Description:
OBJECTIVE: Identify & evaluate data/signal processing techniques and algorithms that will minimize or overcome the system degradation effects caused by dense threat complexes, consisting of large numbers of closely-spaced uninteresting ballistic objects. The intent of this Topic is to develop a Bulk Filtering method where the radar return data for non-threatening objects are de-emphasized, suppressed, or rejected before they are presented to the signal/data processors and tracking software for further acquisition, tracking, or discrimination processing. This proposed Topic is a paradigm shift in Bulk Filtering Debris by rejecting the radar return data at the Detection Level. The expectation is that any final product from this effort will yield improvements in the efficient use of sensor resources and the accuracy of sensor data products. Efficiency is gained by eliminating the resource overhead currently used to process non-threatening objects. Accuracy of the reported data improves by removing the large quantities of non-threatening objects from the RF scene shared with the threatening object. By Bulk Filtering objects at the detection level, the risk that a new detection will be mistakenly correlated & integrated with the track of a threatening object is eliminated, and the degradation in track accuracy due to Debris is Mitigated. DESCRIPTION: Ballistic Missile Defense System (BMDS) performance is dependent on the efficient acquisition, tracking and discrimination of threatening objects. Reducing the resource overhead necessary to process non threatening objects ultimately improves a sensor"s effectiveness and enhances the system probability for a successful intercept. As threat complex numbers, densities and countermeasures increase, it becomes even more important to manage radar resources, and minimize extraneous data. This effort is intended foster improvements in RF discrimination, debris mitigation and track management capabilities for TPY-2 (Forward Based Mode & Terminal Mode) and SBX. Technical areas of interest include, but are not limited to: Bulk filtering techniques and limitations using current detection algorithms Innovative detection algorithms that identify debris and non-threatening objects that will be excluded from further processing Phase I: Develop and conduct proof-of-principle studies and/or demonstrations of discrimination concepts/algorithms that are easily adaptable to a wide range of sensors using simulated sensor data. Phase II: Update/develop algorithm(s) based on Phase I results and demonstrate technology in a realistic environment using data from multiple sensor (as applicable) sources. Demonstrate ability of the algorithm(s) to work in real-time in a high clutter and/or countermeasure environment. Phase II demonstration work will be classified. Phase III: Integrate algorithms into the BMDS and demonstrate the improved total capability of the updated system. Partnership with traditional DOD prime-contractors will be pursued as government applications of this technology will produce near term benefits from a successful program. DUAL USE/COMMERCIALIZATION POTENTIAL: Weather Radar (ability to penetrate debris and look into a storm), Air Traffic Control (ability to reject debris and environmental clutter) REFERENCES: 1. R. Duda, P. Hart, and D. Stork,"Pattern Classification", 2nd Ed., Wiley Interscience, November, 2000. 2. Jenson, Finn V. Bayesian Networks and Decision Graphs. New York: Springer, 2001. 3. Gilks, W.R., Richardson, S. and Speigelhalter, D.J. Markov Chain Monte Carlo In Practive. Boca Raton: Chapman & Hall, 1996. 4. Neapolitan, Richard E. Learning Bayesian Networks. Upper Saddle River: Prentice Hall, 2004. 5. Martinez, David, et.al.,"Wideband Networked Sensors", MIT Lincoln Labs, http://www.fas.org/spp/military/program/track/martinez.pdf, October 2000. 6. D. Hall and James Llinas,"An Introduction to Multisensor Data Fusion,"Proceedings of the IEEE, 85 (No. 1) 1997. 7. D.C. Cowley and B. Shafai,"Registration in Multi-Sensor Data Fusion and Tracking,"Proceedings of the American Control Conference, June 1993. 8. Y. Bar-Shalom and W.D. Blair, Editors, Multi-Target/Multi-Sensor Tracking: Applications and Advances, Vol. III, Artech House, Norwood, MA, 2000. 9. T. Sakamoto and T. Sato,"A fast Algorithm of 3-dimensional Imaging for Pulsed Radar Systems,"Proceedings IEEE 2004 Antennas and Propagation Society Symposium, Vol. 2, 20-25 June 2004. 10. W. Streilein, et al."Fused Multi-Sensor Mining for Feature Foundation Data,"Proceeding of Third International Conference of Information Fusion, Vol. 1, 10-13, July 2000. 11. Mike Botts [ed.], OpenGIS Sensor Model Language (SensorML), OGC 05-086r2. http://www.opengeospatial.org/standards/requests/31. 12. M. Ceruti,"Ontology for Level-One Sensor Fusion and Knowledge Discovery,"8th European Conference on Principles and Practice of Knowledge Discovery in Databases, Pisa, Italy, 2004. 13. Steve Havens [ed.], OpenGIS Transducer Markup Language TransducerML), OGC 06-010r2. http://www.opengeospatial.org/standards/requests/33. 14. Russomanno, D.J.; Kothari, C.; Thomas, O."Sensor ontologies: from shallow to deep models."System Theory, 2005. SSST'05. Proceedings of the Thirty-Seventh Southeastern Symposium on, Vol., Iss., 20-22 March 2005. Pages: 107- 112.
