Correlated received signal strength correction for radio-map based indoor Wi-Fi localization

Abstract

The purpose of received signal strength (RSS) correction in radio-map based Wi-Fi localization is to obtain a set of fine-grain location-dependent RSS fingerprints, and eventually achieve the purpose of highly accurate and reliable localization. To meet this goal, the RSS correction is conducted on the raw RSS samples to eliminate the environmental noise from the radio-map. This paper shows the comprehensive analysis on the autocorrelation property of the chronological RSS samples in the same RSS sequence, and meanwhile presents the correlated RSS correction approach. Furthermore, the correlated RSS correction approach can also be integrated into the conventional radio-map based K nearest neighbor (KNN) and weighted KNN (WKNN) localization algorithms. The experimental results conducted on the real Wi-Fi RSS samples recorded in a representative indoor environment prove that the proposed correlated RSS correction approach can result in the significant improvement of accuracy over the conventional radio-map based localization.

Error reduction in distance estimation of RSS propagation models using Kalman filters

“…We argue that when RSS correction methods are integrated with propagation models the performance of such combined models is expected to get significantly enhanced. A commonly used RSS correction method is the feature extraction based method that considers statistical features of the RSS measurements such as the mean, mode, standard deviation, etc [7]. However, such correction methods give equal importance to all samples without differentiating between noisy and true RSS values.…”