Equalization of Sound Reproduction System Based on the Human Perception Characteristics
Qianqian Fang, Huaxing Xu, Risheng Xia, Junfeng Li, Yonghong Yan · 2015
In this paper, we propose an equalization approach for sound reproduction system (e.g., loudspeaker and headphone) based on the human perception mechanism. Traditional FIR-and IIR-based approaches have difficulties in equalizing the low frequencies, the warped filters provide the feasible design for equalization in a non-uniform resolution way and invovle the high complex frequency. To deal with these issues, a novel equalization approach is presented which takes the human perception properties into consideration. Specifically, a weighted-cost function is designed where the weight is calculated based on the perceptual resolution of human beings, which is followed by a modified quais-Newton gradient search algorithm to determine the optimal coefficients of the equalization filter. The performance of the proposed equalization approach is finally compared to those of the state-of-the-art equalizers.