Research On Target Location Algorithms And Test Method Of Ground Reflection Based On Passive Radar

Multistatic passive radar(MPR) exploits non-cooperative illuminators of opportunity as transmitters to perform target detection and localization. Due to its potential ?four countering’ advantages, the MPR has attracted considerable interests. This work was supported in part by the Basic Research Fund of Beijing Institute of Technology and the National Natural Science Foundation of China. The location algorithm and the problems of measuring reflecting coefficient are studied in this paper which can improve the localization performance in MPR. We introduce two quasi-Newton algorithms, namely Spectral Approach for Nonlinear Equations(SANE) and Nonmonotone Spectral for Nonlinear Equations(NMSGNE), to solve nonlinear equations for target locating in MPR. The effectiveness of the proposed target location algorithms is verified via simulation. A easy method for site measuring is also proposed to measure the reflecting coefficient of the ground surface and modify antenna pitching pattern. Its feasibility is verified via many field experiments. The main research contents are as follows:Firstly, this paper investigates the target location algorithm based on passive radar in the detection mode, and introducing two quasi-Newton algorithms of SANE and NMSGNE to solve the nonlinear equations. It overcomes the shortcomings that Newton iterative algorithm cannot be used if Jacobian matrix of the nonlinear equations is singular. Compare with Newton algorithm via simulation, the introduced two quasi-Newton algorithms have better location accuracy.Secondly, this paper carries out a analysis about antenna beam splitting caused by ground reflection. A method for measuring the ground reflection coefficient based on the amplitude of the upper and lower antenna elements is presented and its accuracy is higher compared with the previous modeling method and look-up table method. The estimated antenna pattern map can be better and the estimation of notch is more accurate. The validity of the proposed method is verified by the analysis of the measured data. The results show that the estimated antenna pattern which is drawn using the reflecting coefficient of the ground surface is very close to the measured antenna pattern painted on the experimental data.