Radar signal adaptive detection method based on autoregressive model

Abstract

The invention discloses a radar signal adaptive detection method based on an autoregressive model. According to the concept, a radar receives coherent pulse sequences of N pulses, the coherent pulse sequences of the N pulses serve as to-be-detected unit echoes z0 of a target, and then a target detection problem of the radar is represented by a binary hypothesis test, wherein H0 indicates that z0 only has an interference hypothesis, H1 indicates that z0 has target and interference hypotheses, and first-order partial derivatives of a joint probability density function f(z0, ZK / theta) of z0 and ZK on two-dimensional column vectors theta r of a target amplitude, an upper left block matrix of a Fisher information matrix J(theta) inverse of a to-be-estimated parameter theta, the maximum likelihood estimator of variance sigma<2> of white complex Gaussian noise and the maximum likelihood estimator of an autoregressive parameter vector a of the M-order autoregressive model are calculated respectively; the detection threshold eta AR-Rao of the autoregressive model based on the Rao detection method is set, and a target detection expression TR, based on the autoregressive model, in z0 is calculated; if the TR is larger than eta AR-Rao, a target exists in z0; otherwise, the target does not exist in z0.

Description

technical field [0001] The invention belongs to the technical field of radar signal processing, in particular to a radar signal adaptive detection method based on an autoregressive model, which is suitable for the adaptive detection of radar signals. Background technique [0002] Today's radar is facing various challenges: active jamming and passive jamming, anti-radiation radar, the rapid development of stealth technology, etc. Adaptive technology is one of the effective ways to deal with these challenges; Adaptive detection techniques have been extensively studied. In the environment of uniform clutter interference, some scholars proposed a generalized likelihood ratio detection method. This method needs to obtain the maximum likelihood estimation of unknown parameters under different assumptions, which has a large amount of calculation. The adaptive matched filter is also known as the two-step generalized likelihood ratio test. First, it is assumed that the covariance ma...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the shortcoming of the traditional adaptive detection method that the detection performance is degraded in the absence of training data in the actual radar working environment, the purpose of the present invention is to propose a radar signal adaptive detection method based on the autoregressive model. Regression model radar signal adaptive detection method to reduce the detection method's requirements for training data, thereby improving the adaptive detection performance of radar signals in the absence of training data

Images