Generalized adaptive comb filter with improved accuracy and robustness properties

Abstract

Generalized adaptive comb filters can be used to identify/track parameters of quasi-periodically varying systems. In a special, signal case they reduce down to adaptive comb filters, applied to elimination or extraction of nonstationary multi-harmonic signals buried in noise. We propose a new algorithm which combines, in an adaptive way, results yielded by several, simultaneously working generalized adaptive comb filters. Due to its highly parallel estimation structure, the new algorithm is more accurate and more robust than the currently available algorithms.