Partial Update Adaptive Filtering Based on Head-Related Model for In-Vehicle Audio Enhancement

Abstract

In-vehicle audio reproduction is a challenging task owing to the music, reverberation, reflection, and noise intentionally and unavoidably propagated in the car. For a multichannel loudspeaker system, the equalizer design is essential for reconstructing a nearly perfect sound field in the reverberation room. In this work, we propose a partial update adaptive algorithm based on the least-mean-square (LMS) approach and head-related transfer functions (HRTFs) to efficiently repress the indirect terms and separate the different timbres resulting from the in-vehicle multichannel playback system. Since HRTFs have played a crucial role in the spatial audio signal processing, introducing this model is shown to facilitate the complexity reduction for a highly nonlinear system. As a result, several examples of numerical design are provided to verify the characteristics of the proposed system.