Robust multi-look HRR ATR investigation through decision-level fusion evaluation

Simultaneous tracking and identification (STID) is impacted by sensor and target dynamics especially in move-stop-move type scenarios. For most scenarios, both moving and stationary targets can be processed into 1D High-Range Resolution (HRR) radar profiles which contain enough feature information to discern one target from another to help maintain track or to identify the vehicle. To meet mission objectives, different decision-level and feature-level classifiers can be designed to achieve performance requirements such as the sensitivity of the number of features for a given location accuracy, identification confidence, timeliness (revisit rate and track length), and throughput of the number of targets tracked. For robust STID evaluation, repeatable scenarios, metrics, and data support is recommended for comparisons. This paper compares the ATR performance of a baseline single-look algorithm to the performance of decision level and feature level fusion ATR algorithms through multilook assessments to assess relative fusion performance gains.