Optimal and self-tuning deconvolution in time domain (Q2723866)

Both the optimal and self-tuning deconvolution problems for discrete signal models using a time domain innovation analysis approach are considered. When the signal model, measurement model and noise statistics are known, a novel approach for the design of optimal deconvolution filter, predictor, and smoother is proposed based on projection theory and innovation analysis in the time domain. This approach unifies the treatment of both stable and unstable signal models and simplifies the calculation of estimates. NEWLINENEWLINENEWLINEWhen the noise statistics, input model and colored noise model are unknown, the self-tuning deconvolution estimators are designed by identifying two autoregressive moving average innovation models: one for the signal plus noise and the other for noise alone. An illustrative example is presented to demonstrate the design of deconvolution estimators and their performance.