| With the rapid development of computer technology, we have entered the era of information. With the help of intellect machines, the truth that a lot of manual work has been taken over has, to some degrees, set us human free from trifles. Differed from traditional perception technologies, Wireless Sensor Network (WSN) can be freely scattered, self-organized and capable of information processing. Wireless node positioning is one of the most important supporting technologies to WSN. In most of applications of WSN, the data collected by sensors is meaningless, unless being related to the position where the data has been collected. In order to describe what particular thing has happened where and when and make the positioning and chasing of out-of-system objects possible in further, sensors in the WSN must known their own exact locations.Nowadays, research into node positioning technology of WSN has been carried out just a few years. There're already several kinds of mature positioning technologies, with many more being brought up and improved by academies and researchers all over the world.According to whether the distance among nodes should be measured or not, positioning methods in WSN can be divided into two categories-Range-Free Positioning Method and Range-Based Positioning Method. In Range-Free Positioning Method, no accurate distance or angle information must be provided; only connection among nodes should be given, to calculate the approximate location of nodes within WSN, with the shortcoming that accuracy is not so good. In Range-Based Positioning Method, such as RSSI, TOA, TDOA and AOA, either point-to-point distance or angle information should be measured at first. Then with the help of mathematical models such as three-edge measurement, three-angle measurement and maximum-resembling estimating, node positions could be generated. Due to distance constraints, the accuracy has been significantly leveled up. With the requirement of high accuracy positioning grows bigger and bigger, Range-Based positioning methods, which provide more acceptable accuracy than Range-Free positioning methods, have been bringing more and more attention from people.Triangle Constraint Positioning Method (TCPM) is Range-Based. The most important discovery in this method is that the definition of nodes has fundamental influence accuracy of positioning. The overall notion is to, first, find all definite topologies between anchor and non-anchor nodes, then, calculate distance between anchor and non-anchor nodes based on these definite topologies and, at last, by three-edge measurement calculate the position of any node with distances to there or more anchor nodes can be definite.As is proved both in theoretical and practical aspects, is of many advantages among the Range-Based positioning methods. This method requires less connectivity in the WSN, provides incredible accuracy of node positioning, and shows less demand on beacon node density. Owing to low connectivity requirement, not too crowed nodes or too large measuring range is needed, which helps cutting both the economic cost and energetic expense. As proved in theory and test, the definition accuracy of TCPM will significantly grow with the accuracy of distance measurement. So, with the improvement and upgrading of distance measurement, TCPM is providing better and better accuracy. The setting process of beacon node is usually very complex. Low demand on beacon node density can make the deployment of whole WSN more efficient.Despite of many advantages, as the same to many other new technologies, some shortcomings that troubles TCPM do exist. There are some tricky problems in the original implementation of TCPM waiting to be fixed. In order to locate the nodes that do not apply to the definitions constraints, topology repairing must be carried out and topology fixing strategy be brought up.The original implementation of TCPM may not be able to locate some nodes even in the WSN of which all nodes meet definition constraints. The error will inevitably affect the use of three-edge measurement. Through detailed analyzing and accurately locating, the reason of this problem has been found. The process that calculating distance based on the triangle searching path is not completely right. The state-changing logic can't suit all types of topology. In this paper, all possible topologies which may be encountered during the distance generating process based on the triangle searching path are sorted into several types. Judging logic to every type of topology is given, correcting the original implementation of TCPM.By solving the problem in the original, all nodes that meet definition constraints can be located with expected accuracy. However, in practical WSN. a large number of nodes which doesn't meet definition constraints, in practical WSN. can't be positioned directly by this method. That is, the appliance of TCPM has been limited by is topology constraints. Topology fixing must be performed as compensation to TCPM. In this paper, a topology fixing method has been brought up. The general ideal of this method is to classify all the nodes into different kinds in the first place. Then calculate the position of wireless nodes the positions which are uniquely definite. After that, only part of all the nodes is left over. To make them definite, new nodes may be brought in to fix the network topology by being deployed into the appropriate areas.There are many possible kinds of topology fixing plan. Deploying new nodes once for all can speed up the deploying process of whole WSN; deploying new nodes by fixing the topology of non-definite node one by one can cut the expense on hardware significantly. The comparison between these two is mentioned in this paper too.In order to research and implement of topology fixing well, simulating program which carries out the topology fixing strategy is a must. Through emulating test on computer and evaluation on the test results, the effectiveness of topology fixing strategy mentioned in this paper is validated. |
PDF Full Text Request