MODULATION CLASSIFICATION FOR EFFECTIVE SPECTRUM SENSING IN CR NETWORKS USING DL

Abstract

Most important problems in spectrum sensing are the quality of service (QoS) issues. There are several QoS issues that addressed in conventional systems for effective spectrum sensing. Type of modulation is most important one that causes unsuccessful dynamic spectrum access of the cognitive radio network (CRN). This work focuses on implementation of effective spectrum sensing method by the perfect modulation type. This work suggests authentication of the best modulation classification to increase the QoS and quality of experience (QoE) of the spectrum sensing under primary user (PU) and secondary user (SU). The proposed method is capable of classifies the performance of 12 various modulation methods in terms of signal to noise ratio and throughput.  Deep learning methods are broadly used in the cognitive radio to reveal the learning capability.  This work utilizes the deep learning convolution neural network (DLCNN) method to select the suitable modulation scheme in the CRN environment. The simulation results show that the proposed method gives the superior performance over conventional approaches.