Optimal update with one step out-of-sequence measurements for wireless multisensor network

• Published in: 2008 International Conference on Wavelet Analysis and Pattern Recognition Vol. 2; pp. 826 - 831
• Main Authors: Xiao-Liang Feng, Quan-Bo Ge, Cheng-Lin Wen
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.08.2008
• Subjects: Algorithm design and analysis; Correlation; Decorrelation; Equivalent measurement; Noise; One step delay; Out-of-sequence measurement; Sensor fusion; Sensor systems; Sensors; Wireless sensor network; Wireless sensor networks
• ISBN: 9781424422388; 1424422388
• ISSN: 2158-5695
• DOI: 10.1109/ICWAPR.2008.4635891

Influenced by the multihop transmission, the routing protocol and the dynamic topology of wireless sensor networks (WSN), the measurements sampled in order by the distributed sensors may arrive in the fusion center out of sequence when they are transported across WSN. In order to gain the optimal global performance, the fusion center should update the state estimate with these out-of-sequence measurements (OOSM). Up to now, the existing OOSM update methods mostly aim at the system with single OOSM or a kind of special multiple OOSMs system with some assumptions, so it results in their limited application abilities. For the defects of these current methods, the multisensor system with arbitrary asynchronous sensors, where multiple one-step-delay OOSMs reach the fusion center in one fusion period, is considered in this paper. Accordingly, a novel OOSM directly sequentially update algorithm, which is optimal in linear minimum mean square error (LMMSE), is proposed by adopting the technology of equivalence measurement and decorrelation idea. The academic analysis shows that the optimal A\ algorithm presented by Y. Bar-Shalom for multisensor system with two asynchronous sensors is a particular case of the proposed algorithm. Simulation example shows the feasibility and the effectiveness of the proposed method.