Recursive Non-Data-Aided SNR Estimation for Coherent Free-Space Optical Systems

We evaluate recursive and non-data-aided (NDA) signal-to-noise ratio (SNR) estimators for coherent free-space optical (FSO) systems. By replacing the block-wise moment accumulations of the M2M4 and skewness-corrected signal-to-noise variance (SNV) estimators with exponentially weighted recursive updates, the proposed framework achieves symbol-rate SNR tracking without requiring large buffers. Simulation validations confirm successful SNR tracking and experimental validations confirm rapid convergence under real hardware impairments.