Optimal Adaptive Detection Method Using Azimuth Reference Cells Under K-distributed Clutter

Moving target detection in sea clutter especially weak target detection has great significance in military and civil fields and is a hard problem in the field of radar target detection simultaneously. In the study of the detailed research on technology of target detection in sea clutter, an adaptive coherent optimum detection strategy is proposed in this paper, which applies Doppler offset based on scanning returns of airborne coherent radar. Its main advantage lies in the full use of the prior information of the detected contiguous beam position data into the detection for the current resolution range cell after Doppler offset.Firstly, an analysis of physical structure mechanism and statistical characteristics of sea clutter is given. In terms of the former, the back scattering mechanism and three-component model of sea clutter are mainly introduced. While in terms of the latter, the amplitude characteristics and coherent properties of sea clutter are discussed. The research and modeling of sea clutter properties provide the preconditions and theoretical basis for the design of target detection and performance improvement.Secondly, the non-coherent integrated and adaptive detection methods are reviewed in sea clutter and their scopes of application as well as disadvantages and advantages are analyzed. The coherent integrated detection methods mainly include CA-CFAR, OS-CFAR and CCS-CFAR. As for the adaptive methods, this paper covers the detection methods of generalized likelihood ratio, adaptive matched filter(AMF), and adaptive normalized matched filter(ANMF). On these basis, our study focuses on seeking for a more applicative detector.Finally, the K-distributed model of sea clutter and ?-AMF are presented to complete the structure of coherent optimum detection scheme in the whole spatial Doppler offset. It is an asymptotically optimum detection for K distribution to perform in Gaussian and non-Gaussian clutter. Due to the high dependence of adaptive detection on the estimate of clutter covariance matrix, three robust methods for clutter covariance matrix are involved. Then, an adaptive optimum detection scheme is put forward which uses the spatial Doppler offset in the condition of airborne beam scanning. After Doppler offset, this schemeappropriately employs the returns of contiguous beam position reference cells which has been detected and serves it as prior information to perform in target detection at current cell under test. Moreover, an estimate method of airborne Doppler offset at resolution cell is proposed so as to eliminate the Doppler effect induced by airborne movement at different cells of radar returns. The experiments using both simulated data and real sea clutter datasets verify the effectiveness of the proposed scheme and show that it has apparent improvement in performance compared with the traditional detection one in which only reference cells along range are selected.