Journal of Systems Engineering and Electronics
Vol. 23, No. 1, February 2012, pp.1–9
New conflict representation model in generalized power space
You He
1
, Lifang Hu
1,2,∗
, Xin Guan
1,3
, Deqiang Han
4
, and Yong Deng
5
1. Research Institute of Information Fusion, Naval Aeronautical and Astronautical University, Yantai 264001, P. R. China;
2. Navy Armament Academy, Beijing 102249, P. R. China;
3. College of Electronic Science and Engineering, National University of Defense Technology, Changsha 410073, P. R. China;
4. Institute of Integrated Automation, Xi’an Jiaotong University, Xi’an 710049, P. R. China;
5. School of Electronics and Information Technology, Shanghai Jiaotong University, Shanghai 200240, P. R. China
Abstract:
The study on alternative combination rules in Dempster-
Shafer theory (DST) when evidences are in conflict has emerged
again recently as an interesting topic, especially in data/information
fusion applications. The earlier researches have mainly focused
on investigating the alternative which would be appropriate for
the conflicting situation, under the assumption that a conflict is
identified. However, the current research shows that not only the
combination rule but also the classical conflict coefficient in DST
are not correct to determine the conflict degree between two pieces
of evidences. Most existing methods of measuring conflict do not
consider the open world situation, whose frame of discernment
is incomplete. To solve this problem, a new conflict representa-
tion model to determine the conflict degree between evidences
is proposed in the generalized power space, which contains two
parameters: the conflict distance and the conflict coefficient of
inconsistent evidences. This paper argues that only when the con-
flict measure value in the new representation model is high, it is
safe to say the evidences are in conflict. Experiments illustrate the
efficiency of the proposed conflict representation model.
Keyw ords: Dezert-Smarandache theory (DSmT), conflict evi-
dence, Dempster rule of combination, information fusion.
DOI: 10.1109/JSEE.2012.00001
1. Introduction
Dempster-Shafer theory (DST) is widely used in many
fields such as information fusion and decision-making
[1]. In the framework of DST, information fusion re-
lies on the use of combination rules allowing the belief
Manuscript received April 8, 2010.
*Corresponding author.
This work was supported by the National Natural Science Founda-
tion of China (60572161; 60874105), the Excellent Ph.D. Paper Au-
thor Foundation of China (200443), the Postdoctoral Science Foundation
of China (20070421094), the Program for New Century Excellent Ta-
lents in University (NCET-08-0345), the Shanghai Rising-Star Program
(09QA1402900), and the Ministry of Education Key Lab of Intelligent
Computing & Signal Processing (2009ICIP03).
functions to combine different propositions. A crucial role
in DST is played by the Dempster’s rule of combination
(DRC) that has several interesting mathematical proper ties
such as commutativity and associativity. However, illog-
ical results may be obtained by DRC when the collected
evidences highly conflict with each other [2].
Since then, many authors used Zadeh’s example to criti-
cize DST either as a whole, or as a motivation for con-
structing alternative combination rules, which mainly di-
ffer from the way possible conflicts are managed [3−21].
Recently, due to the increasing applications of DST in in-
telligent fusion processes, DRC and its alternatives have
been under the microscope again (e.g., [3,4,7]).
Overall, the existing studies can be divided into two ma-
jor categories.
The first is how to use DRC. There exist three kinds
of circumstances. One is how to use the close world of
DRC including the improvement of combination rules and
the modification of the primal data sources [3−14]. An-
other is to support the open world. References [15,16]
are representative in this method. The third is to sup-
port the integration of close and open worlds. So the
hyper-power set and the general power set are put forward
[17−19].
The second is how to reasonably measure the conflict
between evidences [20−22]. Reference [20] proposed an
alternative method to measure the conflict among beliefs
using a pair of values, and the mass of the combined belief
allocated to the emptyset before normalization and the d is-
tance between betting commitments. Reference [ 21] paid
more attention to the essence of conflict in their researches.
Their studies have mainly focused on investigating reasons
for conflicting situations and choosing alternative rules
which would be appropriate for the corresponding con-
flicting situation. This is also the primary idea of this
paper.
剩余8页未读，继续阅读
资源评论
