Robust Channel Estimation for Optical Wireless Communications Using Neural Network

Optical Wireless Communication (OWC) has gained significant attention due to its high-speed data transmission and throughput. Optical wireless channels are often assumed to be flat, but this paper considers very dispersive optical wireless environments resulting in frequency-selective scenarios. To address this, this paper presents a robust and low-complexity channel estimation framework to mitigate frequency-selective effects, then to improve system reliability. This channel estimation framework contains a neural network with strong generalization to provide estimated information about the environment. Based on this estimate and the corresponding delay spread, the selector will activate one of several candidate neural networks to precisely predict this channel. Simulation results demonstrate that the proposed method has improved and robust normalized mean square error (NMSE) and bit error rate (BER) performance in dynamic environments. These findings highlight the potential of extending high-data rate and reliable communications under indoor multi-tap optical channels.