1 Introduction
With the development of computer and communications
disciplines, wireless sensor network technology has a
speeding development since it’s proposed, whether in
theoretical study or engineering fields. In the wireless
sensor network, due to the impact of the network itself and
environmental factors, Information transferring process in
sensor network may appear delay, out-of-sequence even
dropout, so that the information can’t reach the processing
center promptly, thus make the system can’t process the
information effectively. In order to solve this problem, in
recent years, many scholars have put a lot of effort to study
these issues, including network transmission packet loss
rate, delay tolerant control method
[1]
, delay measurement
filter design
[2]
, stability analysis and so on
[3]
.
Research has made some important results for the delay
measurement filtering problems, which can be divided into
the following categories: Re-filtering method
[4,5]
, Discard
the delay measurement method
[6,7]
, Equivalent
measurement method
[8-12]
, The new interest reconstruction
method
[13,14
]etc. The above methods are designed for linear
time-varying systems, however, these filtering methods
either have poor real-time, or need a large amount of
computation. Literature [15,16] make use of the linear
time-invariant systems’ system parameter invariant
properties, designed a class of delay measurement filter.
Although when the delay measurement arrived, this method
only needs to update the estimated value of the system state
by calculating the weighted sum, however, the weighting
coefficients of the measurements need to be recalculated in
each update. Literature [17] designed a new class of delay
measurement filtering algorithms for the linear
This work is supported by National Nature Science Foundation under
Grant
61304258,61273075,61173133,61371064.
*
Corresponding author
time-invariant system, which based on the
pseudo-measurement model library. Under the condition of
a certain maximum number of delay steps, this method
establishes a pseudo-measurement model library between
the possible reached filter measurements and current
filtering time system state by using system parameters
invariant properties. By selecting the pseudo-measurement
model and filter parameters to simplify delay measurement
filtering process. However, the literature [17] only consider
the single sensor system, and multi-sensor system is
widespread, especially in wireless sensor networks, how to
fuse the measurements obtained from the local sensors is a
very interesting research topic. For the multi-sensor linear
time-invariant system, this article proposes a new fusion
filtering method based on pseudo-measurement model
library. First, in the condition of the maximum delay steps,
it builds each local sensor corresponding
pseudo-measurement model library; Second, each local
sensor gets its current state estimate and estimation error
covariance by using pseudo-measurement model library
filtering method; Third, fusion center based on the
weighted matrix method to obtain a global estimation
results. Simulation examples demonstrate the effectiveness
of this method.
This paper remain structured as follows: The system model
is given in the second part. Kalman filter based on local
pseudo-measurement model library design see Part III.
Fusion filter based on a weighted matrix see Part IV. The
fifth part is a computer simulation. The sixth part is the
conclusion.
2 System Description
Consider the following linear time-invariant systems
() (, 1)( 1) (, 1)
() ()() (), 1,2, ,
jj j
xk Fkk xk wkk
kHkxkvk j N
=−−+−
®
=+=
¯
"
˄
1
˅
Fusion Filtering Method Based on Pseudo-measurement Model Library for
Multisensor Systems with Delay Measurements
Ye Haihong
1
, Wen Chenglin
*1
, Feng Xiaoliang
1,2
1. College of Automation, Hangzhou Dianzi University, Hangzhou 310018
E-mail: wencl@hdu.edu.cn
2. College of Electrical Engineering, Henan university of technology, Zhengzhou 450001, China
E-mail: fengxl2002@163.com
Abstract: Information transferring process in sensor network will appear delay, out-of-sequence even dropout. How to
make full use of such information is extremely important to improve the state estimation accuracy. In this paper, it
presents a new method, which can effectively improve the estimation accuracy .Firstly, a local pseudo-measurement
model library is established to deal with the measurement with several steps delay, then sent the local estimates to the
fusion center; Secondly, in the fusion center, under the unbiased linear minimum variances weighting fused rule, the
fusion Kalman filter based on a matrix weighted fusion algorithm processes the local estimates, thus obtain the global
results. Simulations verify the effectiveness of this method.
Key Words: Wireless sensor network; Delay measurement; weighted fusion filter; Pseudo-measurement model library.
5162
978-1-4799-3708-0/14/$31.00
c
2014 IEEE