WiFi Fingerprint Clustering for Urban Mobility Analysis

In this paper, we present an unsupervised learning approach to identify the user points of interest (POI) by exploiting WiFi measurements from smartphone application data. Due lack GPS positioning accuracy in indoor, sheltered, and high rise building environments, rely on widely available access (AP) contemporary urban areas accurately POI mobility patterns, comparing similarity measurements. We propose a system architecture scan surrounding AP, perform demonstrate that it is possible three major insights, namely indoor within building, neighborhood activity, micro users. Our results show aforementioned with fusion GPS, which are not only using GPS.