An important security problem faced by wireless sensor networks (WSNs) is malware propagation. Traditional malware propagation models take less consideration of the impact of the attack and defense processes on malware propagation results. We analyzed the micro-mechanism of malware propagation in WSNs from the perspective of the game theory, established the attack–defense game model of WSNs, derived the mixed Nash equilibrium solution of the game model, and derived the contagion probability of malwares according to the mixed Nash equilibrium strategy of both sides in the game, thus establishing the malware propagation model in WSNs. From the analysis of the theoretical model, we deduced the relationship between the steady-state infection ratio and the game parameters and obtained the conditions for the long-term existence of malwares. In addition, we simulated the propagation process of malwares in WSNs with cellular automaton and the simulation results verified the correctness of the theoretical model. The results have theoretical guiding significance for suppressing the propagation of malwares in WSNs.

无线传感器网络（WSN）面临的一个重要安全问题是恶意软件传播。传统的恶意软件传播模型很少考虑攻击和防御过程对恶意软件传播结果的影响。我们从博弈论的角度分析了无线传感器网络中恶意软件传播的微观机制，建立了无线传感器网络的攻守博弈模型，推导了该博弈模型的混合纳什均衡解，并根据该算法推导了恶意软件的传染概率。博弈双方的混合Nash均衡策略，从而建立了WSN中的恶意软件传播模型。通过对理论模型的分析，推导出稳态感染率与游戏参数之间的关系，并为恶意软件的长期存在提供了条件。此外，我们利用细胞自动机模拟了无线传感器网络中恶意软件的传播过程，仿真结果验证了该理论模型的正确性。该结果对于抑制WSN中恶意软件的传播具有理论指导意义。