Research On SDP-based Node Localization Algorithms For Wireless Sensor Network

The wireless sensor network (WSN) is made of a large number of low-cost, low-power sensor nodes with capacity of sensing, communicating and information processing. Its purpose is to perceive the information of objects in the coverage area of WSN, then send it to the observers through a multi-hop way. The application of wireless sensor network is very extensive, including environmental monitoring and forecasting system, health care, household use and military surveillance. Monitoring information of the target to be useful only in its location is known that resulting in a positioning algorithm. Obviously, the accuracy of the node localization is the prerequisite for the WSN-based applications. Therefore, how to determine the location of sensor nodes is a hot issue.A new localization algorithm based on semidefinite programming (SDP) and multidimensional scaling (MDS) is present in this paper to adapt to the dynamic topology of WSN. It is referred to as:Localization based on MDS and SDP (DLSM). The main idea of the algorithm is to construct several local positioning regions, MDS localization algorithm or SDP positioning algorithm is chose to construct the local relative map in each local positioning regions, and merge the local maps together to form a global map. The estimated position of sensors will be obtained by using the information of anchors. Simulation results show that DLSM can get a high sensor positioning accuracy and reduce the communication, especially under the non-uniform condition, with a small number of anchors.In view of the sensor node energy limited characteristic, the rigid graph theory has been used for the node localization technology in this article. It is combined with semidefinite programming(SDP) to propose a new kind of centralized localization algorithm:SDP localization based on rigid graph (SDP-R). Two different scenes have been used for simulation experiments, and the results show that the SDP-R algorithm can reduce the traffic in the process of the localization with high precision, and it can be applied in sensor network with barriers or empty.