Design and Simulation of Active Noise Cancelling Earphone System Based on FXLMS Algorithm

Abstract

This paper expounds the fundamentals of active noise control and the adaptive feedforward active noise control system based on the FXLMS algorithm, and then uses MATLAB/Simulink to design and simulate the algorithm to obtain a good noise reduction effect. On this basis, a low-pass filter module is added after the residual signal to optimize the model to further reduce the noise. The final simulation results are: when the running time is 10s, the order of adaptive filter is 8, the step factor is 0.5, and the standard deviation of noise passing through the primary path of active noise reduction model is 0.04797, the standard deviation of residual noise is 0.000778, the residual noise of the optimized model is 0.0003826. This shows that the adaptive feedforward active control system based on the FXLMS algorithm is effective in dealing with noise, and it is correct and feasible to add a low-pass filter after the algorithm. In addition, the order and iteration step of the filter are essential to the stability and convergence performance of the algorithm. However, parameter design of FXLMS algorithm will inevitably require a compromise between a system’s performance and stability. As the filter order and the convergence factor increase, so do the speed of system convergence. At the same time, the system may diverge.