Lighthouses: A Warning System for Super-Spreader Events

Abstract

Super-spreader events have been a driving force of the COVID-19 pandemic. Such events often take place indoors when many people come together. Various systems for automated contact tracing (ACT) have been proposed which warn users if they have come near an infected person. These generally fail to detect potential super-spreader events as only users who have come in close contact with the infected person, but not others who also visited the same location, are warned. Other ACT approaches allow users to check into locations, but this requires user interaction. We propose two designs how broadcast-based ACT systems can be enhanced by using location-specific information without the need for GPS traces or scanning of QR codes. This makes it possible to alert attendees of a potential super-spreader event while providing privacy. Our idea relies on cooperating "lighthouses" which cover a large area and send out pseudonyms. In our passive design the health authority (HA) publishes location pseudonyms collected by infected users. In the active design, lighthouses communicate with HAs. After retrospectively detecting an infected visitor the lighthouse notifies the HA which users' stay overlapped.