Study on Node Localization Algorithm Based on Mobile Beacon in Wireless Sensor Network
|School||South China University of Technology|
|Course||Control Theory and Control Engineering|
|Keywords||wireless sensor network nodes localization mobile beacon extended Kalmanfilter DV-Hop algorithm|
Wireless sensor network (WSN) has wide application foreground in militaryreconnaissance, environmental monitoring, agricultural production, intelligent transportation,logistics management etc. with the development of micro-electronics, computers and wirelesscommunication technologies. The premise of most applications is to obtain locationinformation of the sensor nodes, so nodes localization is one of the key technologies forWSN.Present localization algorithms are most based on static beacon nodes to positioning.And the localization accuracy depends on beacon node density, which increases cost of thenetwork. While the mobile beacon can reduce the network cost significantly and also canimprove the network topology structure effectively to increase the localization accuracy,comparing to the static beacon. So the mobile beacon assisted localization algorithms is aresearch focus of the node localization technology in WSN.In this thesis, we focused on the study of node localization algorithms based on mobilebeacons, and the main work as follows:Firstly, we summarized the architecture, features, key technologies, and the practicalapplication for WSN briefly. Also introduced the localization mechanism based on staticbeacons and analyzed the typical range-based and range-free algorithms for node localization.The principle, characteristics and typical mobile model of the assisted localization mechanismbased on the mobile beacon are emphatically introduced.Secondly, according to range-based localization, a mobile beacon assisted localizationalgorithm based on the extended Kalman filter (EKF) was proposed. In this algorithm, themobile model was designed according to the triple cover theory. The mobile beacon travelthrough the sensing area and RSSI ranging method was adopted, and then the initial locationcoordinate of the unknown node is estimated by the weighted centroid algorithm, the EKFwas introduced to improve the localization precision at last. The simulation results showedthat the EKF algorithm can obtain higher localization precision and network coverage with alower network cost. And the performance of the EKF algorithm was obviously superior to thetraditional weighted centroid algorithm. Lastly, according to range-free localization, an improved DV-Hop algorithm based on themobile beacon was proposed. In the new algorithm, a mobile beacon was introduced toinstead of static ones, improved the estimation method of average hop distance, and adoptedthe least square method based on optimal beacon group to achieve the location estimation ofthe unknown nodes. The theory analysis and simulation results showed that the improvedalgorithm had a lower calculation, a higher localization precision and a better stabilitycomparing with the traditional DV-Hop algorithm.