Simultaneous navigation and pathway mapping with participating sensing

Navigating pedestrians to the expected location in an unfamiliar indoor space, like museum, shopping mall, and airport, is often time-consuming and grueling. Anchor position information or a radio map based navigation strategies are useful in some occasions, but such anchors or maps may not be always available for everyone when needed. In this work, We study the possibility of using hints extracted from off-the-shelf sensors on smartphone and widely deployed WLANs to provide a reliable grass-roots solution for indoor navigation. The proposed simultaneous navigation and pathway mapping approach, EasyPop, overcomes these restrictions, i.e., neither placing assistant devices in advance nor the floor plan are needed. Based on Wi-Fi or cellular signals, site-specific calibration in radio-frequency fingerprinting is received by automatically path mapping. Mobile fingerprints are generated on wireless client by probing the neighbored Wi-Fi APs in predefined period, as well as the sensor hints with respect to spatial---temporal features. Pathways can be generated via incremental construction pattern while quickly reducing the expected distance of a randomly-chosen point from POIs. By leveraging a Bayesian probability model to pathway mapping, the optimal direction of motions is demonstrated step by step by crowdsourcing. EasyPop is evaluated in multiple indoor scenarios involving six volunteers with different data density. The experimental results show that the proposed method has comparable performance to the-state-of-arts without requiring any a priori.