Abstract:
In recent years a lot of work has been done in the field of Active Noise Cancellation (ANC) for online estimation of the secondary path. Although vigorous research has been carried out but mostly for Gaussian noise hence in this thesis online estimation of secondary path has been done for the Impulsive Noise with different values of to make the environment more impulsive. Previously most of the techniques presented were for stationary acoustic paths and in practical environment noise is non-stationary hence we carried out simulations for non-stationary environment. Most researchers have worked on Fx-LMS algorithm which is widely used in the field of ANC and then different variants of Fx-LMS have been proposed in past but in this paper we have proposed Filtered x Least Mean Absolute Third (FxLMAT) algorithm for online estimation of the secondary path and further improvements have been added to existing Fx-LMAT algorithm by suggesting two more algorithms i.e. Variable Step-Size Filtered x Absolute Third (VSSFxLMAT) and Variable Step-Size Filtered x Robust Normalized Absolute Third (VSSFxRNLMAT) . Proposed Algorithms outperforms the existing algorithms.
