Design of DBN based Demodulator in Underwater Wireless Optical Communications
2020 IEEE/CIC International Conference on Communications in China (ICCC Workshops)(2020)
摘要
In this work, we present a machine learning based demodulation method for underwater wireless optical communication (UWOC) systems. Specifically, the demodulator is designed based on deep belief network (DBN) network. The DBN demodulator extracts features layer-by-layer through three restricted Boltzmann machines. To make the method more practical, we build a flexible end-to-end UWOC experimental platform to obtain a real data set of signals from ten different modulation methods. The extensive experiments are conducted and our results show that the DBN demodulator enjoys a better performance than other benchmarks.
更多查看译文
关键词
Underwater wireless optical communication,machine learning,demodulation,DBN based demodulator
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络