Research Of Indoor Localization Technology Based On AP-Selection In Grid-Covered Field

Recently, the research and the implementations of indoor positioning based on wireless technologies have drawn lots of attention from the academic communities and industry. For the widespread large indoor scenarios, the localization technology based on the widely deployed Wi-Fi APs has been studied to estimate the location without any additional equipment.Firstly, we study how to use the existing APs to achieve indoor localization. Then we optimize the deployment of APs. In order to suit the indoor complex environment and meet the requirements of high precision, the triangle and centroid localization algorithm is explored and modified. Finally, we combine two proposed algorithms to further improve the positioning accuracy.The main contributions of this thesis are as follows:(1) An indoor localization algorithm based on AP selection was explored. A simple and easy-to-implement localization algorithm based on AP selection, NaviAP, is proposed. The algorithm was also implement on Android system. In addition, the impact of AP deployment to positioning accuracy has been analyzed. Finally the positioning accuracy will be improved in the hexagonal-grid tessellated field.(2) We studied the triangle and centroid localization algorithm. In order to achieve the requirements of high positioning accuracy in some indoor scenarios, we explored the application of traditional triangle and centroid localization algorithms and proposed the CRoC and CTC localization algorithms to address the problem of unable to localization scenarios. The simulation results show that the two proposed algorithms have higher positioning accuracy.(3)We designed the CRoC and CTC localization algorithm co-locating with NaviAP respectively:To improve inaccuracy with NaviAP, we analyzed the computing process and the positioning effect of the proposed algorithms. Besides, considering the coverage characteristics of the APs and the integrated method of different algorithms, we proposed the co-locating method to improve the positioning accuracy without increasing the computation complexity. Finally, on the simulation platform, the theoretical results was verified.Through the extensive simulations, the proposed method of algorithm combined for localization was demonstrated to further improve the positioning accuracy without increasing amount of calculation.