一类高斯混合噪声干扰的系统拆分与滤波方法设计

文章针对这样一类系统:状态方程与观测方程中含有均值为零的高斯噪声且含非零均值的高斯噪声,提出了一种基于系统噪声拆分的滤波算法.首先将状态方程与观测方程分别拆成两个仅含有一个噪声的方程;然后,对两个方程分别进行滤波;最后,进行融合.仿真结果表明相比直接使用卡尔曼滤波方法,系统噪声拆分方法提高了滤波的估计精度.

In this paper, a filtering algorithm based on system noise split is proposed for such a class of systems: the equation of state and the observation equation contain Gaussian noise with zero mean and non-zero mean. First, the state equation and the observation equation are divided into two equations with only one noise;then, the two equations are filtered respectively;finally, the fusion is carried out. The simulation results show that compared with the Kalman filter method, the system noise resolution method improves the estimation accuracy of the filter.