Rapidly Adaptive Intelligent Radar (RAIR)

The sample support problem in space-time adaptive processing (STAP) applicationsarises from the requirement to adapt many spatial and temporal degrees-of-freedom (DOF) to a changing interference environment that includes clutter and jammers. Often, in heterogeneous overland strong clutter environments, the available wide sensestationary sample support is severely limited to preclude the direct implementation of the sample matrix inverse (SMI) approach.In this proposal we outline several approaches to address the sample support problem: (i) Generalized forward/backward sub-aperture-subspace smoothingtechniques to reduce the number of data samples in estimating the samplecovariance matrices (ii) Projection methods using alternating projections orrelaxed projection operators onto desired convex sets to retain the a-prioriknown structure of the covariance matrix.Our initial analysis shows that by combining these approaches with eigenbased techniques, it is possible to reduce significantly the data samples requiredin non-stationary environment and consequently achieve superior target detection.In fact, multiplicative improvement in data reduction compared to directeigen-based methods can be obtained at the expense of negligible loss inspace-time aperture. Phase I efforts will concentrate on obtaining theimprovement in performance by combining these methods and will be supportedby analytical study as well as simulation results.The improved adaptive transmit signal design can be critically important fordetection of extended targets including non-military applications such asmonitoring drug trafficking activities and location identification of cellular systems using limited number of data samples, as well as high resolution SAR technology.