An investigation into the development of a sustainable cognitive internet of things for cyber-physical systems

Abstract The integration of the Cognitive Internet of Things (CIoT) into Cyber-Physical Systems (CPSs) is driving the development of the next generation of intelligent, adaptive, and autonomous infrastructures. CIoT extends traditional IoT by embedding cognitive capabilities, such as data prediction, pattern identification, and decision-making, into networked devices, enabling CPS to dynamically adapt to changing operational environments. A serious utilisation of this transformation is efficient spectrum allocation, which is becoming increasingly complex in highly dynamic and heterogeneous CIoT environments. Due to the exponential growth of connected devices and data-intensive applications, traditional static spectrum management methods are no longer viable. Therefore, the futuristic architecture of CIoT-enabled CPS incorporates cognitive radio technology to adapt to the radio frequency environment. The dynamic spectrum access (DSA) ability allows CPS components to recognise underutilised frequency bands, allocate spectrum resources in real time, optimising communication efficiency and reducing interference. This cognitive spectrum-aware approach enhances the network performance and reinforces the dependability of CPSs against cyber threats. Following this approach, this study presents a simulation-based evaluation of a proactive spectrum-allocation framework for CIoT-enabled CPS, incorporating dynamic parameters such as mobility, interference, and energy constraints into the decision-making state vector. Using reinforcement learning, the system anticipates spectrum demand and optimises channel selection. The simulation results demonstrate that the CIoT model significantly outperforms the random baseline of reducing latency, minimising collisions, and improving spectrum utilisation. Outcomes validate the effectiveness of intelligent, context-aware resource management for enhancing communication reliability and efficiency in CPS environments.