CNN-LSTM based Spectrum Sensing in Full-Duplex-aided CR-IoT Networks

In this paper, we propose a convolutional neural network-long and short-term memory (CNN-LSTM) based secure cooperative spectrum sensing (CSS) in full duplex-aided cognitive radio-Internet of things (CR-IoT) network. It enhances security by separating unauthorized CR-IoT users from authorized CR-IoT users. Avoiding the unauthorized CR-IoT users degrades the detection gain performance. According to simulation results, the proposed CNN-LSTM algorithm performs better detection gain compared to the conventional scheme in terms of security.