Analytical analysis of STAP algorithms for cases with mismatched steering and clutter statistics

Abstract

In the majority of adaptive radar detection algorithms, the covariance matrix for the clutter-plus-noise is estimated using samples taken from range cells surrounding the cell under test. In a nonhomogeneous environment, this can lead to a mismatch between the mean of the estimated covariance matrix and the true covariance matrix for the range cell under test. Further, an inaccurate target steering vector may also be employed. Closed form expressions are provided, which give the performance for such cases when any of a set of popular space-time adaptive processing (STAP) algorithms are used. The expressions are exact for some interesting cases. For some other cases, it is demonstrated that the expressions provide good approximations to the exact performance. To simplify the analysis, the samples from the surrounding range cells are assumed to be independent and identically distributed and these samples are assumed to be independent from the sample taken from the cell under test. A small number of important parameters describe which types of mismatches are important and which are not. Monte Carlo simulations are included which closely match the predictions of our equations. Numerical results demonstrate that steering vector mismatch can offset covariance matrix mismatch in some cases.