An Analytical Method for Dynamic Evolution of Attack Process based on Markov Game
Weicheng Yan · International Journal of Performability Engineering · 2017
Because of the randomness of attacker and defender's strategy selection, the state variation during the network attack process must be a random process.So, the network attack and defense process can be abstracted a confrontation of multi-state based on different gains matrix.This paper describes the random of attack and defense strategy selection with Markov decision, and extends the Markov game model from single-state to multi-state and multi-agent.After that, it proves the existence of equilibrium strategy and gives the solving method of nonlinear programming.Finally, deduction and simulation analysis of the practical example indicate that this model's method is correspond to the actual application and the evaluation result is accurate, so it can be used to have a more detailed simulation to network attack and defense process in reality.