Affiliations: [a] Department of Information Systems, Faculty of Computing, University of Uyo, Uyo, Nigeria | [b] TETFund Center of Excellence in Computational Intelligence Research, University of Uyo, Uyo, Nigeria | [c] Department of Information Systems, Faculty of Computing, University of Uyo, Uyo, Nigeria | [d] Department of Cyber Security, Faculty of Computing, University of Uyo, Uyo, Nigeria | [e] Department of Computer Science, Faculty of Computing, University of Uyo, Uyo, Nigeria | [f] Department of Data Science, Faculty of Computing, University of Uyo, Uyo, Nigeria | [g] Department of Mathematics and Computer Science, Ritman University, Ikot Ekpene, Nigeria
Correspondence:
[*]
Corresponding author: Daniel E. Asuquo, Department of Information Systems, Faculty of Computing, University of Uyo, Uyo, Nigeria. E-mail: [email protected].
Abstract: The proliferation of interconnected devices is driving a surge in the demand for wireless spectrum. Meeting the need for wireless channel access for every device, while also ensuring consistent quality of service (QoS), poses significant challenges. This is particularly true for resource-limited heterogeneous devices within Internet of Things (IoT) networks. Cognitive radio (CR) technology addresses the shortcomings of traditional fixed channel allocation policies by enabling unlicensed users to opportunistically access unused spectrum belonging to licensed users. This facilitates timely and reliable transmission of mission-critical data packets. A cognitive radio-enabled IoT (CR-IoT) network is poised to better accommodate the growing demands of diverse applications and services within the smart city framework, spanning areas such as healthcare, agriculture, manufacturing, logistics, transportation, environment, public safety, and pharmaceuticals. To minimize switching delays and ensure energy and spectral efficiency, this study proposes a hybrid intelligent system for efficient channel allocation and packet transmission in CR-IoT networks. Leveraging Support Vector Machine (SVM) and Adaptive Neuro-Fuzzy Inference System (ANFIS), the system dynamically manages spectrum resources to minimize handoffs while upholding QoS. A Java-based simulation integrates system outputs with interference temperature data to accommodate service demands across 2G–4G spectrums. Evaluation reveals SVM’s 98.8% accuracy in detecting spectrum holes and ANFIS’s 90.4% accuracy in channel allocation. These results demonstrate significant potential for enhancing spectrum utilization in various IoT applications.
Keywords: Cognitive radio network, internet of things, packet transmission, intelligent whitespace detection, optimized channel allocation
