Research On Fusion Localization Algorithm Based On AOA And RSSI

With improvement of living standard and the rapid development of science and technology,the demand for Location Based Service(LBS)is also increasing.Nowadays,the applications based on outdoor location have been well developed,which utilize a series of techniques that are cored on Global Position System(GPS).In the indoor environment,the quality of GPS signal drops significantly due to the blockage of walls and other obstruction,which reduces the localization accuracy and makes it unable to meet the requirement for developing relative applications.Therefore,researchers from different countries are trying to solve the problem of indoor localization through different ways.Due to the advantages of extensive deployment and low equipment cost,Wi-Fi becomes one of the major supporting platforms.Based on the background discussed above and consider the insufficient accuracy of Received Signal Strength Indication(RSSI)localization system based on Wi-Fi platform,this thesis has carried out a research on the fusion localization algorithm of Angle of Arrival(AOA)and RSSI.Its main contents are summarized as follow.Firstly,this thesis studies the transmission model of Orthogonal Frequency Division Multiplexing(OFDM)system and the Channel State Information(CSI)deeply.Then,a joint estimation algorithm of AOA and Time of Arrival(TOA)is constructed based on CSI obtained from multiple carriers of multiple antennas.This algorithm conquers the limitation of the number of antennas to the estimation accuracy in the original system,which enables the system to estimate the AOA accurately with a small number of antennas.At the same time,the two-dimensional smoothing algorithm is studies,which solves the problem of the AOA estimation of coherent signals.The errors caused by sampling frequency offset and carrier frequency offset in the CSI phase are analyzed in detail,and phase error correction algorithm is studied to ensure the accuracy of AOA which is utilized for fusion localization.Secondly,in order to obtain the AOA information of the direct signal from the result of AOA estimation,a joint determination method is studied.It employs the affine propagation clustering algorithm and the outlier detection to analyze the characteristics of the signal propagation path in the angle and time domain.This method can select out the receiver working in the LOS situation and filter out the AOA of direct path between the target and the receiver for localization.Finally,combining the AOA and RSSI information with the Kalman filter,the location of the target is calculated.Furthermore,this thesis builds an experimental platform based on commercial Wi-Fi equipment with three antennas and carries out some experiments in indoor and outdoor environment.The extensive experimental results demonstrate our system can achieve a median accuracy of 0.65 meters and 0.98 meters in an outdoor and indoor environment,respectively.