Indoor Localization Algorithm Based on Iterative Grid Clustering and AP Scoring

Indoor localization is of great importance in daily and commercial applications. This paper proposes a novel indoor localization algorithm based on iterative K-means and grid scoring (KS) and a mechanism of access point (AP) scoring. The basic approach of the proposed algorithm is composed of a two-step iteration. The first step is to randomly select a group of APs. Then, the mobile terminal is located into one cluster based on the received signal strength of the selected APs and the score of all the grids belonging to this cluster. After several iterations, the location estimation is selected as the grid with the highest score. To further improve the localization accuracy, AP scoring (AS) is adopted to select the APs with superior localization capability. The suggested algorithm can locate a position effectively with relatively high accuracy. The expected results are demonstrated using simulations.