Research On Wireless Sensor Networks Localization For Meteorological Observation

At present, the meteorological data monitoring in China mainly depends on automatic meteorological stations across the country, but the distribution of automatic meteorological stations have disadvantage of relatively low distribution density, the deployment of a single, poor mobility deployment, shortcomings of higher deployment costs, etc. However, the wireless sensor networks can be deployed quickly, node density can be adjusted according to demand, and compared with automatic meteorological stations the cost is relatively low. Therefore, the application of wireless sensor networks (WSNs) is very suitable to meteorological observations. It can provide reference and help for traditional ground meteorological observation.Meteorological data acquisition needs location information support is very meaningful, but under normal circumstances, however, it’s not able to predict the position information of a large number sensor nodes in advance. The wireless sensor network’s self-localization algorithm has an important significance. This paper mainly focus on the localization technology of wireless sensor networks for intensive meteorological observation. The main contributions are as follows:(1) As long as the node of a wireless transceiver module can be based on the measurement of the distance between the nodes of RSSI (Received Signal Strength Indicator, RSSI) value, without additional hardware devices. However, due to wireless signal propagation influenced by many environmental factors, just rely on RSSI values to estimate the distance between nodes can cause error greatly. Through the use of the Crossbow’s MicaZ node set up the experimental platform, the exploration verify the relationship between nodes of distance and RSSI values. Experiments show that although the RSSI values do not apply to direct orientation, but can be used to determine the location of the relationship between nodes.(2) To improve the classic APIT localization algorithm based on range free. By leveraging the geometric relationship between the anchor nodes, the introduction of the Voronoi diagram superimposed narrow passage area radius increases caused by the nodes, thereby improving the positioning accuracy.(3) Due to the accuracy of the localization algorithm based on range-free less than range-based ones, improved MDS-MAP localization algorithm, can be used to some applications which have higher requirements. By introducing cluster node density, improved distributed MDS-MAP algorithm can deal with more complex networks. In the range phase, by quadrilateral distance correction, further optimize the localization algorithm accuracy.