论文部分内容阅读
讨论一般ＡＲＭＡ噪声中谐波频率的估计问题，提出了一种基于自相关和奇异值分解－总体最小二乘（ＳＶＤ－ＴＬＳ）的预滤波方法．该方法先两次利用ＳＶＤ－ＴＬＳ求解由自相关获得的超定线性方程组得到噪声的ＡＲ系数估计，然后由滤波输出过程的高阶延时自相关来估计谐波频率．由于应用自相关和ＳＶＤ－ＴＬＳ技术，该方法在短数据情形下具有高分辨率、高稳定性和低信噪比的性能，模拟实验说明了这些性质．
To discuss the problem of estimating the harmonic frequency in ARMA noise, a pre-filtering method based on autocorrelation and singular value decomposition-total least squares (SVD-TLS) is proposed. The method first uses SVD-TLS to solve the overdetermined linear equations obtained by autocorrelation to get the AR coefficient estimation of the noise, and then estimates the harmonic frequency by the high-order delay autocorrelation of the filtering output process. Due to the application of autocorrelation and SVD-TLS, this method has the properties of high resolution, high stability and low signal-to-noise ratio in the case of short data. Simulation experiments show these properties.