【摘要】 为了提高自适应滤波器的实时处理速度,提高算法的稳定性。在此采用最新的DSP Builder软件,在FPGA上实现了解相关归一化LMS(DNLMS)算法。该方法通过在Matlab/Simulink中多次仿真都达到了很好的效果,有效地消除了噪声,运行稳定。在此引进了解相关原理和归一化处理,加快了算法的收敛速度,且稳态误差小。如果将该滤波器从8阶增加到512阶或者1 024阶,可以很好地实现在话音通信中的回声消除等自适应滤波。更多还原

【Abstract】 In order to improve the real-time processing speed of the adaptive filter and the algorithm stability,the latest DSP Builder software is adopted to realize uncorrelated normalized LMS(DNLMS) algorithm based on FPGA.This method is proved to reach good effect by simulation in Matlab/Simulink.The uncorrelated principle and the process of normalization is substituted,the convergence speed could be quickened and the misadjustment could be smaller.If this adaptive is added from 8 stages to 512 stages or 1 024 stages,noise elimination in speech processing or other-adaptive filter is well realized.更多还原