Juxtaposing individual and group mobility from sparse Wi-Fi signatures with cloud-assisted computing: a case study for a multidisciplinary university campus

Abstract

ABSTRACT Understanding human mobility undoubtedly enriches common goods. Among available location-aware technologies, Wi-Fi provides a more sustainable and high-resolution means to study human mobility patterns as it has become conspicuous and affordable recently. Whilst existing studies have shed light on facets of personal mobility, the intricate dynamics of group mobility have garnered comparatively scant empirical scrutiny in real-world settings, especially in the juxtaposition with individual mobility. Moreover, they have often overlooked the multifaceted nature of personal attributes influencing daily routines. This study introduces a comprehensive framework that takes advantage of the readily available Wi-Fi connection data and cloud-assisted computing for juxtaposing individual and group mobility. The framework was tested on auniversity campus that provides representative human mobility patterns with diverse attributes. Two tests that aim to demonstrate the framework’s capability to (1) capture individual mobility patterns by using data processing amenable to the spatiotemporal sparseness and to formulate group mobility and (2) differentiate quantitatively the spatiotemporal signatures of distributions, night activities, transitions, and network topologies were conducted. This study uncovers distinct disparities between individuals and groups, and heterogeneities among different attributes in both an empirical and real-world scenario with a reduction in computation time to approximately 6% of the baseline.