Brownian-Bridge Integrated Kramers-Kronig Framework for Orbital Angular Momentum Mode Demultiplexing and Demodulation in Turbulence

Multiplexing through the orbital angular momentum (OAM) of light offers a significant enhancement in the capacity of free-space optical (FSO) links, addressing the growing demand for high data rates [1]. However, atmospheric turbulence poses a persistent challenge, introducing signal distortion and crosstalk that degrade link performance. In this research, we employ a Brownian-Bridge Neural Network-based Denoising Diffusion Probabilistic Model (BB-DDPM) for turbulence mitigation, integrated with a Kramers-Kronig (KK) framework for phase-amplitude retrieval [2]. This approach effectively reduces the error vector magnitude (EVM), enhances demodulation accuracy, and ensures robust communication under various turbulence conditions.