An Algorithm Based on Bayes Inference And K-nearest Neighbor For 3D WLAN Indoor Positioning

Abstract

This paper proposes a hybrid algorithm based on Bayesian inference and K-Nearest Neighbor to estimate the three- dimensional indoor positioning implemented from a fingerprint technique. Additionally, a comparison was made between the main algorithms discussed in literature. The experiments were conducted in a typical building with two floors with 180m2 and four access points. The proposed solution showed a precision in the location of the rooms of 97% and 90% the estimates were at maximum three meters away from the actual location, furthermore, such method has lower variability than other algorithms, with deviation in relation to the mean reaches of 37.62%.