Cyber-attack intensity interval forecasting on objects of critical information infrastructure

Authors: Krakovskiy Yu. M., Kurchinskiy B. V., Luzgin A. N.

Annotation: In the modern world, cybersecurity issues occupy one of the key and extremely significant strategic niches in a state planning and management system. Cyber-attacks occur daily, and their number grows exponentially. In these conditions, the cybersecurity issues regarding to critical information infrastructure objects of a state are becoming especially topical. The relevance of research in the direction of creating and improving technologies for protection against cyber-attacks on relevant facilities is beyond doubt. In this paper are investigated the results of interval forecasting for cyber-attack intensity through an intelligent algorithm based on a probabilistic neural network with a dynamic updating value of the smoothing parameter. The data on the hourly cyber-attack intensity obtained by means of honeypots from March to September 2013 are used as initial data. Obtained results testify to the high forecasting accuracy of the proposed algorithm. According to the study results, the authors give necessary practical recommendations about interval forecasting results application when protecting against cyber-attacks on critical information infrastructure objects.

Keywords: interval forecasting, cyber-attacks, critical information infrastructure