Research On MDS And Sequence Localization Algorithm In Wireless Sensor Network

Wireless sensor networks, which has been made viable by the convergence of microelectronics, wireless communications and digital electronics, has received increasing attentions in recent years, WSN has been applied widely in many fields, including search and rescue, disaster relief, target tracking and environment monitoring, and so on. For most applications in WSN, the location information of the monitored incidents and the nodes is very important for WSN. Data without location information has limited use. Accurate location estimation can also supply helpful aids for other services, such as routing, information processing, tasking and querying.This paper to deal with the sensor networks with anisotropic or complex terrain conditions, focus on the distributed sensor position method based on multidimensional scaling technique. We introduce a distributed MDS improved algorithm and a novel sequence-based localization technique, which is suitable for Multi-Agent tracing.The paper fulfills the following work:Firstly, we study some situations (for example the sensor network with anisotropic topology), where most existing sensor positioning methods tend to fail to perform well. We propose a distributed sensor positioning method based multidimensional scaling technique to deal with these challenging conditions. Multidimensional scaling and coordinate alignment technique are applied to recover positions of adjacent sensors. The estimated positions of anchors are compared with their true physical positions and corrected. The positions of other sensors are corrected in turn. The transformation from relative coordinate to absolute coordinate is realized finally.Secondly, a novel sequence-based localization technique for wireless sensor networks is proposed. The location space division into distinct regions by the perpendicular bisectors of lines joining pairs of reference nodes. The distinct regions with unique sequence number that represents the ranking of distance from the reference nodes to that region. First we derive expressions for the maximum number of location sequences. Second we present an algorithm to construct the location sequence table. Finally we give the sequence based on the following method: the location of the unknown node is estimated by first determining its location sequence using RSS measurements of RF signals between the unknown node and the reference nodes, and then searching through a predetermined list of the sequence table, to find the region represented by the same one. In simulations we mainly research three beacons in localization areas. The simulation results validate the effectiveness of the proposed algorithm.