Radar Centroid Processing Algorithm with Tracker Feedback Navy SBIR FY2005.2 Sol No.: Navy SBIR FY2005.2 Topic No.: N05-118 Topic Title: Radar Centroid Processing Algorithm with Tracker Feedback Proposal No.: N052-118-0325 Firm: Numerica Corporation PO Box 271246 Ft. Collins, Colorado 80527-1246 Contact: Benjamin Slocumb Phone: (970) 419-8343 Web Site: www.numerica.us Abstract: In surveillance platforms such as the E-2C and upcoming E-2D, it is essential that the Airborne Early Warning (AEW) radar provide a complete track picture of the surveillance region. To maintain continuous tracks throughout the surveillance region, the radar sensor must be able to produce measurements on all targets. Problems occur when there are closely spaced objects (CSOs). Here, the radar signal processor can combine the returns from the targets to produce a single merged measurement. The objective in the proposed research is to develop a new surveillance radar centroid processing algorithm that has an extended merged measurement capability. The algorithm will "parse" the raw range-bearing cell measurements such that object measurements for each target can be generated. The use of tracker feedback enables the processing algorithm to specify the expected number of objects within the combined cluster of measurement primitives. The work will leverage Numerica's prior experience in radar centroid processing and pixel-cluster decomposition research. Benefits: For airborne early warning surveillance radar systems, a primary requirement is that a complete track picture be achieved so that operators and commanders have adequate information to make tactical decisions. Centroid processing is a signal processing function performed within the radar. It converts primitive data (raw measurements) into object reports that feed the tracking system. In cases where targets are closely spaced, a classical problem is the one where the centroid processor merges the primitive data to form only one object report. This can cause a serious problem because some tracks will "starve" of data and eventually be dropped. Thus, the critical need is for an advanced centroid processing algorithm that can "break up" merged measurements and form the right number of reports to support the existing tracks. The thrust of the proposed program is to develop a new centroid processing algorithm that uses tracker feedback to break-up merged measurement reports to provide the right number of object measurements. The availability of this capability will provide a significant benefit to the surveillance radar operators and commanders that use the radar data. With this algorithm, the radar will be able to maintain tracks on closely spaced targets, thereby keeping a complete track picture. The maintenance of the complete track picture is a critical requirement for proper command and control actions. Return