DSSC · Design of secure stream ciphers
FP6 — Marie Curie Actions (Human Resources and Mobility)
- Duration
- 2006-04-01 → 2007-04-30
- EU contribution
- €81,442
- Participants
- 1
- Scheme
- EIF
Lines connect the coordinator with its partners.
Results in brief
Final Activity Report Summary - DSSC (Design of secure stream ciphers)
The areas of information security and cryptography are two of the most active and important research areas worldwide. This project focused on cryptography, and especially on a category of cryptographic algorithms, the stream ciphers. The last two years, stream ciphers have attracted the interest of the research community, in part due to the eStream contest organised by the European Network of Excellence ECRYPT. The project, following the initial proposal, had three main goals. The analysis of the characteristics that a stream cipher must possess, the evaluation of existing attacks against stream ciphers and the proposal of new ones, and finally the impression of the fellow's experience and expertise in a new design proposal. Concerning the first characteristics, the notion of maximum nonlinearity has been study. It consists of an important criterion for measuring the unpredictability and randomness of the stream ciphers output. Both unpredictability and randomness are necessary for the security of a cryptographic algorithm. In terms of the second objective, the algebraic attacks have been mainly studied. A recently proposed algebraic attack has been evaluated. The evaluation revealed inaccuracies. A new attack was proposed that correct mistakes and works in any all cases. Finally, a new stream cipher has been proposed. The new design can be applied in the physical layer, encrypting information that is transmitted through optical fibres, and it bases its security on the laws of physics.
Data: CORDIS, © European Union
Project objective
A cipher is considered secure if it is resistant against all known attacks, and, in the same time, it satisfies a set of statistical tests (linear complexity, universal Mauer test, frequency test, run test, etc).This project deals with both key recovery attacks and statistical tests for stream ciphers. More specifically, we will investigate one of the most promising techniques for attacking a stream cipher, the algebraic attack, and a statistical test for pseudo-randomness evaluation, the minimum nonlinear complexity.The algebraic attack technique was presented in 2002 and applies to a large class of stream ciphers, with linear internal state transition. The technique is based on the solution of an over-defined system of equations. The results of the attack were more than impressive. It was shown that ciphers, widely used in applications, must not be trusted.The minimum nonlinear complexity can be seen as the minimum nonlinear realization of a cipher. In contrast to the linear case, that has been extensively studied, the nonlinear case has been neglected due to its stiffness. Recently, an efficient algorithm has been proposed for the calculation of the minimum nonlinear complexity.Liaisons with existing criteria and attacks will be pointed out, and a new secure stream cipher will be introduced combining the results of the above analysis with existing design guidelines.
Original text from CORDIS.
Participants
- KATHOLIEKE UNIVERSITEIT LEUVEN · LEUVENCoordinatorBelgium
Links
Data: CORDIS, © European Union