eprint.iacr.orgpaperBlind-DLPA: An End-to-End Framework for Profiled Blind Side-Channel Analysis
Blind-DLPA is a multi-task deep-learning framework for profiled blind side-channel analysis, where the attacker has labeled profiling traces but only leakage traces from the target, with no plaintexts or ciphertexts. It estimates Hamming-weight posteriors of an input and a key-dependent intermediate, then ranks key hypotheses against theoretical joint distributions. It recovers keys on unprotected AES and Kyber and on ASCAD (guessing entropy zero with 39 traces), and the authors report collisions in a prior Ascon Hamming-weight model.
TL;DR Presents a deep-learning framework that recovers keys from target power traces alone, without plaintexts or ciphertexts, on AES, Kyber and masked ASCAD.
Organisations named here
- ASCAD
- Blind-DLPA
- Boolean masking
- desynchronization
- guessing entropy
- Hamming weight
- profiled blind side-channel analysis
Technologies named here
- Unprotected AES implementations
- Kyber
- ASCAD dataset
- Ascon
- AES
Related in Hardware & RF
eprint.iacr.orgpaperHAETAE on a Diet: Less Stack, Less Time on Low-End Microcontrollers
github.comrepogeo-tp/ESP32-Bit-Pirate
github.comrepo