Application of Gaussian Mixture Model to Solve Inter-Symbol Interference in PAM8 Underwater Visible Light System Communication

Abstract

Traditional soft or hard decisions ignoring the correlation between symbols removing linear and nonlinear damage directly may lead to lack of information and result in system performance degradation. In order to reduce the performance degradation of the system caused by the lack of information, we propose a method that use the Gaussian mixed model to consider the influence of the correlation between consecutive symbols for underwater visible light communication system. The experimental results show that the performance of the system is greatly improved after the correlation between consecutive symbols is considered. The more consecutive symbols there are, the greater the performance will be. When the forward error correction (FEC) threshold is met, the Q factor of Gaussian mixture models(GMM) that join three adjacent symbols is 1.19 dB greater than without the GMM clustering algorithm. The highest data rate that can be achieved exceeds 1.5 Gbps with GMM.