Knowledge aided (KA) grouped generalized likelihood ratio detection method

Abstract

The invention provides a KA grouped generalized likelihood ratio detection method. According to the method, units to be detected are grouped; a mathematical model of a G-GLRT detector is constructed;a KA model is used to calculate an estimated value M<^> of a speckle covariance matrix; the speckle covariance matrix M in a mathematical expression of a G-GLRT detector is replaced with M<^>, and a mathematical expression of a KA-G-GLRT detector is obtained; a detection statistic quantity S is calculated on the basis of the KA-G-GLRT detector, S is compared with a threshold, and whether there isa target is thus determined. The grouped algorithm is used, uniform samples in the space part are uniformized in the groups, the texture component of sea clutters in each group satisfies inverse gammadistribution, and scale and shape parameters of the texture of each group are different. The KA-G-GLRT detector has a constant false alarm characteristic for the estimated value M<^> of the speckle covariance matrix. Compared with other detectors, the KA-G-GLRT detector can obtain a better detection performance in sea clutter data actually measuring experiments.

Description

technical field [0001] The invention relates to the field of radar target detection, in particular to a knowledge-aided grouping generalized likelihood ratio detection method. Background technique [0002] In sea surface target detection, the adaptive target detection algorithm matching the statistical characteristics of sea clutter is a commonly used technical means. When using a coherent detector to detect useful signals in sea clutter, it is necessary to estimate the covariance matrix of the sea clutter. The method is to use the clutter data near the unit to be detected as a reference unit, and use these limited reference unit samples to estimate the covariance matrix of the unit to be detected. The sea clutter covariance matrix for the cell. For low-resolution radars with large elevation angles, it is usually assumed that the clutter of the unit to be detected has the same statistical properties as the reference unit clutter, and the clutters of adjacent range units hav...

AI Technical Summary

Problems solved by technology

After analysis, the shortcoming of this patent is: in the case of small SCR (signal to clutter ratio, signal to clutter ratio), since the smoothing factor is smooth around 1, the performance advantage of the S-GLRT detector cannot be reflected

After analysis, the disadvantage of this patent is: when the radar resolution is high, the clutter background will deviate from the Gaussian model, and this detection method will no longer be applicable

Images