三星时差定位系统的有源校正算法

针对三星时差（TDOA）定位系统中，由星历误差和时间同步误差导致无法精确定位的问题，提出一种基于正交投影的两步迭代有源校正算法。该算法首先利用正交投影将时间同步误差从观测方程剔除，然后引入一个与辐射源位置和星历误差有关的中间变量，通过对中间变量的最优估计消除星历误差对定位精度的影响，最后利用中间变量求解出目标辐射源位置。仿真结果表明，所提算法定位性能的仿真值和理论值均可以达到克拉美-罗界（CRLB），当TDOA观测噪声较小且有3个以上参考源时，可以完全校正星历误差和时间同步误差对定位精度的影响。

A Tri Satellite TDOA Location Algorithm with Calibration Sources

A location algorithm based on tri-satellite time difference of arrival (TDOA) location system is proposed to calibrate the location error caused by the ephemeris error and synchronization error. Firstly the proposed algorithm eliminates the synchronization error by using the rectangular projection, then eliminates the impact of the ephemeris error on the location error via the optimum estimation of an intermediate variable which is related to the target source and ephemeris error, and finally derives the location of the target source based on the intermediate variable. The simulation results demonstrate that the simulated location error and theoretical location error of the proposed algorithm can both reach Cramer-Rao lower bound (CRLB). Besides, the proposed algorithm can eliminate the location error caused by the ephemeris error and synchronization error completely under the condition that the TDOA noise is much smaller than ephemeris error and the number of the calibration sources is more than or equal to three.