Space-time adaptive processing (STAP) is a well-suited technique to detect
slow-moving targets in the presence of a clutter-spreading environment. When
considering the STAP system deployed with conformal radar array (CFA), the
training data is range-dependent, which results in poor detection performance
of statistical-based algorithms. In this paper, we propose registration-based
compensation using sparse representation (SR-RBC) to generate stationary
training data. This method first converts the estimation of both the unknown
configuration parameters and clutter power distribution into an ill-posed
problem with the constraint of sparsity, and then utilizes the technique of
sparse representation like iterative reweighted least squares (IRLS) to solve
it. Based on this, the transform matrix is designed so that the processed
training data behaves nearly stationary with the test cell. Since the
configuration parameters as well as the clutter spectral response is obtained
with full-snapshot using sparse representation, SR-RBC provides more accurate
clutter spectral estimation and is more robust to the configuration parameters
so that the processed training data is more stationary and achieve better
signal-clutter-ratio (SCR) improvement than that of registration-based
compensation (RBC) method.