Crowdsourced radiomap for room-level place recognition in urban environment

摘要: The proliferation of WLAN infrastructures has facilitated numerous indoor localization techniques using fingerprints. In particular, identifying a room or place in urban environments could be usefully utilized many application domains such as ubiquitous health. However, it is not straightforward how to bootstrap system because fingerprints all places must available advance. this paper, we propose crowdsourcing approach for recognition. key idea build an open participatory through which users can contribute As the database size increases, provide recognition service. We conducted extensive experimental study at university campus demonstrate performance proposed method terms accuracy. also studied factors that undermine fingerprint density, incorrect contribution, uneven and device heterogeneity.