Hybrid Spherical- and Planar-Wave Channel Modeling and Estimation for Terahertz Integrated UM-MIMO and IRS Systems

Integrated ultra-massive multiple-input multiple-output (UM-MIMO) and intelligent reflecting surface (IRS) systems are promising for 6G beyond Terahertz (0.1-10 THz) communications, to effectively bypass the barriers of limited coverage line-of-sight blockage. However, excessive dimensions UM-MIMO IRS enlarge near-field region, while strong THz channel sparsity in far-field is detrimental spatial multiplexing. Moreover, estimation (CE) requires recovering large-scale from severely compressed observations due RF-chains. To tackle these challenges, a hybrid spherical- planar-wave model (HSPM) introduced cascaded integrated system. The multiplexing gains under regions analyzed, which found be by segmented with lower rank. Furthermore, compressive sensing-based CE framework developed, including sparse representation method, separate-side (SSE) dictionary-shrinkage (DSE) algorithms. Numerical results verify effectiveness HSPM, capacity only 5 × 10 -4 bits/s/Hz deviated that obtained ground-truth spherical-wave-model, 256 elements. While SSE achieves improved accuracy than benchmark algorithms, DSE more attractive noisy environments, 1 dB normalized-mean-square error SSE.