Entity

Time filter

Source Type


Gu X.-F.,Yantai Naval Aeronautical and Astronautical University | Jian T.,Yantai Naval Aeronautical and Astronautical University | He Y.,Yantai Naval Aeronautical and Astronautical University | Hao X.-L.,Yantai Electricity and economics Technical Institute
Yingyong Kexue Xuebao/Journal of Applied Sciences | Year: 2013

By generalizing the clutter-clustered estimation method and considering the normalized sample covariance matrix (NSCM), a generalized NSCM(GNSCM) is proposed for covariance matrix structure estimation in correlated compound-Gaussian clutter. A maximum likelihood recursive estimation process of covariance matrix structure is derived in generalized clutter-clustered background. A generalized approximate maximum likelihood (GAML) estimator is then obtained by using GNSCM as the initialized estimation estimated matrix to recursive. GAML is an extension of the existing methods the approximate maximum likelihood (AML) and the constrained recursive clutter-clustered estimator (CRCCE). Simulation results show that, compared with the two previous methods, GAML has higher estimation accuracy, and the corresponding adaptive detector has better constant false alarm ratio (CFAR) property and detection performance. Source


Gu X.,Yantai Naval Aeronautical and Astronautical University | Jian T.,Yantai Naval Aeronautical and Astronautical University | He Y.,Yantai Naval Aeronautical and Astronautical University | Hao X.,Yantai Electricity and economics Technical Institute
Hangkong Xuebao/Acta Aeronautica et Astronautica Sinica | Year: 2012

The problem of detecting a range-spread target in a non-Gaussian clutter modeled as a spherically invariant random vector (SIRV) is investigated when the steer vector is mismatched or unknown. First, the steer vector is assumed known, and the statistics of the normalized matched filter (NMF) of each range cell are obtained by utilizing the generalized likelihood ratio test (GLRT). The NMF integrator (NMFI) is obtained by integrating incoherently the NMF statistics of the range cells which the range-spread target occupies. Then, the steer vector is estimated using the eigenvalue decomposition and the average phase-angle difference by maximizing the detection statistics of the NMFI. Finally, a blind-NMFI (B-NMFI) is obtained using the estimated steer vector. Simulation results show that the detection performance of the NMFI is better than that of GLRT when the steer vector is mismatched, and the B-NMFI can detect a range-spread target effectively and is robust to detect the targets in diverse orientations when the steer vector is unknown. Source


Gu X.,Yantai Naval Aeronautical and Astronautical University | Jian T.,Yantai Naval Aeronautical and Astronautical University | He Y.,Yantai Naval Aeronautical and Astronautical University | Hao X.,Yantai Electricity and economics Technical Institute
Hangkong Xuebao/Acta Aeronautica et Astronautica Sinica | Year: 2013

This paper addresses the adaptive detection of range-spread targets in a structured compound-Gaussian clutter (CGC). In view of the fact that the asymptotic generalized likelihood ratio test in a heterogeneous environment (AGLRT-HTG) suffers a signal to clutter ratio loss in a CGC environment, the structured CGC is modeled as an autoregressive process and a recursive AGLRT in the compound-Gaussian clutter (RAGLRT-CGC) environment is proposed by using the method of asymptotic generalized likelihood ratio test (AGLRT) and the idea of recursive estimation. The analytical formula relating false alarm probability to detection threshold for limit cases is deduced. The simulation results show that the RAGLRT-CGC is robust to different multiple dominant scattered targets and the detection performance of RAGLRT-CGC is obviously better than the AGLRT-HTG. Source


Gu X.-F.,Yantai Naval Aeronautical and Astronautical University | Jian T.,Yantai Naval Aeronautical and Astronautical University | He Y.,Yantai Naval Aeronautical and Astronautical University | Hao X.-L.,Yantai Electricity and economics Technical Institute
Yuhang Xuebao/Journal of Astronautics | Year: 2012

The problem of covariance matrix structure estimation is addressed for radar target adaptive detection in correlated compound-Gaussian clutter environment modeled as spherically invariant random vector. The method of clutter-clustered estimation is generalized and modified, and a generalized recursive clutter-clustered estimation (GRCCE) is proposed. The recursive process of covariance matrix structure estimation is derived in generalized clutter-clustered environment based on the method of maximum likelihood estimation. A generalized clutter-clustered estimation (GCCE) is proposed based on the idea of clutter-clustered estimation. And then GRCCE is obtained by recursion with GCCE as the initialized estimation matrix. The simulation results show that one recursion is sufficient to obtain the desired goals. The estimation accuracy of the GRCCE is improved as first order correlation coefficient becomes bigger, and it is independent of the clutter texture components. Compared with the existing methods, the GRCCE is shown to be better adaptability, higher estimation accuracy and lower computational burden. Source

Discover hidden collaborations