Sensor Fusion Software Tool for CBRN Global Threat Prediction and Assessment

Physical Sciences Inc. (PSI) proposes to develop a software tool that is capable of fusing disparate data sources including hyperspectral standoff sensor data products, weather information, and point sensor data generated from a deployed network of sensors. The proposed software suite will be based on an algorithm that will compute the threat type, spatial and temporal concentration profiles as well as the statistical confidence of the results. This information will be overlaid on a geo-referenced map in order to provide the end-user with an integrated picture of the battle space. The software will be based on the latest developments in statistical learning theory and information fusion. These algorithms will rely on individual sensor performance data (P-d and P-fa) that are generated based on known sensitivity, selectivity, vulnerability, and propensities for false alarm for each individual sensor. As a result of this capability, the software tool will provide the end-user with a network performance prediction of the cloud track and concentration distribution as well as a confidence assessment associated with reported detections across the grid.