On Algorithms And Implementations Of A 4-Channel Active Noise Canceling Window

Chuang Shi,Nan Jiang,Huiyong Li,Dong-Yuan Shi,Woon-Seng Gan
DOI: https://doi.org/10.1109/ISPACS.2017.8266476
2017-01-01
Abstract:The active noise canceling window is an application of the multi-channel active noise control (ANC) system, which aims to provide a quiet living environment while preserving the ventilation of the room. In the implementation of the active noise canceling window, the standard multiple-error LMS algorithm demands too much computational power to be handled by a common digital signal processor (DSP). Hence, the minimax algorithm, which minimize the amplitude instead of the power of the noise field, is revisited. The minimax algorithm is compared with the multiple-error LMS algorithm based on numerical simulations of a case (1, 4, 4) ANC system and real-time experiments with floating-point implementations. The simulation and experiment results show that the minimax algorithm has a distinct advantage in reducing the computational complexity but its trade-off is the relatively slow convergence speed.
What problem does this paper attempt to address?