Performance Analysis of IRS-Assisted Terrestrial MIMO Terahertz Communications over Atmospheric Turbulence Channels

The terahertz (THz) band has received much attention as a potentially usable bandwidth for 6G. Near-Earth THz is susceptible to strong atmospheric interference, and direct links are vulnerable to obstacles, limiting the breadth of its applications. Emerging intelligent reflecting surfaces (IRS) with multi-input multi-output (MIMO) techniques can mitigate the effects of signal interference and atmospheric turbulence through non-line-of-sight (NLOS) links. To address these challenges, this paper introduces IRS to enhance the coverage and system performance of THz links, and thoroughly analyses the performance metrics of the IRS-assisted THz communication system. First, the atmospheric perturbation of the MIMO THz system is modelled to analyse the characteristics of the perturbation. Given that the turbulence intensity in THz wave transmission is positively correlated with the distance, M\'alaga and Gamma-Gamma distributions are used to simulate the atmospheric turbulence channels from the transmitter to the IRS and from the IRS to the receiver, respectively. Subsequently, pointing errors are considered for the potential displacement deviation of the receiving aperture and the NLOS link. Based on this, the probability density function of the channel state for the NLOS link is derived, and the equal-gain combination (EGC) and maximum-ratio combination (MRC) reception schemes are examined. The numerical results indicate that appropriate modulation techniques can improve performance, while pointing errors and turbulence effects have a significant impact on performance. Increasing the number of input and output ports also significantly improves system performance for all receiving schemes. Comparative analysis reveals that the MRC scheme offers a higher gain compared to the EGC scheme, typically achieving a 1-5 dB performance gain. Still, this advantage diminishes as the number of input ports increases. Meanwhile, the effects of phase noise and phase shift on the system performance are analysed. As phase noise and phase shift continue to decrease, their impact on system performance will gradually diminish and approach zero. Consequently, the performance effects due to phase shift and phase noise are weaker compared to the differences in system performance caused by variations in system parameters.