There is a wide range of interdisciplinary intersections between cyber security and artificial intelligence (AI). On one hand, AI technologies, such as deep learning, can be introduced into cyber...

Nowadays, benefit from more powerful edge computing devices, edge artificial intelligence (edge-AI) could be introduced into Internet of Things (IoT) to find knowledge derived from massive sensory data, such as cyber results or models of classification, detection and prediction from physical environments. Heterogeneous edge-AI devices in IoT will generate isolated and distributed knowledge slices, thus knowledge collaboration and exchange are required to complete complex tasks in IoT intelligent applications with numerous selfish nodes. Therefore, knowledge trading is needed for paid sharing in edge-AI enabled IoT. While most existing works only focus on knowledge generation rather than trading in IoT. To address this issue, in this paper, we propose a peer-to-peer (P2P) knowledge market to make knowledge tradable in edge-AI enabled IoT. We firstly propose an implementation architecture of the knowledge market. Moreover, we develop a knowledge consortium blockchain for secure and efficient knowledge management and trading for the market, which includes a new cryptographic currency knowledge coin, smart contracts and a new consensus mechanism Proof-of-Trading (PoT). Besides, a non-cooperative game based knowledge pricing strategy with incentives for the market is also proposed. The security analysis and performance simulation show the security and efficiency of our knowledge market and the incentive effect of knowledge pricing strategy. To the best of our knowledge, it is the first time to propose an efficient and incentive P2P knowledge market in edge-AI enabled IoT.

Social networks are very important social cyberspaces for the human being to establish efficient relations and collaborations for life and production. Currently, information-centric networks (ICN) are main trend of next generation networks, which promote traditional social networks to information-centric social networks (IC-SN). Because of the complexity and openness of social networks, it is a very key issue to filter security services for users. However, existing scheme were proposed just for traditional social networks, which cannot satisfy the new requirements of IC-SN including extendibility, data mobility, non-IP address, flexible deployment, etc. To address this challenge, a fog computing based content-aware filtering for security services, FCSS, is proposed in information centric social networks. Firstly, the assessments and content matching scheme is proposed for IC-SN. Secondly, fog computing based content-aware filtering scheme is proposed for security services, in which the content label technology is used. The simulations and evaluations show the advantage of FCSS in terms of hit ratio, filtering delay, and filtering accuracy.

Edge computational intelligence, integrating artificial intelligence (AI) and edge computing into Internet of Things (IoT), will generate many scattered knowledge. To enable auditable and delay-sensitive IoT services, these knowledge will be shared among decentralized intelligent network edges (DINEs), end users, and supervisors frequently. Blockchain has a promising ability to provide a traceable, privacy-preserving and tamper-resistant ledger for sharing edge knowledge. However, due to the complicated environments of network edges, knowledge sharing among DINEs still faces many challenges. Firstly, the resource limitation and mobility of DINEs impede the applicability of existing consensus tricks (e.g., Poof of Work, Proof of Stake, and Paxos) of blockchain. Secondly, the adversaries may eavesdrop the content of edge knowledge or entice the blockchain to forks using some attacking models (like man-in-the-middle attack, denial of services, etc.). In this article, an user-centric blockchain (UCB) framework is proposed for preserving edge knowledge sharing in IoT. Significant superiorities of UCB benefit from the proof of popularity (PoP) consensus mechanism, which is more energy-efficient and fast. Security analysis and experiments based on Raspberry Pi 3 Model B demonstrate its feasibility with low block generating delay and complexity.

Individually reinforcing the robustness of a single deep learning model only gives limited security guarantees especially when facing adversarial examples. In this paper, we propose DeSVig, a Decentralized Swift Vigilance framework to identify adversarial attacks in an industrial artificial intelligence systems (IAISs), which enables IAISs to correct the mistake in a few seconds. DeSVig is highly decentralized, which improves the effectiveness of recognizing abnormal inputs. We try to overcome the challenges on ultra-low latency caused by dynamics in industries using peculiarly-designated mobile edge computing and generative adversarial networks (GANs). The most important advantage of our work is that it can significantly reduce the failure risks of being deceived by adversarial examples, which is critical for safety-prioritized and delay-sensitive environments. In our experiments, adversarial examples of industrial electronic components are generated by several classical attacking models. Experimental results demonstrate the DeSVig is more robust, efficient, and scalable than some state-of-art defences.

In industrial automation, numerous devices are interconnected in smart factories for further monitoring and control. Various infrastructure devices in industrial automation are usually used for control instruction distribution, data collection, and collaboration of the industrial applications. Recent security threats on industrial automation are more frequent and the industrial control systems (ICS) lack trust mechanism. Blockchain has been introduced due to its decentralization and security promise, but the election results in the original designs could be biased without collaboration trust, which leads the blockchain based industry applications invalid. What's more, in existing solutions, both supernodes and normal nodes in blockchain cannot transfer their control authorities for disaster backup. To address the aforementioned challenges, this paper proposes a collaborative trust based unbiased control transfer mechanism, which can realize a dynamic allocation of industrial control. First, a novel delegated proof of stake, CT-DPoS, based consensus is proposed for determining the authorities of control dynamically and unbiasedly, by designing a lightweight trust propagation protocol. Second, a control transfer mechanism for checking, alarming, and restarting CT-DPoS is devised for the disaster backup. The simulation results demonstrate the CTDPoS is feasible and effective for industrial automation security.