Innovative Signature Exploitation for Long Range Object Discrimination

Technology Service Corporation proposes an approach to placing object discrimination on firm foundations in a systems theory and mathematical framework. Using System Identification concepts, we establish the relationships among threat object characteristics and behaviors; rotational motion features, and sensor observations. The systems framework defines metrics and disciplined trade space in which the chief measures of effectiveness are bias, variance, complexity and uncertainty. Complexity has a direct influence on performance. The mathematical formalism also provides for a rigorous accounting of assumptions and usage of a priori information. In Phase I, the theory will be further developed from the preliminary foundations described in this proposal and signature-derived body rotational motion features will be demonstrated using single sensor data. In a proposed Phase I Option, the theory will be applied to investigating body rotation features that exploit multiple sensors. The product of the Phase I research and development will be a detailed report on the system theory framework design, and the evaluation of discrimination performance characteristics. The method benefits C2BMC by characterizing a disciplined trade space, providing metrics that are linked to a theoretical foundation, and offering a path to unify discrimination approaches and assumptions across diverse sensors and other elements.