Active Intelligent Reflecting Surface for MISO Wireless Powered Communication: NOMA or TDMA?

To enhance the massive connectivity and spectral efficiency of Internet-of-Things applications supporting batteryless devices, we integrate active intelligent reflecting surface (IRS) into multiple input single output (MISO) wireless powered communication networks (WPCNs). The networks are designed with either nonorthogonal multiple access (NOMA) or time division multiple access (TDMA) towards the maximum throughput. For this purpose, we develop alternative optimization (AO) algorithms optimizing energy beamforming (BF) and active IRS BF in wireless power transfer, active IRS BF and receive BF in wireless information transfer, and time allocation iteratively, for both NOMA and TDMA networks. The proposed algorithm for the NOMA network solves each subproblem using either a convex optimization solver or a closed-form expression with the help of with the help of fractional programming. For TDMA networks, we develop a more computationally efficient algorithm than the conventional algorithm for the TDMA network. The results show that active IRS is more effective than passive IRS with a moderate number of IRS elements for MISO WPCNs. In addition, NOMA significantly outperforms TDMA for active IRS aided MISO WPCN although a slight gain of TDMA over NOMA is observed for the single-antenna counterpart.