Weighted Regularized Sparse Recovery Method for Optical Power Monitoring

Monitoring optical power from the broadened optical spectrum acquired by scanning a low-cost tunable optical filter (TOF) is a practical and challenging task for the management of wavelength-division-multiplexing (WDM) networks. Existing methods, such as least squares (LS) and regularized sparse recovery (RSR) methods, suffer from performance deterioration when the power discrepancies between adjacent channels are significant. In this letter, we formulate it as a maximum a posteriori (MAP) estimation problem under Bayesian framework. A weighted-RSR algorithm is, therefore, proposed to estimate the optimal channel power. Experimental results demonstrate that, using commercial low-cost TOFs, the weighted-RSR can accurately monitor the power of mixed 10-G/40-G WDM channels with 50-GHz channel spacing and 10-dB power difference.