奇异值分解降噪的改进方法

利用测试信号构造Hankel矩阵进行奇异值分解(SVD)是消除随机噪声干扰的有效方法,其关键是奇异值数目的选取,但目前尚无成熟有效的确定方法。针对这一问题,提出了一种奇异值分解降噪的改进方法,该方法依据去噪后信号极值点数量随奇异值数目变化的关系,可以准确选取与最优降噪效果对应的奇异值数目。仿真及实验结果表明,该方法准确、有效。利用该方法处理船舶和机械设备振动噪声测试信号,可有效提高其信噪比,最大程度地优化信号去噪的效果,提高分析的可靠性。

An Improved Method for Noise Reduction Based on Singular Value Decomposition

Using Singular Value Decomposition（SVD）of constructed Hankel matrix by measured signal is an effective method for eliminating the random noise.The key is to choose the rank of the Hankel ma trix,but there is no mature and effective method now.An improved method was proposed according to the relationship between the extremum points of the noise elimination signal and the rank of the matrix.It wass easy to choose the rank of the matrix which leads to the best noise elimination result.Simulation and experi ment validated this method.The noise and vibration signal measured on ship and mechanical equipment could be refined through this method to improve the signal to noise ratio,optimize the noise elimination ef fect as much as possible,and the analysis which depends on the signal is more reliable.