MambaGCN-SMI: Protocol state machine inference framework based on Mamba and GCN
Junkang Ren, Qing Li, Ruijuan Chu, Yifan Chen · 2026
Read MambaGCN-SMI: Protocol state machine inference framework based on Mamba and GCN by Junkang Ren, Qing Li, Ruijuan Chu, Yifan Chen (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-0354921.txt.
Description
Junkang Ren and three colleagues present a deep learning system for reverse-engineering how computer protocols behave, in this 2026 paper. Security analysts often face private or undocumented protocols and must infer their internal state machines, the rules for which message can follow which, from observed traffic alone. Their framework pairs a sequence model called Mamba with a graph learning method.
The system first extracts deep features from message exchanges, builds an initial map of protocol states, then merges redundant states in a data-driven way rather than by fixed rules. Tested on five protocols, from web and email traffic to industrial control protocols, it recovered state machine structures accurately and predicted behavior better than mainstream alternatives. Removing either main component hurt performance, confirming both contribute. The authors offer it as a sharper tool for protocol analysis and security testing.
Catalog record
| CREATOR | Junkang Ren, Qing Li, Ruijuan Chu, Yifan Chen |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 73008 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |