Indoor Localization System Design And USRP Implementation

With the development of communication technology,the application scenarios and requirements of indoor positioning technology are becoming more diverse,so the importance of indoor positioning technology research is becoming more and more significant.For indoor positioning algorithms,there are commonly used distance estimation algorithms based on Time of Arrival(TOA)and the Classic Multiple Signal Classification Algorithm(MUSIC).The principle of the former algorithm is simple,but the absolute arrival time of the signal from the transmitting end to the receiving end is difficult to accurately measure and a very small measurement time error will cause a large distance error.The latter uses the phase difference generated by the wave path difference of different antennas in the array to estimate the angle of arrival of the signal to be located to the access point(AP),and estimates the position of the target to be measured by the arrival angle of multiple APs..The positioning accuracy of the algorithm is high,but the premise of the algorithm is that the number of antennas in each antenna array is larger than the number of multipaths in the indoor environment.In an indoor environment,there are usually 6 to 10 multipaths,so each antenna The array requires at least 6 antennas,resulting in higher hardware costs per AP in real-world applications.Firstly,in view of the above problems,the MUSIC algorithm based on the combination of Angle of Arrival(AOA)and Time of Flight(TOA)utilizes the phase difference and signal generated by the arrival time of each subcarrier in the multicarrier orthogonal transmission system.A phase difference is generated between different antennas to estimate the angle of arrival,the number of subcarriers is used to compensate for the shortage of the actual number of antennas,and the angle of arrival from the target to the access point is estimated with fewer antennas.Compared with the TOA-based distance estimation principle,the algorithm calculates the relative arrival time of the target to AP through each path,and uses the relative arrival time and angle of arrival to jointly locate the target to be located.The error will be smaller than the TOA-based distance.Estimation principle;less than the classical MUSIC algorithm requires fewer antennas,and does not need to consider the problem that the number of multipaths is larger than the theoretical number of antennas,reducing the number of antennas actually deployed.Secondly,in the actual implementation,there are still many problems in the MUSIC algorithm combined with AOA and TOF,such as the time when the received signal and the transmitted signal are not synchronized,the sampling offset caused by the transmitting end and the receiving end DAC,and the ADC.In view of the above problems,this thesis proposes a frame synchronization processing and phase cancellation algorithm for the received signal,which solves the problem that the received signal and the transmitted signal are not synchronized,and eliminates the sampling bias caused by the DAC and the ADC at the transmitting end and the receiving end.Secondly,using the CSI spatial smoothing algorithm,the coherence of the received signals of different antennas is reduced,and the theoretical number of antennas is expanded.Finally,a grid search algorithm is proposed to accurately measure the targets to be measured.It is estimated that the positioning error is reduced.Finally,this thesis implements the designed indoor positioning system on the Universal Software Radio Peripheral(USRP).Each module was designed based on LabVIEW,and the system was tested and analyzed in an actual indoor environment.The experimental results show that the algorithm of this system can estimate the arrival angle of each AP more accurately;the angle estimation and position estimation performance is better than the classical MUSIC algorithm.The correctness of the elimination phase offset algorithm on the USRP platform is verified,and the influence of different AP numbers on the positioning error is compared.In summary,based on the MUSIC algorithm combined with AOA and TOF,the indoor positioning system is designed and implemented on the USRP platform,which solves the problem that the receiving and transmitting time is not synchronized;the sampling offset is eliminated.Phase offset;expands the theoretical number of antennas;reduces signal coherence in the middle of receiving different antennas;reduces the number of antennas actually deployed,and can achieve higher positioning accuracy.