We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Compensation strategy for distributed tracking in wireless sensor networks with packet losses.
- Authors
Wang, Yan; Qian, Chen; Liu, Xiangyu
- Abstract
Packet loss is inevitable in wireless sensor networks (WSNs). On the basis of the distributed Kalman filter (DKF) and consensus filter, we propose both biased and unbiased compensation strategies to improve tracking accuracy and reliability for WSNs' packet dropout problem. In the biased compensation strategy, undelivered data from neighbours is compensated by the data of node itself; and in the unbiased compensation strategy, weights of all nodes are updated when packet dropout happens. With these two compensation strategies, DKF can be valid under both mild and poor connectivity conditions. Furthermore, sufficient condition is given to guarantee the convergence of estimation error system. Simulation results show that statistics average estimation errors of the biased and unbiased compensation strategies reduce 37.541 and 37.542 % of state estimation respectively compared to the DKF when the packet loss rate reaches 33.3 %, which demonstrates that the proposed algorithms perform better in filtering packet dropout than the DKF.
- Subjects
WIRELESS sensor networks; DISTRIBUTED tracking; DATA packeting; KALMAN filtering; COMPUTER reliability
- Publication
Wireless Networks (10220038), 2015, Vol 21, Issue 6, p1925
- ISSN
1022-0038
- Publication type
Article
- DOI
10.1007/s11276-014-0884-x