The CFAR adaptive subspace detector is a scale-invariant GLRT

The CFAR adaptive subspace detector is a scale-invariant GLRT Kraut, Shawn ; Scharf, Louis L. "This work was supported by the Office of Naval Research under Contract N00014-89-J-1070 and by the National Science Foundation under Contract MIP-9529050." The constant false alarm rate (CFAR) matched subspace detector (CFAR MSD) is the uniformly most-powerful-invariant test and the generalized likelihood ratio test (GLRT) for detecting a target signal in noise whose covariance structure is known but whose level is unknown. Recently, the CFAR adaptive subspace detector (CFAR ASD), or adaptive coherence estimator (ACE), was proposed for detecting a target signal in noise whose covariance structure and level are both unknown and whose covariance structure is estimated with a sample covariance matrix based on training data. We show here that the CFAR ASD is GLRT when the test measurement is not constrained to have the same noise level as the training data. As a consequence, this GLRT is invariant to a more general scaling condition on the test and training data than the well-known GLRT of Kelly. Colorado State University. Libraries 1999 text ; image application/pdf ECElls00015.pdf FACFECEN100404ARTI eng c1999 IEEE

The CFAR adaptive subspace detector is a scale-invariant GLRT

Kraut, Shawn ; Scharf, Louis L.

"This work was supported by the Office of Naval Research under Contract N00014-89-J-1070 and by the National Science Foundation under Contract MIP-9529050."

The constant false alarm rate (CFAR) matched subspace detector (CFAR MSD) is the uniformly most-powerful-invariant test and the generalized likelihood ratio test (GLRT) for detecting a target signal in noise whose covariance structure is known but whose level is unknown. Recently, the CFAR adaptive subspace detector (CFAR ASD), or adaptive coherence estimator (ACE), was proposed for detecting a target signal in noise whose covariance structure and level are both unknown and whose covariance structure is estimated with a sample covariance matrix based on training data. We show here that the CFAR ASD is GLRT when the test measurement is not constrained to have the same noise level as the training data. As a consequence, this GLRT is invariant to a more general scaling condition on the test and training data than the well-known GLRT of Kelly.

Colorado State University. Libraries

1999

text ; image

application/pdf

ECElls00015.pdf

FACFECEN100404ARTI

eng

c1999 IEEE