| 一种连续型野值的识别与修正滤波算法 |
| |
| 引用本文： | 冯翔,费向东.一种连续型野值的识别与修正滤波算法[J].四川大学学报(自然科学版),2010,47(1). |
| |
| 作者姓名： | 冯翔 费向东 |
| |
| 作者单位： | 四川大学图形图像研究所,四川大学计算机学院 |
| |
| 摘 要： | 通过对量测数据新息的统计数学特性的分析,借助于主成分分析法的思想,根据矩阵的最大特征值反映矩阵的主要信息这一特点,提出了一种检测量测数据中是否含有野值的判别方法.此基础上,修正了Kalman滤波器的状态估计算法.使该算法在出现连续型野值时,能明显地降低误差,提高滤波精度.仿真结果证明了该方法的有效性.
|
| 关 键 词： | Kalman滤波 野值 最大特征值 新息 |
| 收稿时间： | 4/11/2008 2:01:57 PM |
| 修稿时间： | 7/1/2008 12:00:00 AM |
An improved filter algorithm on distinguishing and dealing with continous outliers |
| |
| Feng Xiang and Fei Xiang-dong.An improved filter algorithm on distinguishing and dealing with continous outliers[J].Journal of Sichuan University (Natural Science Edition),2010,47(1). |
| |
| Authors: | Feng Xiang and Fei Xiang-dong |
| |
| Institution: | College of Computer Science,Sichuan University |
| |
| Abstract: | The measurement innovation is analyzed statistically based on the largest eigenvalue which presents the key matrix informain according to the principle component analysis idea.A kalman algorithm improved on state estimation is developed to reduce the error and increase the filter accuracy by distinguishing and dealing with outliers.The simulation results prove the efficiency of the method. |
| |
| Keywords: | Kalman filter outliers largest eigenvalue innovation |
|
| 点击此处可从《四川大学学报(自然科学版)》浏览原始摘要信息 |
| 点击此处可从《四川大学学报(自然科学版)》下载免费的PDF全文 |
