Title: Data-Driven State Estimation for Light-Emitting Diode Underwater Optical Communication

Abstract: Light-Emitting Diodes (LEDs) based underwater optical wireless communications (UOWCs), a technology with low latency and high data rates, have attracted significant importance for underwater robots. However, maintaining a controlled line of sight link between transmitter and receiver is challenging due to the constant movement of the underlying optical platform caused by the dynamic uncertainties of the LED model and vibration effects. Additionally, the alignment angle required for tracking is not directly measured and has to be estimated. Besides, the light scattering propagates beam pulse in water temporally, resulting in time-varying underwater optical links with interference. We address the state estimation problem by designing an LED communication system that provides the angular position and velocity information to overcome the challenges. In this way, we leverage the power of deep learning-based observer design to explore the LED communication's state space properly. Simulation results are presented to illustrate the performance of the data-driven LED state estimation.