Analysis of Wind Field Interpolation Methods for a Real-Time CBRN Track Before Detection Algorithm

Real-time chemical, biological, radiological, and nuclear (CBRN) threat detection is a complex problem due to high false alarm rates from individual sensors and varying noise profiles, both of which contribute to performance variability. The current state of operations requires analysts to fuse information coming from multiple sensors and modalities, define its context, and decide on an actionable path forward. Multi-Node Analytics (MNA) is an emerging tool that leverages advents in Internet of Things (IoT), low Size, Weight and Power (SWaP) computing, and communications to reduce the minimum detectable signal, lower false alarm rate, and decreases time to detect (compared to traditional techniques). Accurate micro-scale wind fields are crucial for correlating CBRN detections across sensors within MNA plume detection algorithms that rely on Gaussian puff propagation. In-situ wind measurements provide a robust foundation for generating these wind fields in real-time but require accurate interpolation between sensors. For the MNA use case, these interpolation schemes must be accurate enough for puff advection and dispersion, yet fast enough to operate in real-time.

Recent advancements in diagnostic wind models (DWMs) make them a viable candidate for real-time wind interpolation for MNA. Areté compared results of various approaches to interpolate microscale wind fields for propagating Gaussian puffs, such as bilinear interpolation and the WindNinja DWM at a variety of locations. Real micrometeorological network wind data from relevant environments were used for comparing wind interpolation methods using a holdout validation approach to determine model performance. Areté will present the results from this study highlighting the importance of fast and accurate wind interpolation methods in the performance of MNA algorithms.