Game theoretical-based task allocation in malicious cognitive Internet of Things

In this chapter we consider a heterogeneous Internet of Things (IoT) scenario in which sensor nodes belong to different platforms and form different clusters managed by their own cluster head (CH). This configuration is exploited to foster nodes collaboration in sensing activities with coordinated resource usage. We assume that the considered nodes have cognitive radio (CR) and exploit device-to-device (D2D) communications. The nodes collaboratively sense the spectrum, through standard energy detection, to find spectrum holes to be opportunistically exploited for task allocation by the CH. Some of the nodes may act as malicious nodes (MNs) trying to disrupt this process by providing tampered data, trying to lead to a higher overall probability of error of the spectrum sensing. At this point, task allocation is performed by means of a game theoretical-based approach considering two elements: the gain that is won for its contribution to sensing and for the execution of the task (in case, it wins the competition), and the cost in terms of energy to be consumed in case the task is executed. In particular, we investigate the impact of tampered data in the former aspect so that the MNs try to gain as much as possible control of the allocated task, thus performing a Denial of Service (DoS) attack. Extensive simulations are performed to evaluate the impact of the main system parameters on the overall performance and provide guidelines for future work.