Intelligent Reflecting Surface Assisted Integrated Data and Energy Multicast System in Terahertz-Bands

Abstract

An intelligent reflect surface (IRS) assisted multi-user integrated data and energy multicast (IDEM) system is investigated. Considering terahertz (THz) downlinks, the spatial asymptotical orthogonality is proved. Our goal is to maximize the minimum achievable rate of data users (DUs) and satisfy the energy harvesting requirements of energy users (EUs) by jointly designing the multicast beam at the transmitter and the reflecting beam at the IRS. Firstly, the solutions obtained by the semidefinite relaxation (SDR) are analysed. Based on the orthogonality, a low-complexity beamforming scheme is proposed. Specifically, the asymptotically optimal analog beamforming structures of IRS and transmitter are proved, respectively. The transmit power is allocated at the transmitter to satisfy the EUs and the asymptotically optimal beamforming structure of the IRS is used to compensate the path-loss of DUs. Then the transmitter and IRS beamformings are obtained in closed form. Finally, a randomization scheme for IRS beamforming is proposed to further improve the IDET performance. The numerical results shows that, the trade-off between wireless data transfer (WDT) and wireless energy transfer (WET) is well achieved. Moreover, the proposed scheme achieves nearly the same spectral efficiency as fully digital transmitter and higher spectral efficiency than the IRS solutions obtained by SDR.