Font Size: a A A

The Research On Automatic Location Algorithm Of Wireless Sensor Network Based On Multidimensional Scaling

Posted on:2008-10-13Degree:MasterType:Thesis
Country:ChinaCandidate:J TianFull Text:PDF
GTID:2178360272468767Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
Wireless Sensor Network is one of hotspot of the communication network research. What is the WSN? It is a new pattern network system that is automatically constructed by lots of nodes which have the abilities of wireless communication and computation. The nodes are distributed randomly. The network system is extensively applied to national defense, environment inspection, transport management and so on. We can say it is a fire-new network technology. At present, the technology is still in the phase of research. The research of orientation of network nodes is very important in the system of the WSN research. In this paper, we investigated the existing orientation arithmetic. And a new arithmetic with more efficiency on the orientation of network nodes is proposed.These days, the orientation arithmetic of WSN is divided into two kinds: range-based and range-free. The mechanism of range-based is entirely based on the distance measurement. In this mechanism, the nodes must have strong communication abilities, so there is high standard for the hardware. This mechanism is not fit for WSN that have the character of small volume, low power. And then, the range-free mechanism appears, this is a kind of estimate arithmetic, and low requirement to nodes'capabilities.In this paper, MDS-MAP algorithms that adopt range-free mechanism are deeply analyzed and compared each other. And combining the ASP algorithm that is applied in ad-hoc network, a new self-orientation algorithm is proposed. There are processes of this algorithm: (1) Plot species maps in whole network. (2) Divide the nodes in one species into different levels according different Euclidean distances, and constitute approximate distance matrixes. (3) Get relative species maps by applying classic MDS on every approximate distance matrixes. (4) Improve the precision of orientation by importing refinement function to revise relative species maps. (5) Unite species maps in the absolute coordinates system. After explaining the design of the algorithm detailedly, we make an emulational experiment on MDS-MAP(S, R) algorithm by MATLAB. And then analyze the result of the experiment in the error of orientation and the complexity of computation. Finally, the characters of accuracy and high efficiency of this new algorithm are validated.
Keywords/Search Tags:Wireless Sensor Network, orientation arithmetic, MDS-MAP, species maps, global maps
PDF Full Text Request
Related items