A Range-free 3D Integrated Localization Algorithm In Wireless Sensor Networks

Wireless Sensor Networks (WSN) with wireless network for communion is a network system which is composed by nodes spread in the stakeout area that are small in volume, cheap in price and low in calculation. Its characteristics are limited energy, self-organized network, huge scope network, the dynamic and so on. There are several research directions about WSN, they could be summarized as Localization, Routing, Security, and energy. Localization is the supportable technology of WSN, so there are more and more research base on location, and they presented many significative thought and many typical algorithms, such as Centroid algorithm and DV-Hop algorithm, but they all have their own limitations. Presently, researchers focus on 2D space. Although the 2D WSN localization algorithm is becoming mature, to consider with the application in reality, the node of WSN is always deployed in 3D space and its localization is not limited in 2D space. Due to measuring the temperature and the salinity in water or the route, the 2D localization algorithms can not satisfy in need in reality. Such as in marine, besides the plane coordinates, the depth of the marine also needs to be measured sometimes. What's more, we also need to consider about the height of mountain beyond the longitude and dimensionality. The 2D localization algorithms show its limitation at this time. So, to enhance the utility of the algorithms, it's more practical to study the 3D localization algorithms.After analysis, we found the truth is that the Centroid algorithm has the virtue of less energy consumption, but its accuracy of positioning is not precise enough at the same time. While the DV-Hop algorithm has precise accuracy of positioning, but needs more energy consumption. The CDLI algorithm synthesizing them together uses a parameter named K to decide the algorithm to be used and combines the virtues of less energy consumption in Centroid algorithm and the precise eaccuracy of positioning in DV-Hop algorithm. So, this thesis is to design the CDLI algorithm in 3D space which is 3D-CDLI algorithm. The 3D-CDLI algorithm is composed by two parts, the Centroid algorithm and the DV-Hop algorithm, and lead in a threshold by the connectivity of the network. When the amount of the beacon nodes not exceeding the area the unknown nodes jumping only once is more the threshold, the algorithm chooses the 3D-Centroid algorithm automatically. In the same manner, when the amount of the beacon nodes not exceeding the area the unknown nodes jumping only once is less the threshold, the algorithm chooses the 3D-DV-Hop algorithm.The threshold that is the foundation for 3D-CDLI algorithm to choose the 3D-Centroid algorithm or the 3D-DV-Hop algorithm is not same in different network environment, and the experiment proves that the selection of the threshold has some connection with the connectivity of network. When the connectivity of network is higher, the threshold should be chosen less for more nodes orientating by 3D-Centroid algorithm, which plays a part in depressing the energy consumption; To the other way round, when the connectivity of network is lower, the threshold should be chosen more for more nodes orientating by 3D-DV-Hop algorithm, which makes the accuracy of positioning higher. As a whole, the choice of the threshold should be chosen evenly in stead of that the majority unknown nodes choose the same localization algorithm. Otherwise, the predominance of the two algorithms can not be shown and we can not synthesize the virtues of the two algorithms together.For the analysis of the 3D-CDLI's precision shown before, the thesis uses MATLAB7.0 to carry on the experiment. The experiment shows that the localization precision of the 3D-CDLI algorithm is higher 3.5%-6.5% than the DV-Hop algorithm. During the process of locating the unknown nodes, a part of unknown nodes use the 3D-Centroid algorithm for locating, and it reduces the cost of the communication in network. The experiment advanced in the thesis of the 3D-CDLI algorithm draws the conclusion of that this algorithm has the virtue both in 3D-Centroid and in 3D-DV-Hop algorithm, which shows that the algorithm enhances the localization precision and reduces the cost of the network. The 3D-CDLI algorithm is designed by synthesizing special algorithms and the virtues themselves for suffering different application environment. I wish the design in this thesis could be taken as reference in other researches, meanwhile keeps its virtue moving on in the reality.