A Robust Indoor Autonomous Positioning System Using Particle Filter Based on ISM Band Wireless Communications

In the field of the ubiquitous computing, positioning systems which can provide users' location information have paid attention as an important technical element which can be applied to various services, for example, indoor navigation services, evacuation services, market research services, guidance services, and so on. A lot of researchers have proposed various outdoor and indoor positioning systems. In this paper, we deal with indoor positioning systems. Many conventional indoor positioning systems use expensive infrastructures, because the propagated times of radio waves are used to measure users' positions with high accuracy. In this paper, we propose an indoor autonomous positioning system using radio signal strengths (RSSs) based on ISM band communications. In order to estimate users' positions, the proposed system utilizes a particle filter that is one of the Monte Carlo methods. Because the RSS information is used in the proposed system, the equipments configuring the system are not expensive compared with the conventional indoor positioning systems and it can be installed easily. Moreover, because the particle filter is used to estimate user's position, even if the RSS fluctuates due to, for example, multi-paths, the system can carry out position estimation robustly. We install the proposed system in one floor of a building and carry out some experiments in order to verify the validity of the proposed system. As a result, we confirmed that the average of the estimation errors of the proposed system was about 1.8 m, where the result is enough accuracy for achieving the services mentioned above.