Nikolai A. Grusho

Problems of modeling in the analysis of covert channels

Sometimes the analysis of covert channel is weakly dependent on the correctness of probabilistic models, but more often the result of such analysis is seriously dependent on the choice of aprobabilistic model. We show how the problem of detection of covert communications depends on the correctness of the choice of probabilistic model. We found the dependence of judgments about invisibility of covert communication from the bans in a probabilistic model of the legal communication.

Generation Of Probability Measures With The Given Specification Of The Smallest Bans.

A ban means a sequence which has zero probability in a finite space. Generation of probability models is carried out, as a rule, from simpler models by introduction of additional restrictions. However fulfilment of required properties for stochastic processes requires the proof in case of introduction of additional restrictions. In particular, the proof is required that restrictions on admissibility of trajectories don’t destroy the property of being a random process that is to satisfy to the Kolmogorov’s theorem. The paper deals with conditions under which introduction of restrictions on trajectories of random sequences according to the given specification of the smallest bans again generates random process. When the probability measure Q is generated by restrictions defined by bans we consider testing of sequence of hypotheses H0, n : Qn against H1, n : Pn, where Q possesses the specification of the smallest bans, P is a uniform measure and Qn, Pn are projections of measures Q and P . The existence of consistency of sequence test defined by bans is investigated.

Criteria On Statistically Defined Bans.

The method of statistical determination of smallest bans of probability measures on discrete spaces is offered. Consistency of the constructed estimates is shown. The application of the received estimates for testing of statistical hypotheses in the discrete spaces is constructed. It is shown that estimates of bans can generate consistent sequences of criteria in some sense.

Information security architecture synthesis in distributed information computation systems

This work considers the task of synthesizing information security architecture in the distributed information computation systems. The synthesis has been based on elementary structures described in the form of diagrams of solving action protection problems in the process of computation in the computer system and during network exchange.

Intelligent data analysis in information security

At present, there are a large number of trends and developments in artificial intelligence systems. This paper deals with the available intelligent data analysis in information security and the possibilities provided by data mining. Examples of tasks that can be solved using artificial intelligence systems are presented, including safe resource management, prediction of critical states and failures, resolution of conflicts in computational processes and information security regulations, and forensics.

Alternative mechanisms for information security

The usage of doubtful data for information protection is considered. It is shown that by means of unauthenticity it is possible to protect confidentiality, integrity and availability. The basic methods of usage of unauthenticity are mentioned. We introduce the mathematical models and present estimations of quality for information protection with help of injection of unreliable information.