Channel Estimation for Intelligent Reflecting Surface Enabled Terahertz MIMO Systems

With the development of sixth generation (6G) wireless systems, terahertz (THz) communication has been envisioned as an emerging technology pillar to provide large bandwidth and support diverse application scenarios. However, due to the severe path attenuation and poor diffraction of THz waves, THz communication links are easily interrupted by the obstacles when it is applied to indoor scenarios. To tackle this challenge, an intelligent reflecting surface (IRS), which is able to control the propagation direction of THz waves by adjusting the discrete phase shifts of IRS elements, is considered as an available alternate to mitigate blockage vulnerability and enhance the coverage capability. To begin with, the hardware characteristics of graphene-enabled IRS is investigated and the IRS-assisted THz multiple-input multiple-output (MIMO) system model is developed. Then, a low complexity compressed sensing (CS) based channel estimation scheme, namely iterative atom pruning based subspace pursuit (IAP-SP), is proposed for channel state information (CSI) acquisition. Concretely, the IAP-SP scheme reduces the computational burden by eliminating the redundant columns of sensing matrix during the iterative process. Simulation results demonstrate that, in contrast with conventional subspace pursuit (C-SP) scheme, the developed IAP-SP maintains basically consistent channel recovery performance while realizes extra 99.51% complexity reduction.