LL Logo



Strategies for Minimal
Sample Support STAP



S. Unnikrishna Pillai, Y.L. Kim, and Joseph R. Guerci
Polytechnic University
Five Metrotech Center
Brooklyn, New York 11201
tel: (718) 260-3732
email: pillai@arma.poly.edu

Abstract This paper provides a comprehensive strategy for maintaining effective STAP performance while utilizing a parsimonious sample support for adaptive weight training. The need for such minimal sample support techniques is particularly acute in overland operations, where the usual wide sense stationarity assumption with down range is rarely satisfied, often leading to the so-called undernulled clutter problem [1]. The mitigation approach described herein is two pronged: The first element consists of a new closed form expression for the optimal loading factor for generally rank deficient sample covariance matrices. The expression also yields as special cases both the traditional SMI covariance estimate when the sample support is very large relative to the number of adaptive degrees-of-freedom (DOFs), and the Hung-Turner projection when the sample support is small [2]­[4]. A new computationally efficient inversion procedure is established for the small sample support case, which can yield up to an order of magnitude reduction in complexity as compared to a direct matrix inversion (DMI) of a diagonally loaded sample covariance matrix. This approach is then combined with new forward-backward smoothing techniques resulting in STAP interference cancellation performance equivalent to SMI with a sample support many times greater than the DOFs employed. All results are based on both synthetic and real data sets, the latter originating from the DARPA Mountaintop radar.

[1] J.R. Guerci, S.U.Pillai and Y.L. Kim, "Empirical Findings of Post-STAP Clutter Cancellation Residual Statistics: A Case for Adaptive Sub-CPI Processing," Post-STAP Detection Technical Interchange Meeting, MIT Lincoln Laboratory, November 8, 1995.

[2] S.U. Pillai, Y.L. Kim and J.R. Guerci, "An Efficient Implicit Interference Removal Technique for STAP," Proceedings of the Adaptive Sensor Array Processing (ASAP) Workshop, 12­14 March, MIT Lincoln Laboratory, Lexington, MA 1997.

[3] M. Zatman, "Properties of Hung-Turner Projections and their Relationship to the Eigencanceler," Proceedings of the Thirtieth Annual Asilomar Conference on Signals, Systems and Computers, Monterey, CA, October 29­November 1, 1996.

[4] C.H. Gierull, "Performance Analysis of Fast Projections of the Hung-Turner Type for Adaptive Beamforming," Signal Processing, vol. 50, 1996.




Direct comments and questions to: webmaster@ll.mit.edu

MIT Lincoln Laboratory. All rights reserved.