Access Point Reselection and Adaptive Cluster Splitting-Based Indoor Localization in Wireless Local Area Networks

Indoor localization technology has received increasing attention because the hobbies and interests of human can be mined from the location data. Wireless local area network (WLAN)-based fingerprinting localization methods have become attractive owing to their advantages of open access and low cost. However, for localization in realistic large areas, three problems persist: (1) excessive memory requirements in the offline phase; (2) high computational complexity in the online phase; and (3) how to select the access point (AP) sets with best distinction capability. A novel method of localization based on adaptive cluster splitting (ACS) and AP reselection is proposed in this paper. The suggested method can significantly reduce the requirements of offline storage and online computing capacity while improving the localization accuracy. The expected result is demonstrated in a theoretical deduction, simulation, and with experiments in a realistic environment.