Wireless positioning based on hierarchical symbolic dynamic filtering of RSSI time series

Abstract

Hierarchical symbolic dynamic filtering (HSDF) is used in literature for anomaly detection by unsupervised classification of time series data. This paper proposes a novel wireless positioning method based on the HSDF concept. This approach aims to tackle the main problems in fingerprinting-based localization methods by eliminating the offline phase, which is the most time-consuming part of a fingerprinting localization. Online learning of Received Signal Strength Indicator (RSSI) time series data causes this method to be efficient in dealing with indoor environments’ dynamical behavior, which is another problem in the localization methods domain. Due to SDF (symbolic dynamic filtering)-based structure, it shows robustness to noise and multipath effect. This paper demonstrates these achievements by results obtained from both simulation of noisy datasets and real-time positioning in an indoor testbed.