An adaptive observer-based estimator for multi-sinusoidal signals

This paper deals with a novel robust estimation methodology yielding the amplitudes, frequencies and phases of the components of a biased multi-sinusoidal signal in presence of a bounded disturbance on the measurement. The proposed method is based on a suitable adaptive observer in which the parameters´ adaptation law is equipped with an excitation-based switching logic. The stability analysis shows the existence of a set of tuning parameter guaranteeing that the estimator´s dynamics is input-to-state stable with respect to bounded measurement disturbances. The effectiveness of the algorithm is illustrated by some simulation examples also reporting a few comparison results.