Orthogonal Frequency Division Multiplexing (OFDM) Sub-Band Channel Classification Based on Adaptive Thresholding

Abstract

The radio frequency (RF) spectrum has experienced increased congestion due to a growing number of wireless devices and escalating demand for higher data rates. Cognitive radio (CR) enables wireless devices to sense their environment and efficiently utilize available spectrum resources. However, spectrum sensing in CR systems still remains a challenge due to the increased demands for higher data rates and efficiency. Existing research often trades complexity for accuracy, however a balance between accuracy and complexity is required for a good wireless communication. This study proposes a low-complexity, high-accuracy spectrum sensing method for Orthogonal Frequency Division Multiplexing (OFDM) based cognitive radio. It combines a Fast Fourier Transform (FFT) filter bank for signal decomposition with a novel convex-optimized adaptive threshold algorithm to classify sub-band occupancy. Then, an optimal Adaptive threshold technique is designed through convex optimization, enhancing classification accuracy by optimizing statistical properties. Simulation results show improved accuracy compared to other related works, with a higher probability of detection achieving a probability of detection (????????) of 0.9686 at a probability of false alarm (????????????) of 0.1 and an SNR of 10 dB indicating the proposed technique's promise for efficient spectrum utilization.