A Method for Threshold Setting and False Alarm Probability Evaluation for Radar Detectors

A threshold setting and false alarm probability evaluation method is proposed for radar detectors for which an analytical relationship between thresholds and false alarm probabilities is impossible or difficult to obtain due to complicated detection schemes or non-Gaussian noise and clutter. Importance sampling is performed to increase the sampling efficiency of low-probability samples. Then, the weighted piecewise parabolic (P²) interpolation method is used to rapidly plot the relationship curve between the threshold and the false alarm of the detector, with a low storage requirement and high accuracy. The proposed method effectively enhances the detector threshold computing scale and addresses the statistical verification of the false alarm probability for radar detectors with a low false alarm rate.