A fairly classical example of readily de-anonimisable data is location, at postal-code resolution.
With nothing more than two postal codes (ZIP codes in the US), representing home and work, it's possible to individually identify about 90% of the population. That information is available via, say, geocoded location using mobile phone or tracking cookie data.[1]
In the case of device or vehicle tracking such as with drones, it's highly likely that a given device would be used within a particular jurisdiction or activity region. If you can pinpoint specific locations and times within those (e.g., police activity around an address + drone activity, activity following specific infrastructure such as power lines or gas distribution, training locations, etc., etc.) you can probably come up with a strong idea of who operates or owns the equipment.
1. The home+work ZIP anecdote is the one I recall. I'm not finding that specifically though this paper mentions four spatio-temporal locations sufficing for 95% of the population: Yves-Alexandre de Montjoye, César A. Hidalgo, Michel Verleysen & Vincent D. Blondel, "Unique in the Crowd: The privacy bounds of human mobility", Scientific Reports volume 3, Article number: 1376 (2013) https://www.nature.com/articles/srep01376
