Waveform Design for Kalman Filter-Based Target Scattering Coefficient Estimation in Adaptive Radar System

作     者：Chen, Peng Qi, Chenhao Wu, Lenan Wang, Xianbin 

作者机构：Southeast Univ State Key Lab Millimeter Waves Nanjing 210096 Jiangsu Peoples R China Columbia Univ Dept Elect Engn New York NY 10027 USA Southeast Univ Sch Informat Sci & Engn Nanjing 210096 Jiangsu Peoples R China Western Univ Dept Elect & Comp Engn London ON N6A 3K7 Canada 

出 版 物：《IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY》 (IEEE运载工具技术汇刊)

年 卷 期：2018年第67卷第12期

页      面：11805-11817页

核心收录：

学科分类：0810[工学-信息与通信工程] 0808[工学-电气工程] 08[工学] 0823[工学-交通运输工程] 

基　　金：National Natural Science Foundation of China [61801112, 61871119, 61471117, 61601281] Natural Science Foundation of Jiangsu Province [BK20180357, BK20161428] Open Program of State Key Laboratory of Millimeter Waves at Southeast University [Z201804] Fundamental Research Funds for the Central Universities 

主　　题：Adaptive radar system Kalman filter non convex optimization waveform optimization 

摘      要：The temporal correlation of target can be exploited to improve the radar estimation performance. This paper studies the estimation of target scattering coefficients in an adaptive radar system, and a novel estimation method based on Kalman filter (KF) with wavetbrm optimization is proposed for the temporally correlated target in the scenario with both noise and clutter. Different from the existing indirect methods, a direct optimization method is proposed to design the transmitted waveform and minimize the mean square error of the KF estimation. Additionally, the waveform is optimized subject to the practical constraints including the transmitted energy, the peak-to-average power ratio, and the target detection performance. With clutter and noise, the waveform optimization problem is non-convex. Therefore, a novel two-step method is proposed and converts the original non-convex problem into several semidefinite programming problems, which are convex and can solve efficiently. Simulation results demonstrate that the proposed KF-based method with waveform optimization can outperform state-of-art methods and significantly improve the estimation performance.