Interesting, I learnt something new today. My only comment was the noise floor of the simple graph was very high-25dBm which (without having a clue how the physics works and skimming the article) sounds about right for something with no proper RF front-end.
Where do you see -25 dBm noise floor? But chance do you mean a 25 dB SNR?
The picture here <https://scx1.b-cdn.net/csz/news/800a/2026/rydberg-atoms-dete...> though I just saw the X-axis refers to audio frequencies, kind of confusing on first glance.
That's dB not dBm, and the article says it's normalized to whatever power each of the two options received at. Nothing about absolute power.