Modelling and Experimental Assessment of Inter-Personal Distancing Based on Shared GNSS Observables.

In the last few years, all countries worldwide have fought the spread of SARS-CoV-2 (COVID-19) by exploiting Information and Communication Technologies (ICT) to perform contact tracing. In parallel, the pandemic has highlighted the relevance of mobility and social distancing among citizens. The monitoring of such aspects appeared prominent for reactive decision-making and the effective tracking of the infection chain. In parallel to the proximity sensing among people, indeed, the concept of social distancing has captured the attention to signal processing algorithms enabling short-to-medium range distance estimation to provide behavioral models in the emergency. By exploiting the availability of smart devices, the synergy between mobile network connectivity and Global Navigation Satellite Systems (GNSS), cooperative ranging approaches allow computing inter-personal distance measurements in outdoor environments through the exchange of light-weight navigation data among interconnected users. In this paper, a model for Inter-Agent Ranging (IAR) is provided and experimentally assessed to offer a naive collaborative distancing technique that leverages these features. Although the technique provides distance information, it does not imply the disclosure of the user's locations being intrinsically prone to protect sensitive user data. A statistical error model is presented and validated through synthetic simulations and real, on-field experiments to support implementation in GNSS-equipped mobile devices. Accuracy and precision of IAR measurements are compared to other consolidated GNSS-based techniques showing comparable performance at lower complexity and computational effort.