Hybrid SOMP-MUSIC-Based Channel Estimation Scheme for Terahertz Massive MIMO-OFDM Systems

The channel estimation procedure is quite challenging in Terahertz (THz)-band communication systems due to the beam-split effect. This occurrence emerges in ultra-broadband THz signals that have large fractional bandwidths. In such a scenario, the components of the THz path at various subcarriers split into diverse spatial directions with the consequence of array gain loss. This results in a beam split. Most of the previous channel estimation techniques documented in the literature neglect the effect of the beam split which is not realistic in the real-time scenario. In this paper, a compressive sensing technique that considers the beam split effect is proposed for channel estimation in THz Massive Multiple Input Multiple Output (MIMO) Orthogonal Frequency Division Multiplexing (OFDM) system. The proposed estimator is based on the hybrid of simultaneous orthogonal matching pursuit (SOMP) and MUltiple SIgnal Classification (MUSIC) algorithms and is named the hybrid (H) SOMP-MUSIC channel estimator. It employs a time-delay network to mitigate the effect of beam split. By using the inherent sparsity in the THz channel, the proposed HSOMP-MUSIC estimator can estimate the channel with few pilots than the traditional OMP and SOMP techniques. In comparison with another method that incorporated the beam split effect into the estimation procedure, and two other methods that neglect the effect, the proposed HSOMP-MUSIC-based estimator exhibits better achievable performance.