一种基于高斯牛顿迭代的单站无源定位算法

由于初始估计误差大、可观测性弱,且可得到的观测量受限等特点,对运动辐射源的快速单站被动定位一直是个难题。针对单站无源定位特点,对IEKF(iterated extended Kalman filter)算法进行改进,该算法对IEKF中的测量更新按照高斯牛顿迭代方法进行修正,从而减小非线性滤波的线性化误差,改善滤波算法性能。所提算法和UKF(unscented Kalman filter)I、EKF以及EKF(extended Kalman filter)算法的仿真比较表明,提出的算法可以用更小的计算量得到和UKF相当甚至更好的定位性能,在定位精度和收敛速度上明显优于IEKF以及EKF。

Gauss-Newton iteration based algorithm for passive location by a single observer

Due to large initial estimation error,low observability and limited achievable measurements,etc.,fast passive location of the moving emitter by a single observer is still intractable.Considering the peculiarity of passive location,IEKF(iterated extended Kalman filter) is modified by providing a new measurement update based on Gauss-Newton iteration,thus the linearity error is reduced and the filtering performance is improved.Computer simulation,in which the new iterative algorithm is compared with UKF(unscented Kalman filter),EKF and IEKF,is conducted and the results demonstrate that the proposed algorithm can achieve an equivalent or even better performance than UKF in location accuracy and convergence speed,and is markedly superior to EKF and IEKF.