PERFECT : Per sonalized F ederated L e arning for C BRS Radar De t ection
Shafi Ullah Khan 1 , Madan Baduwal 2 , Vini Chaudhary 2 , and Debashri Roy 1 · 2026
Read PERFECT : Per sonalized F ederated L e arning for C BRS Radar De t ection by Shafi Ullah Khan 1 , Madan Baduwal 2 , Vini Chaudhary 2 , and Debashri Roy 1 (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-03199.txt.
Description
Shafi Ullah Khan, Madan Baduwal, Vini Chaudhary, and Debashri Roy propose PERFECT, a personalized federated system for detecting naval radar in the shared CBRS band. Sensors must spot faint radar amid heavy phone network traffic, while central training raises privacy worries and sensor data differ by location.
PERFECT trains locally on each sensor and personalizes models where sites see distinct radar types, so raw data never leaves the sensor. In simulations it reached the required ninety-nine percent radar recall, matching central training while improving privacy, efficiency, and scale. The design fits the varied, spread-out sensing network that guards the band.
Catalog record
| CREATOR | Shafi Ullah Khan 1 , Madan Baduwal 2 , Vini Chaudhary 2 , and Debashri Roy 1 |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 44 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.03199