Rate satisfaction-based power allocation for NOMA-based cognitive Internet of Things

Abstract

The transmission performance of Internet of Things (IoT) has been limited by the spectrum resource shortages. Integrating cognitive radio (CR) in IoT, cognitive IoT (CIoT) can increase available spectrum by accessing the licensed spectrum for primary user (PU), providing that the interference to the PU can be well controlled. In this paper, a clustering CIoT based on Non-orthogonal Multiple-access (NOMA) is proposed to improve transmission performance, where the cluster heads use NOMA to relay the data from the cluster nodes to the data center. The power for cluster heads and cluster nodes is jointly optimized to maximize the average total transmission rate of CIoT, while guaranteeing satisfactory rate for each cluster and controlling the interference power to the PU. The clustering algorithm for the CIoT is proposed to classify the nodes and select the cluster heads. The optimal number of the clusters is obtained to decrease the rate loss, and the cluster head replacement is presented to avoid energy exhaustion of each cluster head. The simulation results have indicated that the NOMA-based clustering CIoT can improve transmission performance while guaranteeing satisfactory rate for each CIoT node.