Development of a cognitive model for analysis of the impact of threats on the security level of a computer network

• Дата: 2019
• Автори: Saliieva, O. V., Yaremchuk, Yu. E.
• Формат: Стаття
• Мова: Ukrainian
• Опубліковано: Інститут проблем реєстрації інформації НАН України 2019
• Теми: інформаційна безпека; загрози безпеці; комп’ютерна мережа; когнітивне моделювання; нечітка когнітивна карта; information security; security threats; computer network; cognitive modeling; fuzzy cognitive map
• Онлайн доступ: http://drsp.ipri.kiev.ua/article/view/199268
• Назва журналу: Data Recording, Storage & Processing

It is impossible to imagine modern life without the use of computer networks, which offer many opportunities for their users. However, computer networks and their resources are constantly at risk of being infected by malware or from various types of network attacks. Therefore, for the proper functioning of the network and the provision of quality services, special attention is paid to the organization of network security. To solve problems related to the protection of computer networks, it is convenient to apply a cognitive approach based on the creation and analysis of a fuzzy cognitive map that reflects the status of the situation under study. Considering the importance of this issue, a cognitive model has been developed that reflects the security level of a computer network when exposed to potential threats. A matrix of interaction of concepts is formed and on the basis of it the basic system indicators are calculated, such as: consonance, dissonance and influence of concepts on the system. The analysis of these indicators revealed the most dangerous threats to network security such as: malware, physical impact on the network by the attacker, unintentional actions and errors of network users. Increasing the impact of these concepts on the network, various scenarios have been simulated, resulting in a relative change in the security level of the network. It is revealed that with increased influence of the concept of «unintentional actions, mistakes of network users» the security will be reduced the most, and with the increase of the concept of «physical influence on the network by the attacker» the fault tolerance of network operation will worsen as much as possible. The results of this study will contribute to the implementation of the necessary mechanisms to prevent, protect and control access at the appropriate levels of network infrastructure. Tabl.: 2. Fig.: 5. Refs: 13 titles.