DocumentCode :
2839757
Title :
An optimal sequential filter for the linear system with correlated noises
Author :
Feng, Xiao-Liang ; Ge, Quan-bo ; Wen, Cheng-lin
Author_Institution :
Inst. of Inf. & control, Hangzhou Dianzi Univ., Hangzhou, China
fYear :
2009
fDate :
17-19 June 2009
Firstpage :
5073
Lastpage :
5078
Abstract :
The sequential filter is a kind of useful estimate fusion algorithm. Traditional sequential filters are mainly utilized for the linear systems with the assumption of uncorrelated noises. Recently, some effective algorithms have been presented for the linear system with multiple sensors and correlated noises. Unfortunately, they can´t perfectly solve the optimal filtering estimate in linear minimum mean square error (LMMSE) for the systems with correlative measurement noises which are also cross-correlated with the process noise one time step apart and seldom discussed in present researches. And a novel sequential filter, which is optimal in LMMSE, is proposed in this paper for the linear dynamic system with correlative measurement noises which are also cross-correlated with the process noise one time step apart. The kernel of the novel optimal sequential filter is to decorrelate these correlations by use of the equivalent measurement function. Synchronously, the computer simulations are also presented to illustrate its performance.
Keywords :
correlation methods; filtering theory; least mean squares methods; linear systems; noise; correlated noises; correlative measurement noises; linear dynamic system; linear minimum mean square error; optimal sequential filter; Computer simulation; Decorrelation; Filtering; Kernel; Linear systems; Mean square error methods; Noise measurement; Nonlinear filters; Sensor systems; Time measurement; Correlated Noises; Decorrelation; Information fusion; Linear Minimum Mean Square Error(LMMSE); Sequential Filter;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Control and Decision Conference, 2009. CCDC '09. Chinese
Conference_Location :
Guilin
Print_ISBN :
978-1-4244-2722-2
Electronic_ISBN :
978-1-4244-2723-9
Type :
conf
DOI :
10.1109/CCDC.2009.5194967
Filename :
5194967
Link To Document :
بازگشت