A Deep Convolutional Network Demodulator for Mixed Signals with Different Modulation Types

Xuming Lin, Ruifang Liu, Wenmei Hu, Yameng Li, Xin Zhou, Xiao-Xin He

Abstract: In recent years, deep learning is becoming more and more popular. It has been widely applied to fields including image recognition, automatic speech recognition and natural language processing(NLP). In the field of communication, signals are considered to be temporal data, which can be learned with deep learning to recognize its patterns inside. In this paper, a Deep Convolutional Network Demodulator (DCND) is proposed. This model attempts to respectively demodulate symbol sequences from mixing signals. The data composes of signals modulated with different signal modulation techniques and the same carrier frequency for the purpose of our project. In this condition, the proposed model can give contribution to reduce the bit error ratio(BER), demodulate signals successfully which cannot be recognized by correlation demodulation for additive white Gaussian noise(AWGN), and resistance to frequency interference.

External IDs:dblp:conf/dasc/LinLHLZH17