Uplink resource allocation for multicarrier grouping cognitive internet of things based on K-means Learning

Abstract

Integrating cognitive radio into internet of things (IoT), cognitive IoT (CIoT) can achieve more available spectrum resources by accessing the licensed frequency bands of primary user (PU), providing the communication of the PU is not disturbed. In this paper, a multicarrier grouping CIoT is proposed to decrease the interference to the PU, where the CIoT nodes are divided into several groups by K-means Learning and each group transmits data to the base station in the allocated time slot. In the underlay case, the uplink resource allocation is formulated as an optimization problem, which seeks to maximize the transmission rate of the CIoT under the constraint of the interference power to the PU, by jointly allocating subcarrier and power for each node. The optimization problem is solved by group power optimization and node power optimization. The upper and lower bounds of the optimal number of groups are calculated. In the overlay case, the lower bound of sensing time is achieved to guarantee the spectrum sensing performance, and the transmission rate of the CIoT is maximized without considering any interference. The simulations show the predominance of the multicarrier grouping CIoT under the strict interference restriction.