Throughput Maximization for NOMA-Based Cognitive Backscatter Communication Networks With Imperfect CSI

Cognitive radio and backscatter communication (BackCom) have been viewed as two promising technologies for the future green Internet of Things (IoT). The combination of these two technologies can not only enhance spectrum efficiency but also improve energy efficiency. However, most of the existing resource allocation (RA) algorithms in cognitive BackCom networks consider ideal channel state information and a time division multiple access protocol, which is unrealistic in practical systems and can not support the massive number of accessing users. To this end, in this article, we study a robust chance-constrained RA problem for nonorthogonal multiple access (NOMA)-based cognitive BackCom networks to overcome the influence of channel estimation errors and support for large-scale IoT nodes. Specifically, cognitive backscatter users (CBUs) can not only share the spectrum resource owned by primary users but also harvest surrounding radio frequency and transmit their own information over the primary signals. Moreover, CBUs can use the harvested energy to actively transmit information via an NOMA protocol. The robust RA problem with outage-probability constraints is formulated to maximize the total throughput of CBUs by jointly optimizing the transmission time, transmit power, and the reflection coefficients of CBUs. To tackle the nonconvex problem, the original problem is converted into an equivalent form by applying the linear objective function, an inequality transformation approach, and an auxiliary variable method. Then, an iteration-based RA algorithm is proposed to solve it. Simulation results demonstrate the effectiveness of the proposed algorithm by comparing it with the benchmark algorithms.