN) and Rician fading channels.

키워드