Matched Subspace Detectors英文清晰版.pdfVIP

2146 IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL. 42, NO. 8, AUGUST 1994 Matched Subspace Detectors Louis L. Scharf, Fellow, IEEE, and Benjamin Friedlander, Fellow, IEEE Abstract-In this paper we formulate a general class of prob- to include subspace interferences. These problems involve lems for detecting subspace signals in subspace interference and unknown parameters in the mean and covariance of a mul- broadband noise. We derive the generalized likelihood ratio tivariate normal (MVN) distribution. For each problem in the (GLR) for each problem in the class. We then establish the invariances for the GLR and argue that these are the natural class, we establish invariances for the GLR and find that they invariances for the problem. In each case, the GLR is a maximal are identical to the natural invariances for the problem. We invariant statistic, and the distribution of the maximal invariant show that a monotone function of the GLRT equals one of statistic is monotone. This means that the GLR test (GLRT) is the uniformly most powerful invariant (UMP-invariant) tests the uniformly most powerful invariant detector. We illustrate the derived in 111. This means that the GLRT is itself UMP- utility of this finding by solving a number of problems for de- invariant. In addition to tying up the theories of invariance and tecting subspace signals in subspace interference and broadband noise. In each case we give the distribution for the detector and the GLRT, our results generalize and extend previous wor