Distributed robust channel allocation for clustered cognitive radio-based IoT networks using graph theory

Abstract

The exponential growth in the Internet of Things (IoT) devices for the Internet of Everything (IoE) services demands more operating spectrum. Utilizing the unlicensed spectrum by a large number of IoT networks leads to congestion in the unlicensed spectrum. To mitigate the scarcity of radio spectrum for IoT networks, integration of the cognitive radio technology with IoT networks allows IoT devices to operate and share the licensed spectrum with primary users (PUs). For efficient licensed-spectrum sharing, a cognitive radio-based spectrum assignment algorithm is proposed for IoT networks, which minimizes network interference and ensures connectivity against the PUs activity. For interference reduction, a conflict graph is used to determine the potential interfering links in the network, and channels are accordingly assigned to the radio interfaces of each IoT device. To ensure connectivity in the network, an ordered pair of channels is assigned to the radios of the IoT devices such that the network topology is robust to the presence of the PUs on multiple channels. The robustness of the network topology avoids frequent channel switching and improves the energy efficiency of the network. Simulation results show that the proposed algorithm minimizes overall network interference, and achieves 100% successful packet transmission, compared to other channel assignment algorithms. The algorithm shows that the network is not partitioned due to the PUs’ presence on up to half of the available licensed channels, which significantly reduces the amount of channel switching required, and conserves energy in the IoT devices.