【摘要】 本文从理论上证明了互相关函数的Ｙｕｌｅ－Ｗａｌｋｅｒ方程，并在此基础上提出了互谱ＡＲ模型参数估计的Ｌｅｖｉｎｓｏｎ递推算法。该算法可有效地克服传统互谱估计的ＦＦＴ方法和周期图法存在的谱分辩率低、谱估计方差大等缺点。仿真试验表明，该方法还可有效地克服有色观测噪声对测量精度的影响。更多还原

【Abstract】 The Yule-Walker equation based on cross-correlation is proposed theoretically, and the Levinson algorithm is developed. The algorithm can efficiently overcome the shortcomings of the classical FFT and cross-periodogram methods for cross-spectral estimation. Finally, the method presented in this paper is compared with the well-known Cadzow’s SVD method by a numerical example.更多还原