Underwater vortex optical multiplexing communication demodulation based on random forest algorithm

Orbital Angular Momentum (OAM) enables mode division multiplexing due to the orthogonality of its different mods, which enhances channel capacity. However, this multiplexing approach can lead to mode crosstalk, degrading the performance of the communication system. To address this issue, this paper proposes a demodulation scheme based on the Random Forest (RF) algorithm. Compared to existing equalization methods, this approach does not require channel estimation and can handle nonlinear distortions. In this paper, numerical simulations are employed to model transmission experiments in a marine turbulence environment. The RF algorithm is compared with the Minimum Mean Square Error (MMSE) algorithm to analyze their error performance in communication systems under varying turbulence intensities and transmission distances. The results indicate that the RF demodulation algorithm significantly improves the error performance of the communication system, surpassing the MMSE algorithm. These findings offer theoretical insights for the development of underwater vortex beam multiplexing communication demodulation technologies.