Research On Key Technologies In Localization Of Wireless Sensor Networks

A Wireless Sensor Networks (WSNs) consists of many tiny-sized sensor nodesthat have computation power, communication capability, and sensing functions.Because of its powerful and versatile functions, the WSNs have been widely used inmany areas such as military affairs, patient healthcare, and environment inspection,etc. In many applications, location information of the sensor node is very important.In addition, the location information of the deployed sensor nodes can be desirablebesides the sensed data. Hence, how to find the locations of the sensor nodes becomesone of the most critical issues in WSNs.In order to achieve the positioning of WSNs, the commonly used positioningtechnology includes the received signal strength indication (RSSI), the time of arrival(TOA), the time difference of arrival (TDOA), the angle of arrival (AOA) and hybridpositioning method. Combined with RSSI and AOA location technologies, an AOAestimation algorithm based on RSSI is proposed on the basis of summarizing andcomparing four kinds of traditional positioning technology (RSSI, AOA, TOA andTDOA). Two rotating directional antennas, which two pattern overlap, receive RSSIvalues respectively, and then AOA of the object node can be estimated by using thetwo pattern difference method. The algorithm proposed in this dissertation is of lesscomplexity, requires less hardware cost and is suitable for node localization in WSNs.The experimental results demonstrate that the mean errors of AOA estimation for theindoor and outdoor are6.7and0.6degrees, respectively.The experimental results demonstrate that multipath interference is the mainfactor causing positioning error. In order to eliminate the multipath interference, thisdissertation first introduces the methods of multipath time delay estimation based oncorrelation analysis, and analyses the two super resolution time delay estimationalgorithm——the EM algorithm and WRELAX algorithm. These algorithmsdecouples the multidimensional optimization problems into a series ofone-dimensional iterative optimization problems. While the time delay estimationprecision is higher, it needs large amount of calculation because of iterative operation.Then this dissertation introduces a method of multipath time delay estimation basedon FRFT and its computational complexity is much smaller than WRELAX algorithm. Finally we present a highly efficient algorithm with subchip resolution capability,which is referred to as FRFT-WRELAX. The basic step of this algorithm is to firstfind reliable initial estimates of the time delay and amplitude by using FRFT and thenrefine them by using the last step of the WRELAX algorithm. The validity of thisproposed algorithm is demonstrated via simulations. FRFT-WRELAX outperformsboth FRFT and WRELAX in time delay estimation precision and provides higheraccuracy than FRFT in amplitude estimation. Furthermore, the proposed method canbe implemented at less computational cost than WRELAX.