Tutorial 3.6
Thursday PM, 27 April 2006

Advanced STAP Concepts: Thinking Outside the Sample Covariance
Instructor: Joseph Guerci, DARPA SPO, USA

Synopsis: It has been well established that space-time adaptive processing (STAP) for radar performance can be significantly degraded when confronted with challenging nonstationary terrain clutter and dense target backgrounds. In this tutorial, an in-depth overview of the newly emerging advanced techniques such as knowledge-aided STAP will be explored. Through the exploitation of a new real-time embedded computing architecture incorporating a high-fidelity physical database, many adaptation problems due to nonstationary clutter environments can be solved in an entirely new manner. Details of the “look-ahead” scheduling for real-time database management developed under the DARPA KASSPER program will be provided along with results for both simulated and measured data. In particular, methods for augmenting the sample data matrix in a manner that still allows for conventional QR factorization will be highlighted.

Dr. Joseph Guerci is the Director of the Special Projects Office (SPO) at the Defense Advanced Research Projects Agency (DARPA), where is engaged in the development of next generation adaptive sensors and systems for both terrestrial and space based applications. He has eight US patents, numerous peer reviewed technical publications, is a Member of the IEEE Radar Systems Panel, a Fellow of the IEEE (for “Contributions to advanced radar theory and its embodiment in real-world systems”), and is author of Space-Time Adaptive Processing for Radar (Artech House, 2003) that is based on a popular course he has taught in academia, government and industry.