摘要
本文首次把近代谱估计方法引入到互谱估计中，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参数谱估计的矩估计方法和Ｌｅｖｉｎｓｏｎ递推估计方法。该方法可以有效地克服传统的互谱ＦＦＴ算法和互周期图法存在的谱分辩率低、谱估计方差大等缺点。文中还给出了信噪比为－３０ｄＢ的正弦信号参数估计的仿真实例。
This paper introduces the modern spectral estimation method in to cross-spectral estimation,the cross-correlation-based Yule-Walker equation is proposed. Based on this contribution, the Moment estimation method and Levinsion recursive algorithm are developed. The new algorithms can overcome efficiently the shortcomings of low resolution and high variance which are existing in both FFT and Cross-periodogram methods. Finally, we compare the proposed algorithm with the well-known methods, Pisarenko and SVD methods, by using the parameter estimation results of sinusoids in strong measurement noise (SNR = -30dB) .
出处
《电子学报》
EI
CAS
CSCD
北大核心
1994年第1期1-8,共8页
Acta Electronica Sinica
基金
国家教委博士基金
关键词
功率谱估计
弱信号
测量
噪声测量
Cross-spectral estimation, Measurement noise, Weak signal mesurement, Levinsion algorithm
