Font Size: a A A

Wireless Sensor Network Node Positioning Technology To Achieve

Posted on:2009-01-31Degree:MasterType:Thesis
Country:ChinaCandidate:Y M SongFull Text:PDF
GTID:2208360245461328Subject:Access to information and detection technology
Abstract/Summary:PDF Full Text Request
With the profound development and wide application of wireless sensor networks, they are attracting more and more attention. A wireless sensor network is harmony worked self-organizing with thousands of nodes randomly distributed. The information of positions of nodes is crucial to the monitoring of sensor networks, thus, estimating nodes'positions is fundamental function of wireless sensor networks. Localization approaches based on Multidimensional Scaling (MDS) could work efficiently, and a robust estimation method based on Least Absolute deviations (LAD) rule could work better, so this dissertation focuses on these two methods, and does some work to probe into implement of the platform of location system.First of all, based on the study of plenty of related literatures, current research status of location algorithms and systems for wireless sensor networks are summarized. The categories of recent representative localization approaches are discussed.Then, this dissertation compares classical MDS algorithm which using centroid of network as reference point with MDS algorithm which using arbitrary network node as reference point, and extends to a unified framework, which is applied to any MDS algorithm with arbitrary point of the plane is used as reference point, and also derives the Cramér-Rao bound of the estimation error in wireless sensor networks.Furthermore, this dissertation modifies the cost function to LAD form based on discussion of problems of MDS algorithm. By utilizing semidefinite programming (SDP) relaxation model, a robust node localization method which is based on LAD rule is proposed. Simulation results prove that the proposed approach is more effective and performance of location is improved compared with conventional MDS algorithm in the influence of pulse noise.At last, this dissertation does some tentative work to the experimental platform of location system. To obtain distance measurements, S51 microcontroller is used to control TOA (time of arrival) of altrasonic signal, and wireless transceiver nRF905 is used to transmite trigger signal. Experiment results are also showed in this dissertation.
Keywords/Search Tags:Wireless sensor networks, node localization, MDS, LAD estimation, altrasonic signal
PDF Full Text Request
Related items