A k-nearest neighbors (KNN) approach to the design of radar detectors is investigated. The idea is to start with either raw data or well-known radar statistics as feature vector to be fed to the KNN decision rule. Formulas for the probability of false alarm and probability of detection are derived and used to show that the former detector possesses the constant false alarm rate (CFAR) property with respect to the power of the disturbance in the cell under test while the latter is CFAR with respect to the overall matrix of the disturbance. Simulation examples together with results obtained using real clutter recordings are provided to illustrate the behavior of detectors derived using the proposed approach.
A k-nearest neighbors approach to the design of radar detectors
Coluccia A.;Fascista A.;Ricci G.
2020-01-01
Abstract
A k-nearest neighbors (KNN) approach to the design of radar detectors is investigated. The idea is to start with either raw data or well-known radar statistics as feature vector to be fed to the KNN decision rule. Formulas for the probability of false alarm and probability of detection are derived and used to show that the former detector possesses the constant false alarm rate (CFAR) property with respect to the power of the disturbance in the cell under test while the latter is CFAR with respect to the overall matrix of the disturbance. Simulation examples together with results obtained using real clutter recordings are provided to illustrate the behavior of detectors derived using the proposed approach.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
