Author(s)

Gong Dejing

Corresponding Author

Gong Dejing

Abstract

When formulating network attack strategy, there is an uncertainty in target network information, and the attacker lacks comprehensive, reliable and real-time attack basis, so it is difficult to achieve the attack effect. Therefore, a scientific and complex network attack method was proposed. The earnings, loss, cost and risks of the attacker were analyzed, and the index system was established, and then a dynamic Bayesian network was used to comprehensively evaluate the attack effect of network nodes, thus overcoming the shortcoming that traditional node importance degree evaluation methods conducted static evaluation by relying on the single index of network topology or target node. Therefore, the simulation experiments indicated that this method integrated more node relations and observation information when attacking, avoiding the gap between the actual attack effect and the theoretical expectation when carrying out attack by means of static evaluation, and at the same time, the attack precision was more accurate and the attack efficiency was higher.

Keywords

Network attack, Dynamic Bayesian network, Network nodes, Comprehensive evaluation