An Effective Hybrid Spectrum Sensing Methodology in Cognitive Radio Network Using Deep Temporal Convolution Network

“Spectrum sensing is a crucial component in Cognitive Radio Networks (CRN)”. CRN systems' capacity to reliably and quickly sense the principal signal is a vital requirement. One effective technique to detect activity in Primary Users (PU) is Hybrid Spectrum Sensing (HSS). It involves various detectors to reach an agreement regarding the PU state. The ineffective use of the permitted frequency makes CR a promising technology for current and future communications. The ability to use the available bandwidth of other wireless networks for communication to boost their use is the reason for this. A novel methodology for HSS in CR utilizing deep learning is offered to get around this. The proposed framework comprises several phases; initially, the model utilizes the energy from energy detection. The collected data are fed into the deep learning model as Deep Temporal Convolutional Network (DTCN) for the prediction objective. Finally, the validation is carried out and compared with different schemes. Hence, the proposed model outperforms with better energy detection than existing radio technologies.