Toward Secure Transmission in Fog Internet of Things Using Intelligent Resource Allocation

Abstract

Thanks to the full play of the synergy of local node sets, fog Internet of things technology has become the focus of many researchers. Unfortunately, the data transmission of the sensing terminal therein has the risk of being eavesdropped, due to the openness of the wireless channel and the large-scale deployment of different types of terminals. Meanwhile, the encryption method based on cryptography is difficult to apply to the Internet of things terminals with limited capacity because of its high computational complexity. This paper considers the use of physical layer security technology where the fog node improves the secrecy capacity of the desired channel by sending artificial noise to the untrusted third party to ensure the reliable transmission of data. Specifically, an intelligent resource allocation method based on deep reinforcement learning is proposed to realize the rapid allocation of link resources and interference noise power in the scene, where the relevant elements, such as state, action and reward, are reasonably designed. Simulation results demonstrate that the presented algorithm could effectively resolves the allocation problem, and its transmission delay obviously outperforms that of multiple comparison methods.