FXLMS Algorithm-based Feedforward Active Noise Control System Design

Active noise control is a way to offset the original noise by artificial creation of secondary noise. Robust stability and computational complexity is the main problem facing in an active noise control system in practical applications. In order to improve its robust stability and noise attenuation under the conditions of a reasonable computational complexity, an in-depth analysis of the methods of Eriksson, Zhang, Akhtar and Davari, who proposed a common active noise control system based on FXLMS algorithm, is given in this paper, then the structure of the system is researched, and on this basis, a new method of active noise control is proposed. By introducing the dynamic adjustment factor in the step iterative process of the control filter and the modeling filter, This method improves the convergence rate of the control filter and the modeling filter, Better noise attenuation is implemented at a reasonable computational complexity conditions. In broadband, narrowband and interference conditions, the proposed method is comprehensively compared with the above four methods in simulation and performance. The results are presented to show the robustness and effectiveness of the proposed method. The system achieves a higher overall performance.