Signal Game and Compact Particle Swarm Optimization Based on Pareto Distribution for Attack-Defense Eavesdropping Confrontation Analysis in Internet of Things

The information transmission in internet of things (IoT) communication brings eavesdropping attacks to people. How to take defense strategy to defend eavesdropping attacks in IoT is a significant research content. In order to analyze the attack-defense confrontation of eavesdropping, a signal game is established in this paper. Nash equilibrium (NE) is important for analyzing signal game. The NE of the signal game from both macro and micro levels is analyzed in this paper. Because solving the NE is difficult, a particle swarm optimization algorithm based on Pareto distribution (pCPSO) is proposed to solve this hard problem. It is a new idea to use intelligent computing to solve NE. The pCPSO not only improve the optimization performance, but also reduce the use of memory. The simulation experiment analyze how the changing value of parameters direct benefit, defense benefit and defense cost influence the profit of the defender. The defense benefit has greater influence on the profit of the defender than direct benefit. The experimental results prove the reliability and availability of the model. The experimental results also show that the defender using pCPSO can obtain higher profit than using Posterior, Prior, PSO, Nonlinear optimization, Convex optimization and Gradient ascent. This paper provide an important idea for defender to analyze attack-defense confrontation of eavesdropping.