Research On Node Localization Algorithms Based On Wireless Sensor Network

As a new technique, Wireless Sensor Network has become very popular in the researchfield. It covers sensor technology, embedded computing technology, distributed informationprocessing technology, communication technology and micro-monitor technology, and is ofgreat value to application in military, industrial, medical, transportation, environmentalprotection and many other areas. It can get kinds of sensor data.But they will be nonsensewithout location information. The GPS can locate the object precisely, but it’s too expensive.So we cannot make each of the nodes in wireless sensor network work with GPS. So the nodelocalization technique in wireless sensor network generates.Node Localization based on wireless sensor networks is different from traditionalwireless localization. It faces many challenges, such as self-configuration, dynamic topologychange, surrounding environment influence and stringent constraints in computation,communication and energy resources.For wireless sensor network positioning technology generally divided into two categories:one is based on the distance measurement (Range-Based) positioning algorithm, anothernon-Ranging (the Range-free) positioning algorithm. In contrast, the Range-Basedlocalization algorithm, after distance measurement estimation, the node location is highpositioning accuracy with appropriate positioning algorithm. The advantage for theRange-free positioning algorithm does not require additional hardware to measure thedistance between the nodes. So it is more and more popular now.This paper focus on technology development project of Jilin Province “Based onwireless sensor networks located in the tracking system underground staff”(20090425),Basedon wireless sensor network technology, locate the unknown node, we analyze and simulate theMaxlikelyhood algorithm, and apply it into the real world for finding a good result.The major work in this paper can be separated to two parts:The first Part: We do some research on the Maxlikelyhood Localization Algorithm whichis widely used in the node localization. We found that, in the equations of the Algorithm, theparameter in the last equation is very important to the localization result. In this paper, theanchor node whose location and range substitute the last equation called reference anchornode.According to the result of this experiments, we found that: Through the matlab6.5simulate experiment, when the range error for the reference anchor node is not zero, if thereference anchor node is in the center of the localization area, the localization error is the least, if the reference anchor node is on the edge of the localization area, the localization error is theMax; when the range error for the reference anchor node is zero, if the reference anchor nodeis in the center of or near the localization area, the localization error is the Max, if thereference anchor node is on the edge of the localization area, the localization error is the least.And under the same situation, the localization error for the later is less than theformer.According to the result of the experiment above; we can get better localization result ifwe choose the appropriate reference anchor node.The second Part: To prove the result above, we do the real localization experiment in theempty environment.To get the relationship between receive power and distance, we have to do theexperiment to get the receive power data according to the varied distance before thelocalization experiment. During the localization experiment, we will calculate the distancebetween the nodes with the relationship. The experiments result shows that the localizationerror is much smaller when the ranging error of the reference node is much smaller. But thelocalization error doesn’t have any change because of the location of the reference node.