Higher-order spectra for on-line identification of ANC plants

In the paper a new approach to the on-line identification of electro-Acoustic plants models for active noise control systems, based on the higher-order spectra aided by signal averaging is presented. Two solutions are proposed: first employing the higher-order spectra only and second employing the both higher and second-order spectra. Proposed identification methods allow to reduce an influence of disturbances on identification results and obtain strongly consistent estimates in contrast to classical second-order identification algorithms. Signal averaging enhances signal-to-noise ratio without deteriorating noise attenuation. Results obtained with the proposed identification methods, computed on the basis of data acquired in the laboratory experiments, are provided and compared with results of classical second-order identification methods.