Towards Scalable Indoor Localization with Particle Filter and Wi-Fi Fingerprint

Abstract

This work aims to design and implement a scalable and easy-deployed indoor localization system based on particle filter and Wi-Fi fingerprint techniques. Specifically, our system leverages particle filter to estimate user's location and automatically scans Wi-Fi fingerprints. Then, we utilize the collected fingerprints to speed up the convergence of particles. Finally, the system iteratively refines the collected fingerprints by evaluating their performance duration the on-line localization phase, which is able to further enhance the positioning accuracy. We implement the system on Android platform and give a comprehensive performance evaluation by setting up the system in our lab area and comparing the algorithm with conventional fingerprint-based solutions. Experimental results demonstrate the scalability and effectiveness of the proposed solution.