| In this context, a new technique called block augmented matrix (BAM) inversion method is introduced that allows for updating as each pulse is received-in contradistinction to conventional STAP methods that process the pulse returns contained in a coherent processing interval (CPI) in a batch, thereby achieving significant computational throughput gains. Additionally, auxiliary innovation covariance information is formed which can be used as an on-line adaptive stopping criterion, thereby forming the basis for an adaptive reduced-rank STAP processor.; For the sample support problem, the intrinsic interference subspace removal (ISR) algorithm based on the diagonally loaded sample matrix inversion (LSMI) is developed and its recursion algorithms are derived. Also, the optimal loading factor is derived in terms of the singular values of sample covariance matrix. The complexity of ISR is the same as that of the Hung-Turner projection (HTP) method, which is the simplest and fastest algorithm, and the performance is very close to the eigen-based subspace methods, the optimal solution.; Finally, subarray-subpulse schemes using forward and backward data vectors are introduced to overcome the data deficiency problem. It is shown that multiplicative improvement in data samples can be obtained at the expense of negligible loss in space-time aperture of the steering vector. The utility of these new approaches is established using both simulated and actual adaptive array radar data obtained from the ARPA Mountain Top Dataset. |
