论文信息 - Ambiguity resolution in sparse linear prediction

Ambiguity resolution in sparse linear prediction

We present some results of our analysis of Kumaresan's (1982) sparse linear prediction method for estimation of frequencies of sinusoids. Refinements of Kumaresan's method are proposed for the case of two sinusoids which are not close in frequency. When the data is corrupted by additive white Gaussian noise, the probability of correctly resolving ambiguities is used to evaluate the performance. Comparisons between statistical performance analyses and computer simulations demonstrate that the analyses are accurate.<<ETX>>

[1] R. Kumaresan,et al. Estimation of frequencies of multiple sinusoids: Making linear prediction perform like maximum likelihood , 1982, Proceedings of the IEEE.

[2] G. W. Lank,et al. A Semicoherent Detection and Doppler Estimation Statistic , 1973, IEEE Transactions on Aerospace and Electronic Systems.

[3] T. Ulrych,et al. Time series modeling and maximum entropy , 1976 .

[4] Albert H Nuttall. Scientific and Engineering Studies; Spectral Estimation. , 1977 .

[5] Elizabeth Ralston,et al. Ambiguity Resolution in Interferometry , 1981, IEEE Transactions on Aerospace and Electronic Systems.

[6] Albert H Nuttall,et al. Spectral Analysis of a Univariate Process with Bad Data Points, via Maximum Entropy and Linear Predictive Techniques , 1976 .

[7] Ramdas Kumaresan,et al. Singular value decomposition and spectral analysis , 1981, 1981 20th IEEE Conference on Decision and Control including the Symposium on Adaptive Processes.

[8] R. Kumaresan. Accurate frequency estimation using an all-pole filter with mostly zero coefficients , 1982, Proceedings of the IEEE.