Font Size: a A A

Research Of Optimization Localization Algorithm On Wireless Sensor Network

Posted on:2010-12-25Degree:MasterType:Thesis
Country:ChinaCandidate:L L ZhangFull Text:PDF
GTID:2178360308478402Subject:Signal and Information Processing
Abstract/Summary:PDF Full Text Request
In the Wireless Sensor Networks, the location information of sensor nodes is an important issue.The data collected by the nodes will make sense only when the locations are exact. Recently, the major techniques of sensor nodes localization are Range-baced and Range Free. Range-baced has higher precision while it requires additional hardware cost; Range-Free not only provides sufficient precision, but also reduces the complexities of hardware. Range-Free has more advantage on cost and power than the former. This thesis focuses on the algorithm of Range-Free in the WSN.Comparing with the characteristics of several typical algorithms, the disadvantages are generalized in the thesis by focusing on the distant unrelated algorithms which sets up the base work of later research.Based on the relationship between RSSI Range-Baced and classical multi-dimensional scaling algorithm, the RSSI-GA algorithm based on RSSI is proposed. It begins with the analysis of the distinctive among entities and the geometric constraint relationship of the distance among each node and establishes the mathematical model with the location of unknown nodes. The RSSI-GA algorithm uses genetic algorithm to work out the optimal path, thereby directly calculates node coordinates. Simulation result shows that:this algorithm greatly reduces the calculation costs and improves the Precision of localization.In the research of multi-sink node localization algorithm, the thesis adopts a clustering method of network topology organization and presents a multi-sink node clustering localization algorithm. Firstly, the algorithm devides the network into several sectors. Secondly, it calculates the location of the head node and upgrades it as the anchor node. Then, it works out the location of the node. This algorithm increases the stability and effectiveness of single-sink node, avoids the information conflict when nodes send the report at the same time and reduces overhead to transmit while improving the accuracy.
Keywords/Search Tags:Wireless sensor networks, Multidimensional scaling, Node location, Genetic algorithm, Clustering
PDF Full Text Request
Related items