FedMamba-IoMT: Federated state space models with differential privacy and byzantine resilience for privacy-preserving intrusion detection in Internet of Medical Things
Yasser Mohammad Al-Sharo, Mohammed Tawfik, Ali Mansour Almadani, Amr H Abdelhaliem, Islam S Fathi, Gaber Hassan · 2026
Read FedMamba-IoMT: Federated state space models with differential privacy and byzantine resilience for privacy-preserving intrusion detection in Internet of Medical Things by Yasser Mohammad Al-Sharo, Mohammed Tawfik, Ali Mansour Almadani, Amr H Abdelhaliem, Islam S Fathi, Gaber Hassan (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/plos-pone-0355601.txt.
Description
Yasser Mohammad Al-Sharo and co-authors present a federated intrusion detector for networks of medical devices that learns without gathering private data. Hospitals cannot pool patient-linked traffic, yet shared learning catches more attack types. Heavy transformer models detect well but cost too much for small devices. Their 2026 design swaps them for state-space sequence blocks with linear cost.
Traffic features are reshaped into short token sequences for the stacked blocks. Formal privacy noise is added during training, and the merging step resists sabotage by rogue participants. Explanation tools operate at several levels so analysts can audit alerts. Al-Sharo and colleagues aim to square accuracy, privacy guarantees, and robustness in one framework.
Catalog record
| CREATOR | Yasser Mohammad Al-Sharo, Mohammed Tawfik, Ali Mansour Almadani, Amr H Abdelhaliem, Islam S Fathi, Gaber Hassan |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 82763 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |