Research On High Accuracy Passive Localization Using TDOA And FDOA With Multiple Receivers

In modern high-tech warfare, the passive localization system using multiple receivers has been widely studied by experts and scholars because of its advantage to locate the target stealthily, timely and accurately. In order to conquer the problems that affect system localization performance such as the receiver position and velocity errors or the restriction of the receiver amount and achieve fast and high-precision location of non-cooperative target, this dissertation deeply studies the localization algorithm based on receiver parameter errors correction and the sequence localization algorithm with restrict amount of receivers. The researches in this dissertation provide some guidance for the application of passive localization system and the main contents can be described as follows:1. The localization algorithm base on no-calibration receiver parameter errors correction is studied. In order to reduce the bad impact of receiver parameter errors in localization system, the localization algorithm using no calibration emitters based on Chan algorithm is discussed on the basis of location model and Cramer-Rao Lower Bound with receiver parameter errors. The localization performance degradation in the algorithm is pointed out because of the pseudo-linear processing of the observation equations. Then, the localization algorithm using no calibration emitters based on multidimensional scaling analysis is studied. The TDOA and FDOA scalar product matrixes are established and the linear equations related to target parameter is set up by using matrix eigenvalue decomposition. The influence of the statistical property of receiver parameter errors is added into the least squares weighting matrix to correct the error and locate the target. Algorithm analysis shows that the MDS localization algorithm using one-step is able to improve localization accuracy by reducing the performance loss caused by the two-step solution to the target. Simulation results show that the MDS localization algorithm achieves good performance for both far-field and near-field targets and obtains higher localization accuracy with large TDOA and FDOA measurement errors and receiver parameter errors.2. The localization algorithm base on receiver parameter errors correction using calibration emitters is studied. In order to reduce the bad impact of receiver parameter errors and improve the localization accuracy, the localization algorithm with auxiliary calibration emitters is discussed. Two different cases of using calibration emitters with accurate parameters or erroneous parameters are introduced. The corresponding location model and CRLB are given and the algorithms are analyzed. Algorithm analysis shows that, no matter the calibration emitter parameters are accurate or not, using auxiliary calibration emitters is able to reduce the bad impact of receiver parameter errors effectively. Then, the localization algorithm based on the self-calibration of receiver parameter errors is studied. The location model of receiver parameter errors self-calibration is set up and the CRLB is deduced. The linear equations related to receiver parameter errors are established by using the TDOA and FDOA measurements among the receivers. The linear minimum mean square error estimation of receiver parameter errors are obtained and the calibrated value of receiver parameters are used in the localization algorithm using no calibration emitters to achieve target parameter estimation. The errors correction and target location without auxiliary calibration emitters is accomplished. Simulation results show that the self-calibration localization algorithm is able to achieve good localization performance for both far-field and near-field targets.3. The localizations algorithm using TDOA and FDOA sequence with restrict amount of receivers are studied. In order to solve the instantaneous location problem with restrict amount of receivers(less than 4 receivers), the location models to space uniform motion target by three fixed/moving receivers(TDOA/TDOA&FDOA) are set up and the corresponding CRLBs are deduced. Then, the hypothesis value of the target parameters(initial position and velocity) are introduced and the linear equations related to hypothesis value errors are built by using Taylor series expansion. The weighted least squares estimations of the target parameters are obtained and the target location is accomplished without searching or iteration operations. Simulation results show that the TDOA/TDOA&FDOA sequence localization algorithms achieve good localization performance with small TDOA and FDOA measurement errors and receiver parameter errors.