Cognitive Backscatter NOMA Networks With Multi-Slot Energy Causality

In this letter, we consider a multi-user cognitive backscatter network. Backscatter devices (BDs) not only share the spectrum with the primary user, but also harvest energy from and transmit their own information over the primary signal. The multi-BD system adopts non-orthogonal multiple access (NOMA) to further improve the spectral efficiency. Moreover, BDs involved are assumed to have energy storage capability so that the energy harvested in the current block can be stored for future use. Under this assumption, a multi-slot energy causality constraint is proposed. The transmit power of the primary user and the reflection coefficients of BDs are jointly optimized to maximize the sum-rate of the secondary system over multiple slots. Based on the concave-convex procedure (CCCP), we obtain a near-optimal solution by iteratively maximizing the sum-rate. Besides, the effect of imperfect SIC on our proposed resource allocation schemes is also studied. Numerical results show that the multi-slot operation can achieve a remarkable performance gain over the single-slot operation.