Social Mix-zones: Anonymizing Personal Information on Contact Tracing Data

Abstract

In many different contexts, the encounter between two or more individuals opens a window in which information can be exchanged. Considering Mobile Ad hoc Networks (MANETs) scenarios, encounters - also called contacts - are used to transfer data between nodes (the users). In more recent cases, tracing contacts between individuals has shown to be a strong strategy in mapping the transmission of contagious diseases, such as COVID-19. However, sharing contact data can impose threats to the safety of participants regarding their social and mobility behavior. As an example, we can infer acquaintances, as well as home and work locations. This work presents a strategy to anonymize contact tracing data by utilizing mix-zones, a well-defined concept to anonymize data in a given region. Called social mix-zones, it considers the number of contacts happening in a location, producing anonymized data and protecting the personal integrity of the individuals. We validate the proposal using two real contact tracing data, showing that social mix-zones can cover a large portion of contacts, reducing the risk of malicious location attacks.