APSIPA Transactions on Signal and Information Processing > Vol 14 > Issue 1

Spatial Active Noise Control Based on Kernel Interpolation With Individual Directional Weighting

Kazuyuki Arikawa, The University of Tokyo, Japan, Shoichi Koyama, National Institute of Informatics, Japan, koyama.shoichi@ieee.org , Hiroshi Saruwatari, The University of Tokyo, Japan
 
Suggested Citation
Kazuyuki Arikawa, Shoichi Koyama and Hiroshi Saruwatari (2025), "Spatial Active Noise Control Based on Kernel Interpolation With Individual Directional Weighting", APSIPA Transactions on Signal and Information Processing: Vol. 14: No. 1, e28. http://dx.doi.org/10.1561/116.20250034

Publication Date: 09 Oct 2025
© 2025 K. Arikawa, S. Koyama and H. Saruwatari
 
Subjects
Adaptive signal processing,  Audio signal processing,  Sensor and multiple source signal processing,  Statistical/Machine learning,  Kernel methods,  Adaptive control and signal processing
 
Keywords
Adaptive filterspatial active noise controlkernel ridge regressioninterpolation
 

Share

Open Access

This is published under the terms of CC BY-NC.

Downloaded: 4 times

In this article:
Introduction 
Problem Formulation 
Spatial ANC Based on Kernel Interpolation of Totak Sound Field 
Spatial ANC Based on Individual Kernel Interpolation 
Experiments 
Conclusion 
References 

Abstract

An active noise control (ANC) method for reducing noise over a region in space based on the individual kernel interpolation of primary and secondary sound fields is proposed. The kernel-interpolation-based spatial ANC based on the estimation and synthesis of a three-dimensional sound field using multiple microphones and secondary sources has practical advantages in an arbitrary array configuration and efficient time-domain algorithms. It is possible to incorporate prior information on primary noise source directions into the kernel function by directional weighting to enhance the estimation accuracy of a sound field; however, in a previous study, the performance improvement by this directional weighting is limited owing to the mismatch of the source direction in the sound field generated by secondary sources. We propose a spatial ANC method based on the individual interpolation of the sound fields generated by primary and secondary sources. An adaptive filtering algorithm for individual kernel interpolation is also developed. The experimental results indicate that the proposed method outperforms the previous kernel-interpolation-based spatial ANC.

DOI:10.1561/116.20250034