Julian Nowakowski Profile
Julian Nowakowski

@nowakow_ski

Followers
43
Following
21
Media
2
Statuses
12

PhD student @ruhrunibochum | Cryptanalysis

Bochum
Joined February 2018
Don't wanna be here? Send us removal request.
@CASA_EXC
CASA - Cluster of Excellence for Cyber Security
2 years
Day two of #Eurocrypt24 in Zurich, and yes, the rumors are true – #CASAfants are out! 😱🐘 But don't worry, those people from our team in the picture are still around. And they are all happy to provide you with information about #Cybersecurity Research in Bochum.
0
3
10
@4ndre3sser
Andre Esser
2 years
There is a new workshop in town! ***Advances in Asymmetric Cryptanalysis*** (AAC '24, @AAC_Workshop) is an ACNS24 (@acns_2003 ) affiliated workshop. It covers the whole spectrum of cryptanalysis! More infos at https://t.co/h1ymy5PrJj. Submission deadline: 15th of November!
Tweet card summary image
aac-24.tii.ae
AAC 2024 fills the current gap in the cryptographic community by providing a dedicated platform for cryptanalysts.
0
10
23
@nowakow_ski
Julian Nowakowski
2 years
On the technical side, we exploit the high redundancy of McEliece secret keys to construct an error-correcting code with very high decoding radius. Our attack then simply uses this code to perform error-correction on our leak data. 3/3
0
0
1
@nowakow_ski
Julian Nowakowski
2 years
We assume access to a side channel, that leaks the elementary row operations of the Gaussian elimination step inside McEliece keygen. Even if our side channel is extremely noisy and flips each bit with probability ~40%, we can recover the secret key in a matter of minutes. 2/3
1
0
1
@nowakow_ski
Julian Nowakowski
2 years
Our new paper "Leaky McEliece: Secret Key Recovery From Highly Erroneous Side-Channel Information" is now on eprint ( https://t.co/vgsnkIFIXo). Joint work with @lambdafu, Chitchanok Chuengsatiansup, @AlexanderMay10 and @yuvalyarom. 1/3
1
3
12
@nowakow_ski
Julian Nowakowski
2 years
The main idea behind our algorithm is a new *automated* variant of Coppersmith's method. Previous Coppersmith-type results typically required highly technical, ad-hoc lattice constructions, whereas our novel approach constructs these lattices fully automatically. 6/6
0
0
1
@nowakow_ski
Julian Nowakowski
2 years
We show that given a (static) CDH-like oracle, that outputs 54% (CSIDH) / 76% (CSURF) of the MSBs of the shared curve, we can break CSIDH / CSURF in polynomial time. 5/6
1
1
3
@nowakow_ski
Julian Nowakowski
2 years
In (2), we define and analyze the Commutative Isogeny Hidden Number Problem, which is the natural analogue of the Hidden Number Problem in the CSIDH and CSURF setting. 4/6
1
0
0
@nowakow_ski
Julian Nowakowski
2 years
We simplify the involved lattice constructions of the original paper, and provide a significantly faster implementation - finally making the original framework practical for real world attacks. 3/6
1
0
0
@nowakow_ski
Julian Nowakowski
2 years
In (1), we revisit the amazing Crypto 20 paper "LWE with Side Information", which introduced a framework for integrating side channel information into the primal lattice attack. 2/6
1
0
1
@nowakow_ski
Julian Nowakowski
2 years
Very happy to announce that our two papers (1) "Too Many Hints — When LLL Breaks LWE" (with @AlexanderMay10), (2) "Solving the Hidden Number Problem for CSIDH and CSURF via Automated Coppersmith" (with Jonas Meers), got accepted at Asiacrypt. 1/6
2
4
30