Double-talk robust VSS-NLMS algorithm for under-modeling acoustic echo cancellation

In this paper, we proposed a variable step-size normalized least-mean-square (VSS-NLMS) algorithm in the context of acoustic echo cancellation (AEC), taking into account the under-modeling and double-talk scenarios simultaneously. The recursive relation for updating the tap-weight vector was derived, and the robustness of the algorithm was analyzed. The proposed algorithm requires neither any addition information about the acoustic environment nor a separate double-talk detector and has low computational complexity and low level data storage. So, it´s very practical for real-world AEC applications. The algorithm proposed has good convergence speed and excellent double-talk immunity. The simulation results demonstrate the good performance of the proposed algorithm.