混合卡尔曼滤波在外辐射源雷达目标跟踪中的应用

为了提高无迹卡尔曼滤波(UKF)中误差协方差矩阵的估计精度，该文结合外辐射源雷达目标跟踪模型，提出了一种混合卡尔曼滤波(MKF)算法，首先通过UKF对目标状态进行一次后验估计，然后重新建立一个观测方程，把UKF滤波输出的状态估计值转化为新建观测方程的量测值，并通过线性卡尔曼滤波对状态进行二次最优估计。实验结果表明，与扩展卡尔曼滤波(EKF), UKF相比，MKF明显提高了外辐射源雷达目标跟踪的精度。 

Application of Mixed Kalman Filter to Passive Radar Target Tracking

To improve the estimation accuracy of the error covariance matrix in Unscented Kalman Filter (UKF). With the passive radar target tracking model, a novel Mixed Kalman Filter (MKF) is proposed, Firstly, the UKF is used to conduct a posteriori estimate for target state, and then re-establish a measurement equation, the posteriori estimated value of state by UKF is transformed into a measured value of the new measurement equation, and through linear Kalman Filter the state is best estimated secondly, improving the precision of target state estimation. Experimental results indicate that MKF algorithm significantly improves the performance of passive radar target tracking, compared with the Extended Kalman Filter (EKF) and UKF.