Construction of evolutionary stability and signal game model for privacy protection in the internet of things.

The research focuses on privacy protection in the Internet of Things environment. A model based on evolutionary theory game and signal game mechanism is proposed to analyze and optimize privacy protection strategies. The study introduces evolutionary game theory and signal game mechanism to construct a game model between users, devices, network operators, and attackers. Detailed discussions are conducted on factors such as privacy protection needs, information asymmetry, and privacy leakage risks. The proposed Multi-stage Signal Game and Deep Learning Model for IoT Privacy Protection (IoT-PSGDL) performed the best on privacy protection effectiveness, at 98.25% on the CIC IoT dataset, with a policy update speed of 7.42 updates/second and a system response time of 35.12ms. Compared with other models, the proposed model performed well in multiple metrics, such as privacy protection persistence (97.56%), communication latency (54.12ms), and data storage security (96.75%). In addition, privacy protection strategies such as data encryption performed the best in the experiment, with a privacy protection success rate of 96.72% and the lowest privacy leakage probability of only 2.14%. The significance of the research lies in providing an efficient and dynamically optimized privacy protection strategy that can effectively respond to various privacy threats in complex Internet of Things environments.