A Low-Cost Collaborative Indoor Localization System based on Smartphone Platform

While GPS has been well served in outdoor localization for a long time, the indoor localization is still an open challenge in both academia and industry. To solve the indoor localization problem, we propose a novel and low-cost system, named Collaborative Indoor Localization System (CILS), using sound as the medium to localize smartphones. The proposed CILS is composed of three parts, smartphones, sound sources and a server. The sound sources are fixed at known positions and repeatedly generate sound, which can be any existing sound generators, such as a ringing bell or electrical elevator. The smartphones distinguish the sound periods from noise in the recorded signals from microphones using statistical features and upload the sound periods to the server with timestamps. The server collects all smartphones' recorded sound and detects every sound period's source by a Hidden Markov Model (HMM) and a Gaussian Mixture Model (GMM). The server calculates every pair of smartphones' positions together using the time difference of signals' arrivals from the same sound source. The system is implemented in the Android system and tested in different indoor scenarios. It is shown that the CILS can achieve a high accuracy of localization via extensive experiments.