Geometries, invariances, and SNR interpretations of matched and adaptive subspace detectors

Geometries, invariances, and SNR interpretations of matched and adaptive subspace detectors

Géométries, invariances et interprétations par le rapport signal à bruit de détecteurs en sous-espaces adaptés ou adaptatifs

Louis L. Scharf Shawn Kraut 

Department of Electrical and Computer Engineering University of Colorado, Boulder, Colorado 80309-0425, USA

1 April 1998
31 December 1998
| Citation



Matched subspace detectors generalize the matched filter by accommodating signals that are only constrained to lie in a multidimensional subspace. There are four of these detectors, depending upon knowledge of signal phase and noise power. The adaptive subspace detectors generalize the matched subspace detectors by accommodating problems where the noise covariance matrix is unknow, and must be estimated from training data . In this paper we review the geometries and invariances of the matched and adaptive subspace detectors. We also establish that every version of a matched or adaptative subspace detectors can be interpreted as an estimator of output signal-to-noise ratio (SNR), in disquise.

1. Introduction
2. Matched Subspace Detectors: Geometries And Invariances
3. Matched Subspace Detectors: SNR Interpretations
4. Adaptive Subspace Detectors: Geometries And Invarlances
5. Adaptive Subspace Detectors: SNR Interpretations
6. Conclusions

[1] B . Picinbono and G . Vezzosi, «Détection d'un signal certain dans un bruit non stationnaire et non gaussien», Ann . Telecommun . vol . 25 pp . 433-439, 1970 .

[2] L .L. Scharf and D . W. Lytle, "Signal detection in gaussian, noise of unknown level : An invariance application", IEEE Trans . Information Theory, vol . 17 pp . 401-411, July 1971 .

[3] G . Vezzozi and B . Picinbono, «Détection d'un signal certain dans un bruit sphériquement invariant, structure et caractéristiques des récepteurs», Ann. Telecommun., vol . 27, pp . 95-110, 1972 .

[4] B . Picinbono, «General detection and estimation theory in an adaptive context», in, Underwater Acoustics and Signal Processing, NATO Advanced Studies Institute Series, Leif Bjorno, Ed . D . Reidel Publishing, Holland, 1980 .

[5] L .L . Scharf, Statistical Signal Processing, Chapter 4, Addison-Wesley, 1991 .

[6] L.L. Schar and B. Friedlander, "Matched subspace detectors", [EE Trans. Signal Processing vol. 42, no. 8, pp. 2146-2157, 1994.

[7] S . Kraut, L .L . Scharf, and L .T. Mc .Whorter, "Matched and adaptive subspace detectors for radar, sonar, and data communication", IEEE Trans. Signal Processing, submitted for review, July 1998 .

[8] E . Conte, M . Lops, and G . Ricci, "Asymptotically optimum radar detection in compound-gaussian clutter", IEEE Trans . On Aerospace and Electronic Systems, Vol . 31, no .2, pp . 617-625, 1995 .

[9] E . Conte, M. Lops and G . Ricci, "Adaptive matched filter detection in spherically invariant noise", IEEE Signal Processing Letters, vol. 3, no. 8, ppp . 248-250, 1996.

[10] B .C.Picinbono, "A geometrical interpretation of signal detection and estimation", IEEE Trans, Information Theory, vol . 26, no 4, pp . 493-497, July 1980.

[11] S . Kraut and L .L. Scharf, "The CFAR adaptive subspace detector in a scaleinvariant GLRT", IEEE Trans . Signal Processing, accepted for publication .

[12]F.C . Robey, D .R . Fuhrmann, E . J . Kelly and R .A. Nitzberg, "CFAR adaptive matched filter detector", IEEE Transactions on Aerospaces and Electronic Systems, vol . 28, no .l, pp . 208-216, 1992 .

[13] Wai-Sheou Chen and Irving S . Reed, "A new CFAR detection test for radar" , Digital Signal Processing, vol . 1, no. 4, pp . 198-214, 1991 .

[14] E.J. Kelly, "An adaptive detection algorithm", IEEE Transactions on Aerospace and Electronic Systems, vol. 22, no . 1, pp . 115-127, 1986.

[15] L .L Scharf and L . Tood Mc Whorter, " Adaptive matched subspace detectors and adaptive coherence", in Proceedings of the 30th Asilomar Conference on Signals, Systems, and Computers, Pacific Grove, CA, November 1996 .