350 rub
Journal Neurocomputers №6 for 2010 г.
Article in number:
Prevention of intrusions in local information-measuring system
Authors:
G. F. Malykhina, O. V. Shalaevsky
Abstract:
Results of measurement give the objective information on a condition of technical, ecological or biological system. Accuracy of performance of measurement has crucial importance in control systems of technological processes or the commercial account. The increase in an error of measurement at the account of production of gas or oil branch can lead to mullions-strong losses for the state. Therefore the big attention is given development of system of metrological services which activity is directed on maintenance of unity of measurements and creation of conditions of objectivity of the measuring information.
A little bit smaller attention is given protection of the measuring information against deliberate updating. Updating of results of measurement in the commercial account is caused by attempts of theft, updating of data in monitoring systems of technological processes can lead to misuse of the equipment and resources and even to technogenic accident. Modern information-measuring systems often use architecture of local networks; unite some workstations for operators, a database for storage and system of the analysis for processing the measuring information; interfaces of communication with measuring installations and interfaces of telecommunication systems for transfer of the information to the central offices. Dangerous for IMS are represented: infringements of integrity of data which can be considered as attempt of theft of resources, and also with the purpose of concealment of approach of dangerous modes of operation. Probably going removal of results of measurement and addition of new irrelevant data.
Division of the signals describing information processes in local IMS allows to receive a number of advantages. More precisely to define the fact of intrusion, using time characteristics of development of attack; to use an artificial neural network for allocation of a signal of attack in real time; to diagnose type of intrusion, being based on the signatures presented in the form of СММ which parameters are estimated at investigation phase IMS.
Pages: 40-42
References
- Cichocki, A, Amari, Sh.,Adaptive blind signal and image processing // John Wiley& Sons, LTD. WestSussex. England. 2002.
- Бочков М. В. Статистический метод идентификации состояний компьютерной сети в условиях программных атак // Компьютерные системы. 2004. № 3.
- Овчинникова Е. В. Обнаружение аномалий на основе вычисления дисперсионного расстояния. Проблемы информационной безопасности // Компьютерные системы. 2006. № 3.