A Model for Quantifying the Agent of a Complex Network in Conditions of Incomplete Awareness

Abstract

Purpose of the article: development of a mechanism for quantitative evaluation of elements of complex information systems in conditions of insufficient information about the presence of vulnerabilities. Research method: mathematical modeling of uncertainty estimation based on binary convolution and Kolmogorov complexity. Data banks on vulnerabilities and weaknesses are used as initial data for modeling. The result: it is shown that the operation of an element of a complex network can be represented by data transformation procedures, which consist of a sequence of operations in time, described by weaknesses and related vulnerabilities. Each operation can be evaluated at a qualitative level in terms of the severity of the consequences in the event of the implementation of potential weaknesses. The use of binary convolution and universal coding makes it possible to translate qualitative estimates into a binary sequence – a word in the alphabet {0,1}. The sequence of such words — as the uncertainty function — describes the possible negative consequences of implementing data transformation procedures due to the presence of weaknesses in an element of a complex system. It is proposed to use the Kolmogorov complexity to quantify the uncertainty function. The use of a Turing machine for calculating the uncertainty function provides a universal mechanism for evaluating elements of complex information systems from the point of view of information security, regardless of their software and hardware implementation.