A Visible Light Based Indoor Positioning System

Abstract

In this paper, we propose a novel indoor localization scheme that exploits ubiquitous visible lights, which are necessarily and densely deployed in almost all indoor environments. We unveil two phenomena of lights available for positioning: 1) the light strength varies according to different light sources, which can be easily detected by light sensors embedded in COTS devices (e.g., smart-phone, smart-glass and smart-watch); 2) the light strength is stable in different times of the day thus exploiting it can avoid frequent site-survey and database maintenance. Hence, a user could locate oneself by differentiating the light source of received light strength (RLS). However, different from existing positioning systems that exploit special LEDs, ubiquitous visible lights lack fingerprints that can uniquely identify the light source, which results in an ambiguity problem that an RLS may correspond to multiple positions. Moreover, RLS is not only determined by device's position, but also seriously affected by its orientation, which causes great complexity in site-survey. To address these challenges, we first propose and validate a realistic light strength model that can attributes RLS to arbitrary positions with heterogenous orientations. This model is further perfected by taking account of the device diversity, influence of multiple light sources and shading of obstacles. Then we design a localizing scheme that harness user's mobility to generate spatial-related RLS to tackle the position-ambiguity problem of a single RLS, which is robust against sunlight interference, shading effect of human-body and unpredictable behaviours (e.g., put the device in pocket) of user. Experiment results show that our scheme achieves mean accuracy $1.93$m and $1.98$m in office ($720m^2$) and library scenario ($960m^2$) respectively.