Search

A novel dual-polarization and dual-scan (DPDS) classification algorithm is developed for clutter detection in weather radar observations. Two consecutive scans of dual-polarization radar echoes are jointly processed to estimate auto- and cross-correlation functions. Discriminants are then defined and estimated in order to separate clutter from weather based on their physical and statistical properties. An optimal Bayesian classifier is used to make a decision on clutter presence from the estimated discriminant functions. The DPDS algorithm is applied to the data collected with the KOUN polarimetric radar and compared with the existing detection methods. It is shown that the DPDS algorithm yields a higher probability of detection and lower false alarm rate in clutter detection. [ABSTRACT FROM AUTHOR]

A novel dual-polarization and dual-scan (DPDS) classification algorithm is developed for clutter detection in weather radar observations. Two consecutive scans of dual-polarization radar echoes are jointly processed to estimate auto- and cross-correlation functions. Discriminants are then defined and estimated in order to separate clutter from weather based on their physical and statistical properties. An optimal Bayesian classifier is used to make a decision on clutter presence from the estimated discriminant functions. The DPDS algorithm is applied to the data collected with the KOUN polarimetric radar and compared with the existing detection methods. It is shown that the DPDS algorithm yields a higher probability of detection and lower false alarm rate in clutter detection.