Adaptive Data Fusion for Real-time Threat Assessment Navy SBIR FY2010.3 Sol No.: Navy SBIR FY2010.3 Topic No.: N103-224 Topic Title: Adaptive Data Fusion for Real-time Threat Assessment Proposal No.: N103-224-0063 Firm: SEA CORP 62 Johnny Cake Hill Aquidneck Corporate Park Middletown, Rhode Island 02842 Contact: John Murphy Phone: (401) 847-2260 Web Site: www.seacorp.com Abstract: Electronic Support (ES) systems are among the fundamental instruments used for threat detection onboard Navy surface, sub-surface, and air platforms. They are tasked to sense the RF environment, to sort out all emitters, measure key associated parameters, and contribute to a comprehensive situational awareness with respect to all activity occurring in the RF spectrum. As threat systems evolve, numerous ES systems are developed or refined to contend with the ever varying RF landscape. To keep pace with these changes, ES systems need to evolve to deal with increasing sophistication of emitters, creating an inherent gap between the ways emitters are represented in different ES systems. Because of this ongoing engineering evolution, it is not uncommon to have two or more heterogeneous ES systems aboard a war fighting vessel causing the same emitter sensed from one ES system to have a representation in a second ES system that is different and not directly comparable. A system designed to perform adaptive data fusion is well-suited to addressing this problem because it will significantly improve the operator's ability to detect and analyze valid threat signals in and electronic environment that is growing at a rapidly increasing rate. Benefits: Adaptive data fusion algorithms and their ability to aid in decision making using multiple independent data sources are extremely useful in complex environments. These systems, using adaptive algorithms, reduce processing power and operator workload while providing improved results over human in the loop systems. Integrating different classification schemes (Linear, Non-linear, Kernel Equivalence) and the ability of the algorithm to adapt to changes in the data sources significantly improves the effectiveness of the system. Due to the generic nature of these data fusion algorithms, and their ability to adapt to different environments, this system is well suited for almost any environment that requires a data fusion and decision making system. Return