Neural cryptography: Difference between revisions

Revision as of 15:59, 31 May 2016

Revision as of 04:17, 4 June 2016

Line 17:

There are currently no practical applications due to the recent development of the field, but it could be used specifically where the keys are continually generated and the system (both pairs and the insecure media) is in a continuously evolving mode.<br />

In 1995, Sebastien Dourlens applied neural networks cryptanalyze [[Data Encryption Standard|DES]] by allowing the networks to learn how to invert the S-tables of the DES. The bias in DES studied through Differential Cryptanalysis by [[Adi Shamir]] is highlighted. The experiment shows about 50% of the key bits can be found, allowing the complete key to be found in a short time. Hardware application with multi micro-controllers have been proposed due to the easy implementation of multilayer neural networks in hardware.<br />

One example of a public-key protocol is given by Khalil Shihab. He describes the decryption scheme and the public key creation that are based on a [[backpropagation]] neural network. The encryption scheme and the private key creation process are based on Boolean algebra. This technique has the advantage of small time and memory complexities. A disadvantage is the property of backpropagation algorithms: because of huge training sets, the learning phase of a neural network is very long. Therefore, the use of this protocol is only theoretical so far.

== Neural key exchange protocol ==

Line 93:

=== Permutation parity machine ===

The permutation parity machine is a binary variant of the tree parity machine.<ref name="Reyes">{{cite web |url=http://iopscience.iop.org/1751-8121/42/19/195002 |title=Permutation Parity Machines for Neural Synchronization |authorauthor1=Oscar Mauricio Reyes, |author2=Ingo Kopitzke, and |author3=Karl-Heinz Zimmermann |last-author-amp=yes }}</ref>

It consists of one input layer, one hidden layer and one output layer. The number of neurons in the output layer depends on the number of hidden units K. Each hidden neuron has N binary input neurons:

Line 115:

Other configurations of the output layer for K>2 are also possible.<ref name="Reyes" />

This machine has proven to be robust enough against some attacks<ref name="Reyes2">{{cite web |url=http://pre.aps.org/abstract/PRE/v81/i6/e066117 |title=Permutation Parity Machines for Neural Cryptography |authorauthor1=Oscar Mauricio Reyes and |author2=Karl-Heinz Zimmermann |lastauthoramp=yes }}</ref> so it could work as a cryptographic mean, but a recent implementation of a probabilistic attack has shown that a key-exchange protocol based on PPM can be violated.<ref name="Seoane">{{cite web |url=http://pre.aps.org/abstract/PRE/v85/i2/e025101 |title=Successful attack on permutation-parity-machine-based neural cryptography |authorauthor1=Luís F. Seoane and |author2=Andreas Ruttor |lastauthoramp=yes }}</ref>

=== Security against quantum computers ===

Line 137:

* [http://www.springerlink.com/content/kbpxkbnkgtk4ymhh/ Analysis of Neural Cryptography] - Analysis of neural cryptography in general and focusing on the weakness and possible attacks of using synchronized neural networks.

* [http://www.opus-bayern.de/uni-wuerzburg/volltexte/2007/2361/ Neural Synchronization and Cryptography] - Andreas Ruttor. PhD thesis, Bayerische Julius-Maximilians-Universität Würzburg, 2006.

* {{cite journal | authorauthor1=Andreas Ruttor, |author2=Wolfgang Kinzel, |author3=Rivka Naeh, and |author4=Ido Kanter |last-author-amp=yes | year=2006 |

title=Genetic attack on neural cryptography | journal=Physical Review E | url=http://link.aps.org/abstract/PRE/v73/e036121| doi = 10.1103/PhysRevE.73.036121 | volume=73}}

* {{cite journal | author=Khalil Shihab | year=2006 |