A fuzzy-logical model for assessing the state of an information system (IS) during DDoS attacks is proposed. The model is based on the use of a system of linguistic variables that describe critical parameters of network traffic and the operational characteristics of the IS. To form an assessment of the current risk, Mamdani-type fuzzy inference was applied, followed by defuzzification using the centre-of-gravity method. A simulation experiment was conducted, implemented in the Python environment using the scikit-fuzzy library. The results obtained confirmed that the proposed model adequately reflects the dependence of the integral risk on changes in load parameters, ensures continuity of assessment and sensitivity to critical combinations of traffic factors. The results obtained provide grounds for considering the developed model as a basis for the synthesis of an IS status monitoring module.
Valerii et al. (Mon,) studied this question.