Federating Digital Contact Tracing using Structured Overlay Networks

Abstract

One of the biggest challenges of today is to slow down the spreading of SARS- CoV-2 virus producing Covid-19 pandemic; Prevention, Testing and Tracing are the main pillars of the solution. Contact Tracing of an infected person is essential to control the spread of the disease. Through this process Health Authorities identify, notify, and monitor people who came in close contact with an individual who was tested positive for an infectious disease, like Covid-19, while he/she was infectious. Also, contact tracing data helps medical experts to find the origin of the virus, learn more about the nature of the virus and estimate the speed of how fast the virus is spreading. Contact tracing has mostly been done manually since many centuries ago. Identifying contacts is done through an interview with the person infected with the virus, after each contact is called by phone. Due to the highly contagious nature of the SARS-CoV-2 virus and the fact that symptoms can manifest after many days (or even never, e.g. asymptomatic cases), manual contact tracing does not give satisfactory results. In the situations when the virus is spreading to fast, Health departments and authorities do not have enough employees to do manual contact tracing. For these reasons digital contact tracing has been considered already at the beginning of the Covid-19 pandemic. There is a plethora of digital contract tracing application. They are developed on very different paradigms, centralized vs. decentralized, GPS based vs. Bluetooth [1, 7]. The rush to make these applications work in the shortest time led to their great diversity. The most important open problem is their interoperability. There are many ongoing efforts to make a federation of these different systems. Herein, we address this problem and propose a solution based on mathematical models of overlay networks. We present a design of the system for connecting different digital contact tracing applications, called BubbleAntiCovid19 (BAC19). The model is inspired by Chord [8] and Synapse [3] Structured Overlay Networks. The correctness and efficiency of lookup procedures of this protocols was in the focus of several papers, e.g. [4, 5, 6, 8]. We prove that BAC19 provides a complete and fully exhaustive retrieving procedure. Hence, BAC19 is proven to be a simple yet powerful interconnection of already existing digital contact tracing applications that - by construction - do not communicate with each others as such providing their efficient interoperability. More details on this work can be found in [2].