论文部分内容阅读
在理想情况下，平稳白噪声在某一循环频率α上的循环自相关函数为零，因此对于具有循环平稳特性的信号利用循环自相关函数取代传统的自相关函数可以有效地抑制噪声。但在实际场合中由于数据长度有限，噪声循环自相关函数的估计量并不为零，因此将影响循环平稳方法的性能。本文推导了有限长数据下噪声循环自相关函数与数据长度之间的量化关系，给出了相应的物理解释，并通过Ｍｏｎｔｅ－Ｃａｒｌｏ实验验证了有关结论。
Under ideal conditions, the cyclic autocorrelation function of stationary white noise at a certain cyclic frequency α is zero. Therefore, using the cyclic autocorrelation function to replace the traditional autocorrelation function can effectively suppress the noise for the signal with cyclostationarity. However, due to limited data length in practical situations, the estimation of the autocorrelation function of the noise cycle is not zero, which will affect the performance of the cyclostationary method. In this paper, the quantitative relation between the noise cycle autocorrelation function and the data length is deduced under the limited length of data, and the corresponding physical explanation is given. The relevant conclusions are verified by Monte-Carlo experiments.