Title :
UMP invariance of the multirank adaptive coherence estimator
Author :
Kraut, Shawn ; Scharf, Louis L.
Author_Institution :
Dept. of Math. & Stat., Kingston, Ont., Canada
Abstract :
The adaptive coherence estimator, or ACE, is a uniformly most powerful (UMP)-invariant detection statistic is presented in this paper. The proof that ACE is a one-dimensional maximal invariant was based on a relatively simple extension of an earlier work done by the authors on maximal invariant statistics for detection of rank-1 signals. In this paper we consider the multirank version of ACE, needed for scenarios when the signal waveform is uncertain. It also can be shown to be a maximal invariant, and by extension UMP-invariant, but the proof requires a more detailed analysis of the transformation group properties.
Keywords :
adaptive estimation; adaptive signal detection; coherence; covariance matrices; statistics; covariance matrices; multirank adaptive coherence estimator; rank-1 signal detection; signal waveform; transformation group properties; uniformly most powerful-invariant detection statistic; Clutter; Covariance matrix; Mathematics; Probability; Signal analysis; Statistical analysis; Statistics; Testing; Training data; Uncertainty;
Conference_Titel :
Signals, Systems and Computers, 2004. Conference Record of the Thirty-Seventh Asilomar Conference on
Print_ISBN :
0-7803-8104-1
DOI :
10.1109/ACSSC.2003.1292305