LL Logo



The CREST Challenge ­ Overview and Analysis

Peter A. Zulch and Joseph R. Guerci 1
USAF Rome Laboratory
Advanced Technology Group
Rome, NY 13441-4514
email: zulchp@rl.af.mil

1 Science Applications International Corporation
Air Defense Technology Group
4001 No. Fairfax Drive, Suite 400
Arlington, VA 22203

Abstract In support of the DARPA Advanced Signal Processing Program a challenge had been issued to the radar and signal processing community. Several simulated data sets that emulate a flying phased array surveillance radar were created as a challenge to the STAP community to resolve all target information. There were four data sets created for the challenge using the Rome Laboratory Space Time Adaptive Processing Tool (RLSTAP). The first three data sets contained returns from a single coherent processing interval (cpi) over different clutter types and target/jammer scenarios. The fourth data set contained ten cpi's worth of data and simulated a typical airborne radar scenario where the radar would revisit a point on the ground ten different times with a 30 sec revisit time. The first portion of this presentation will describe the CREST challenge data sets and how to access them via the World Wide Web for future STAP investigations. The second part of the talk will be a discussion of the approach employed by the winning strategy. Issues highlighted will include robust and computationally efficient strategies for STAP rank selection, weight calculation, as well as post-STAP and CFAR processing.



LL Logo Disclaimer

Direct comments and questions to: webmaster@ll.mit.edu

MIT Lincoln Laboratory. All rights reserved.