Machine learning–based cognitive radio for spectrum detection in the optical 5G network for 256-QAM

Abstract The rapid expansion of 5G optical networks demands advanced spectrum sensing techniques to maximize radio frequency spectrum utilization and support diverse applications. Intelligent hybrid spectrum sensing integrates multiple sensing methodologies, such as energy detection, matched filtering, and cyclostationary feature detection, with machine learning algorithms to enhance detection accuracy, reduce latency, and improve network performance. By leveraging historical data, machine learning enables adaptive decision-making, predicting spectrum availability in real-time while dynamically adjusting to varying signal conditions. This hybrid approach overcomes the limitations of individual techniques, such as sensitivity to noise and high computational complexity, by optimizing the sensing process and ensuring reliable spectrum access. Furthermore, intelligent hybrid spectrum sensing facilitates the coexistence of licensed and unlicensed users, mitigating interference and enhancing spectrum efficiency. This capability is crucial for supporting high-speed data transmission, ultra-low latency communication, and emerging applications like autonomous systems, smart cities, and the Internet of Things (IoT). As 5G networks evolve toward higher-order modulation schemes, such as 256-QAM, robust spectrum sensing becomes increasingly vital in maintaining network reliability and efficiency. The implementation of intelligent hybrid spectrum sensing in optical 5G networks ensures enhanced spectral efficiency, seamless connectivity, and improved network resilience, making it a key enabler for future wireless communication advancements.