Model for a secure contact tracing app
Simulation of effectiveness against corona
In the fight against the Covid-19 virus, the interdisciplinary research team ContacTUM has developed a model for a secure contact tracing app. Simulations are now testing whether the app really can help to reduce the number of infections with the SARS-CoV-2 virus.
Safe and globally compatible digital contact tracing using an app may offer a chance to slow down the spread of the virus. The interdisciplinary team from the fields of physics, computer science, law, mathematics and medicine lead by the physicist Prof. Elisa Resconi, uses a concept with a coded calculation method.
Warning contact persons
The basic principle of tracing contacts using an app, is to be able to quickly inform the contact persons, when someone has become infected. To this purpose, mobile phones which have the app installed, exchange randomly generated and permanently changing TCNs (temporary contact numbers) using Bluetooth technology.
The contact tracing app collects these TCNs on the individual devices and saves them for a time period of around two weeks. Should a person become infected with the virus, then the contacts can be anonymously notified using the data saved in the app.
Security through coding
ContacTUM has been working on an app using the decentralized approach to the collection of contact data. In this way, an infected person releases only the TCNs transmitted by their own device to a server. These TCNs are then transmitted from the server to all devices on which the app is installed. The cross-check to determine whether any of the "infected" TCNs were received therefore takes place only locally on the individual mobile devices. Consequently, the only party with knowledge of a possible contact with an infected individual is the contacted person themselves – and not a central server.
In the app model from ContacTUM, the cross-checking of TCNs of infected individuals against those collected on other mobile phones also takes place without having to download the TCNs of the infected individual onto the devices of the contact persons. This is possible due to an encryption process known as "private set intersection cardinality", which does not require information to be exchanged in plain text.
This concept has the advantage, that contact persons can be warned about a possible infection risk, without their mobile devices being able to recognize the "infected" TCNs among the TCNs stored on their device. As a result, it is much more unlikely, that the anonymity of an infected person is endangered.
Simulations on the effectiveness of the app
Effective reproduction number $R_{eff}$ in dependency of tracing probability
$p_{trace}$ ($\approx$ app abundance) and social distancing $sd$
Parallel to these developments, part of the ContacTUM-Team, led by the physicist Stefan Schönert and mathematics Professor Johannes Müller, has created simulations, which aim to show under which conditions the app can truly help to reduce the number of Covid-19 infections.
From their initial simulations, the scientists believe that, for a reduction in infections to be achieved, at least 60 percent of the population would have to install and use the contact tracing app. Their results also showed that the contacts of an infected person's contacts would also have to be notified without delay in order to break the infection chain.
TUM and ITO develop app prototype
In order to develop a user-friendly prototype of the app, ContacTUM is in close contact with ITO – an Open-Source community of around 30 international software developers, whose activity is openly accessible and transparently communicated.
A prototype of the app has already been tested on Android operating systems, the code is publicly available. However, it is expected to be some weeks, until a totally secure and technically flawless app is ready for use.
Further, detailled information can be found in the press release of the TUM: Encryption system for a secure contact tracing app.