Title :
Security of neural cryptography
Author :
Mislovaty, R. ; Klein, E. ; Kanter, I. ; Kinzel, W.
Author_Institution :
Dept. of Phys., Bar-Ilan Univ., Ramat-Gan, Israel
Abstract :
In this paper we analyze the security of neural cryptography, a novel key-exchange protocol based on synchronization of neural networks. Various attacks on this protocol were suggested by Shamir et al., and the protocol was shown to be secure against them (Mislovaty et al. (2002)). A new attack strategy involving a large number of cooperating attackers, that succeeds in revealing the encryption key was recently found.
Keywords :
cryptography; neural nets; protocols; synchronisation; attack strategy; cooperating attackers; encryption key; key-exchange protocol; neural cryptography; security; synchronization; Arithmetic; Artificial neural networks; Cryptographic protocols; Cryptography; Equations; Hebbian theory; Machine learning; Neural networks; Physics; Random variables;
Conference_Titel :
Electronics, Circuits and Systems, 2004. ICECS 2004. Proceedings of the 2004 11th IEEE International Conference on
Print_ISBN :
0-7803-8715-5
DOI :
10.1109/ICECS.2004.1399654
