Channel Modeling and Intelligent Anti-Interference in Underwater Visible Light Communication

Abstract

The paper discusses the latest developments in underwater visible light communication (UVLC), with an emphasis on channel modeling and intelligent anti-interference algorithms. It initially points out the shortcomings of classical Beer-Lambert-based models, which assume stationary and uniform water conditions and neglect important effects such as turbulence, temperature stratification, salinity variations, and suspended matter. It then summarizes developments in data-driven multi-physical-field refined modeling, which uses measured oceanographic data, turbulence effects, chlorophyll concentration, and turbidity to better reflect real marine environments. Next, it traces the evolution of anti-interference methods from traditional optimization to intelligent regulation using AI, such as spatial diversity, modulation and coding optimization, deep learning to identify interferences and nonlinear compensation, and adaptive transmission reinforcement learning. Lastly, major challenges and future trends are addressed, including physics-data hybrid modeling, AI-enabled interference suppression, integration and interdisciplinary innovation, with 6G space-air-ground-sea networks. This review is a concise source of UVLC research and engineering.which gives a good overview of what has been achieved to date.