WLAN Environment for Indoor Localization

This paper investigates the deployment of WLAN for indoor localization. K-Nearest Neighbor algorithm is adapted to predict the location of a user in an indoor environment. The accuracy of K-Nearest Neighbor in predicting user's location in an indoor environment is evaluated. As resistance in indoor environment such as walls and movement of objects adversely affect the performance of the algorithm, emphasis is placed on RSS sample vector fluctuation correction. Two simulations were carried out, one adapting the fluctuation correction algorithm and one without fluctuation correction algorithm. The results of the investigation shows that deployment of fluctuation correction algorithm improves the prediction accuracy. The number of access points (APs) deployed in the investigated area also contribute to the prediction accuracy.