Energy-Efficient Joint Spectrum Sensing and Power Allocation in Cognitive IoT Under SSDF Attack

Massive wireless connections for Internet of Things (IoT) terminals require large amounts of spectrum resource, Cognitive Radio-Enabled IoT (CR-IoT) is emerged as an effective solution to alleviate the scarcity of spectrum resources. However, CR-IoT faces several problems, such as spectrum sensing data falsification (SSDF) attacks, when using of cooperative spectrum sensing, and energy efficiency due to equipped with low-powered IoT terminals. To tackle these challenges, secure power allocation is considered a promising technique for improving energy efficiency. In this article, a joint spectrum sensing and secure power allocation scheme is proposed under SSDF attacks. First, to ensure fair resource allocation and resistance against SSDF attacks, a weighted data transmission time allocation scheme is designed based on the trust degree, where the trust degree is obtained by using an online learning algorithm and is updated in the sensed power allocation process. Then, a joint optimization problem of spectrum sensing time, number of cooperative nodes, and transmit power is formulated to maximize the CR-IoT sum energy efficiency, subject to the constraints of CR-IoT node's average transmission power, minimum data rate, and the interference to the primary user. To handle the nonconvexity of the resulting problem, an alternating iterative optimization algorithm is proposed to iteratively solve three subproblems: 1) spectrum sensing time optimization; 2) number of cooperative nodes optimization; and 3) transmission power optimization. Simulation results demonstrate that the proposed scheme is effective in improving energy efficiency under various SSDF attack scenarios.