Trustworthy task allocation in IoT: a cognitive game-theoretical use case

This Chapter analyzes how malicious attacks affect the performance of a heterogeneous Internet of Things (IoT) system where cognitive devices collaborate to negotiate task assignments. In the reference scenario, the involved devices create clusters, each managed by a Cluster Head (CH). Whenever a task is required, the CH triggers spectrum sensing to detect spectrum holes that can be opportunistically exploited by the nodes of the cluster for task allocation. In this scenario, not all the nodes are Honest Nodes (HN). Indeed, Malicious Nodes (MNs) may hinder the process and try to disrupt it by providing tampered data, which would lead to a higher likelihood that the spectrum sensing is not performed correctly. When the spectrum is considered free, the cluster nodes negotiate to execute the required task by means of an auction-based game theory approach. The negotiation takes into account two factors: the reward gained from contributing to the execution of the task, which is provided to the node that wins the competition, and the energy cost to perform the task. Specifically, the Chapter investigates how MNs affect the reward aspect when they try to gain maximum control over the task and potentially launch a Denial of Service (DoS) attack. Extensive simulations are run to assess the effect of the key system parameters on the overall performance and provide recommendations for future research.