Target tracking in over the horizon radar

This paper describes the application of the Deep Target Extractor (DTE), developed from the Maximum Likelihood Probabilistic Multiple Hypothesis Tracker (ML-PMHT), to measurements from an over-the-horizon radar (OTHR) observing a highly maneuvering target (HMT). We describe the motion model for HMTs that start at very large speeds and makes extremely sharp turns. We then present a measurement model for OTHR that uses ray-tracing software to produce detections across multiple signal paths resulting from ionospheric refractions. Next, we describe the DTE operation in the multiple signal path framework of the OTHR. Finally, we present a test scenario where the DTE performance shows low root-mean-square error (RMSE) for HMT motion parameter estimation.