Cutset Sampling Topologies for Intelligence, Surveillance, and Reconnaissance

Distributed data gathering and information fusion are at the heart of localization and efficient monitoring of spatiotemporal phenomena of interest in wide range of applications. They are thus enabling tools for detection, containment and remediation for various threats, ranging from biological/chemical and radiological weapons, through other mobile entities such as troops and (unmanned) reconnaissance and tactical objects. When missions involve reconnaissance/surveillance in irregular warfare, in addition to the resource (e.g., energy) constraints, data gathering may need to be carried out under secrecy, with limited access time as well as other adverse conditions. The goal of the proposed research is to develop novel constraint-aware approaches for designing dynamic structures and distributed algorithms to be used in missions involving geospatial data gathering tasks. The proposed approaches will seamlessly integrate the management of: (1) spatial sensor deployment and data communication constraints, (2) signal characteristics (e.g., continuity, expected size and shape), and (3) unconventional tradeoffs in terms of balancing information gains with spatiotemporal constraints. We aim at achieving accurate and robust signal reconstruction, interpolation, and analysis for the purpose of information extraction. The proposed technology will significantly enhance naval capabilities for remote detection and assessment of various security threats, spanning from motion of actual physical entities through weapons of mass destruction.