Security Configuration and Pricing Scheme for Satellite-Terrestrial IoT: A Stackelberg Game

With the introduction of the concept of 6G ubiquitous intelligence, a network infrastructure that can provide ubiquitous intelligence services has been expected widely. Among all the promising technologies, satellite-terrestrial Internet of Things (IoT) networks are one of the key enablers of the implementation of 6G IoT, offering multiple services for remote IoT applications, such as disaster rescue and remote area monitoring in global coverage. As the satellite-terrestrial link is vulnerable to eavesdropping which can greatly damage users' privacy, implementing information security protection for reliable communication is extremely necessary. Therefore, we introduce the security service provider to offer encryption services for remote IoT users. However, higher security configuration can lead to higher overhead for the service provider and higher prices for users. Thus, to find the optimal service price and encryption security configuration, this paper models the interaction between IoT users and the service provider as a Stackelberg game. To achieve the Nash equilibrium, we formulate the decision-making process as a Markov Decision Process. Then, we apply the ‘Wolf-PUC’ multi-agent reinforcement learning algorithm to learn the optimal security configuration and pricing strategies. Finally, the feasibility and performance of the algorithm are demonstrated with our simulation results.