Thursday, February 10, 2011
Localization for Mobile Sensor Networks (Part 1)
Localization is an important step for many applications such as vehicle tracking, environment monitoring and location-based routing. Although there has been a lot of work addressing localization, they do not consider scenarios where the nodes in the network experience non-uniform and uncontrolled motion. Although localization in the latter case appears to complicate the process, this paper proposes a method which exploits the mobility of the nodes to estimate their locations. The paper uses a sequencial monte carlo method which recursively filters the location samples of the nodes and concieves a resonable location estimate.The central idea of the approach is to use the velocity of the nodes to estimate their potential posterior locations and recursively filter impossible estimates using new observations that nodes recieve from seeds.
Subscribe to:
Post Comments (Atom)
The mobility of the sensor nodes have been used efficiently to find the accurate and precise localization of the nodes. The Monte Carlo localization method is so efficient algorithm that it has also been used in robotics to locate a mobile robot. The range information is obtained from the anchor nodes that are one and two hops away from the unlocalized node, by which the authors show that they can improve the accuracy of localization. Though this mechanism shows improved accuracy, the computation cost for localization is high.
ReplyDeleteI feel that WSNs usually aren't mobile just because of the design constraint of a sensor node being battery powered. So it seems as though no matter what type of localization technique you use, if a node is going to be considered mobile then it will probably consume all of its energy relatively fast because it will keep having to update its position.
ReplyDelete