论文部分内容阅读
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。
In order to improve the real-time processing speed of adaptive filter and improve the stability of the algorithm. This is where the latest DSP Builder software is used to understand the relevant normalized LMS (DNLMS) algorithms on the FPGA. The method achieves good results through multiple simulations in Matlab / Simulink, effectively eliminating the noise and running steadily. Introduced here to understand the relevant principles and normalized to speed up the convergence rate of the algorithm, and the steady-state error is small. If this filter is increased from 8th order to 512th order or 1 024 orders, adaptive filtering such as echo cancellation in voice communication can be well achieved.