一种基于循环小波累积量的谐波恢复方法

Wavelet Cyclic Cumulants Based Approach to Harmonic Retrieval

循环统计量是乘性和加性噪声中谐波恢复问题的主要分析工具之一。本文在其基础上提出了一种新的理论与方法——基于循环小波累积量的谐波恢复方法。首先,提出了循环小波累积量的定义。然后,建立了基于一阶、二阶、三阶循环小波累积量的谐波分量的频率和相位的估计方法。在此基础上,采用了一种可调参数的小波,通过调节其参数来改善传统方法中频率分辨率低的缺陷。最后给出了模拟实验结果。

Based on cyclic statistics ,this paper proposes a new scheme,namely wavelet cyclic cumulants,for harmonic retrieval in multiplicative and additive noise. At first,the definition of wavelet cyclic cumulants is proposed. Then,the estimation for harmonics pa- rameters based on the kth-order wavelet cyclic cumulants are presented for k = 1,2,3. The present approach improves the low resolution rate of tradtional method by using a wavelet. Finally,the simulation results are given.