Behrouz Safarinejadian - Shiraz University of Technology
Kazem Hasanpoor

In this paper, two mobile agent based distributed Kalman filter algorithms are proposed for optimal information fusion in sensor networks. In the proposed methods, mobileagents (MAs) play an important and fundamental role in information fusion. A mobile agent travels among the sensorsand collects the necessary information. The collectedinformation is carried to a fusion centre in which the optimal state estimates are computed. Finally, the proposed methodsare applied on a radar tracking system with five sensors arranged in two rings with some faulty sensors. Performance of the proposed methods will be compared with the centralized Kalman filter

Optimal information fusion, mobile agent, fault tolerance, moving agent, sensor network, Kalman filter