Adaptive Thresholding of the GIP Statistic to Remove Ground Target Returns from the Training Data for STAP Applications

摘要

This paper presents an adaptive thresholding algorithm that can be used in conjunction with the multi-pass GIP-based editing method to adaptively eliminate ground moving target returns ('non-homogeneities') from the training data used for space-time adaptive processing (STAP) applications such as adaptive radars. The algorithm exploits a property of the generic structure of the ordered GIP statistic the origins of which have been theoretically developed by the authors and a single adjustable parameter related to the Type I error of incorrectly excising target-free training data to adaptively determine the thresholds for excising target-contaminated training data. An iterative application of the method improves the distinction between non- homogeneities and background clutter which increases the number of target- contaminated training bins excised. Moreover, the editing method has been extended to reduced degree-of-freedom (DoF) STAP implementations such as multi- bin post-Doppler STAP. As a consequence the method is practically implementable for realistic scenarios with very limited a priori knowledge (i.e. only the number of DoFs is required).