Performance Evaluation of Selective Fixed-filter Active Noise Control based on Different Convolutional Neural Networks

Abstract

Due to its rapid response time and a high degree of robustness, the selective fixed-filter active noise control (SFANC) method appears to be a viable candidate for widespread use in a variety of practical active noise control (ANC) systems. In comparison to conventional fixed-filter ANC methods, SFANC can select from a variety of pre-trained control filters for different types of noise. Deep learning technologies, thus, can be used in SFANC methods to enable a more flexible selection of the most appropriate control filters in the presence of varying incoming noise. Furthermore, deep neural network parameters can be learned automatically from noise data rather than through trial and error, significantly simplifying and improving the practicability of ANC design. Therefore, this paper compared the performance of SFANC using several different neural networks. Two-dimensional and one-dimensional neural networks are used to classify noise types in the frequency and time domains, respectively. Additionally, we compare training from scratch versus fine-tuning training to determine which network training strategy is superior in SFANC methods.