Title :
A resolution threshold for multidimensional parameter estimates
Author :
Clark, Michael P.
Author_Institution :
AFIT/ENG, Air Force Inst. of Technol., Wright-Patterson AFB, OH, USA
fDate :
31 Oct-2 Nov 1994
Abstract :
This paper presents a metric useful in analyzing the resolution of multidimensional parameter estimators which produce normal, unbiased estimates. The metric is based on the distance between ellipsoidal level curves of the individual estimates´ densities. The resolution threshold is deemed to be the lowest SNR at which ellipsoids enclosing a given amount of probability mass have disjoint interiors. This paper develops an algorithm which computes the threshold by studying the problem of finding the distance between two ellipsoids
Keywords :
parameter estimation; signal processing; spectral analysis; SNR; algorithm; disjoint interiors; ellipsoidal level curves; multidimensional parameter estimates; probability mass; resolution threshold; unbiased estimates; Chromium; Covariance matrix; Density functional theory; Ellipsoids; Frequency estimation; Multidimensional systems; Noise level; Parameter estimation; Signal resolution; Signal to noise ratio;
Conference_Titel :
Signals, Systems and Computers, 1994. 1994 Conference Record of the Twenty-Eighth Asilomar Conference on
Conference_Location :
Pacific Grove, CA
Print_ISBN :
0-8186-6405-3
DOI :
10.1109/ACSSC.1994.471407
