Identification of the Traffic Model Parameters for Network and Cloud Platform Security Management

The paper suggests a technique for identification of incomplete and inconsistent knowledge about the traffic model parameters. The technique is applied for network and cloud platform security management and is based on neural network transformation. The elements of the stochastic transition probability matrix - the main component of the traffic model - are subject to identification. The proposed technique considers uncertainties of a non-stochastic nature. It makes possible to increase the reliability of modeling and the degree of validity of decisions made to manage the security of networks and cloud platforms.