Selecting an access point for indoor localization system using frequency analysis

Abstract

Even though Indoor localization system has been continuously developed, there have also been some amounts of errors in the positioning due to the variance of the signal. This paper proposes a method to help selecting an access point to reduce the variance of received signal strength indicator (RSSI) using frequency analysis and applied genetic algorithm to search the optimal weights for weighted distant fingerprint algorithm (WDF). Experiments were conducted in the indoor environment using android mobile received signal strength from the access point and the proposed algorithm compared with K-Nearest Neighbor (KNN) algorithm and conventional weighted distant fingerprint (WDF) algorithm. Results have demonstrated the proposed algorithm that can improve an increased accuracy up to 89.75% for more accurate identification.