I was thinking along similar lines. Maybe it's possible to use BLE for this (don't know if the standard PXP profiles would be applicable or you'd need something

I was thinking along similar lines. Maybe it's possible to use BLE for this (don't know if the standard PXP profiles would be applicable or you'd need something else). You could also have app-level 24-hour rolling identifiers to prevent non-infected people's contacts from being correlated over time, even locally—if that would make people more comfortable using such an app. (The app would have to keep track of all its previous identifiers; if the user is found to be infected, all their previous identifiers would be marked as such in the database.) Some challenges off the top of my head: - What polling rate is needed? (How do disease experts define a "contact"?) What's the battery impact? - What fraction of the population has BLE-capable phones? - What fraction of the population keeps their phones physically on their person as they go about their day? - Can distance be roughly inferred from RSSI? Does the mapping of RSSI to distance vary much depending on the transmitting radio / phone model? - If you live in an apartment, you might be identified as a "contact" of your neighbor even if you never breathe the same air, etc.