Title :
A fundamental algorithm for the power estimation of closely separated spectral sources
Author :
Silverstein, Seth D.
Author_Institution :
GE Corporate Res. & Dev., Schenectady, NY, USA
Abstract :
A new algorithm is based upon division of the eigenstructure of the sample covariance into approximate signal and noise subspaces due to fluctuations caused by finite data samples. These fluctuation effects are calculated using stochastic perturbation theoretic techniques. These results show that the moments of the MUSIC null spectrum can be approximated by linear functionals of the source SNR and the number of snapshots. All theoretical predictions are in accord with the simulation results. A featured simulation demonstrates accurate source power estimates for three sources separated by sub-Rayleigh resolution spatial frequencies with a weak source of 0 dB SNR sandwiched between two much larger sources of 40 and 20 dB
Keywords :
array signal processing; eigenvalues and eigenfunctions; fluctuations; functional equations; parameter estimation; perturbation techniques; stochastic processes; closely separated spectral sources; eigenstructure; fluctuations; fundamental algorithm; linear functionals; power estimation; sample covariance; simulation; stochastic perturbation theoretic techniques; sub-Rayleigh resolution spatial frequencies; Covariance matrix; Direction of arrival estimation; Fluctuations; Frequency estimation; Multiple signal classification; Phased arrays; Research and development; Signal to noise ratio; Spatial resolution; Vectors;
Conference_Titel :
Statistical Signal and Array Processing, 1992. Conference Proceedings., IEEE Sixth SP Workshop on
Conference_Location :
Victoria, BC
Print_ISBN :
0-7803-0508-6
DOI :
10.1109/SSAP.1992.246874
