Taylor Expansion Aided Gradient Descent Schemes for IRS-Enabled Terahertz MIMO Systems

Terahertz (THz) communication has been regarded as an emerging technology for diverse sixth generation (6G) application scenarios, which possesses large bandwidth and provides ultra-high data transmission up to several terabit per second (Tbps). With regard to the severe path attenuation of THz waves, the short-distance indoor scenario is an appropriate choice for THz communication. However, the THz signals are readily blocked by the indoor obstacles, leading to a communication interruption. To tackle this issue, an intelligent reflecting surface (IRS) is proposed to control the propagation direction of THz beam and enhance the coverage performance by smartly adjusting the phase shifts of reflecting elements. For sake of a desired phase shift combination of IRS elements in the IRS assisted THz multiple-input multiple- output (MIMO) system, an efficient Taylor expansion aided gradient descent (TE-GD) scheme is developed by dynamically updating the step size during the iterative process, where the step size is determined by the coefficients of second-order Taylor expansion formulation. Simulation results shows that, in contrast with the exhaustive search method, the developed TE-GD scheme realizes the basically consistent spectral efficiency with ignorable complexity. Moreover, compared with conventional gradient descent (C-GD) with fixed step size, the developed TE-GD scheme achieves 4.3 bps/Hz spectral efficiency improvement and 80.17% complexity reduction.