基于最小均方算法的破片测速信号处理方法

Signal Processing of Fragments Velocity Measurement Based on Least Mean Square Algorithm

针对激光光幕战斗部破片测速中信号噪声大的问题,提出了基于最小均方(LMS)算法的破片测速信号处理方法。该方法将变步长与变换域结合,变步长LMS算法通过合理调整步长因子有效解决了传统LMS算法收敛速度和稳态误差之间的矛盾;变换域LMS算法通过多尺度小波变换降低输入信号矩阵相关性,提高传统LMS算法的收敛速度。仿真验证表明,新算法收敛速度加快35.04%,稳态误差改善27.12%,信噪比差值提高0.72倍,测速精度提高了0.54%。

n view of the problems of the big noise in the laser screen warhead fragments velocity testing signal, a method of signal processing based on the least mean square （LM.S） algorithm was propose. This method combined with the varia- ble step size and the transform domain. The contradiction between convergence rate and steady-state error of traditional LMS algorithm was effectively solved by variable step size LMS algorithm through adjusting the step size factor reasona- ble. Transform domain LMS algorithm could reduce the correlation of the input signals through multi-scale wavelet trans- form, so the convergence rate would be improved. Simulation results showed that the convergence rate was improved to 35. 04% and the steady-state error was improved to 27. 12%,the difference between the was improved as 0. 72 times as be- fore. The accuracy of the velocity measurernent could be improved 0. 54 times than the traditional algorithm of LMS.