Pith. sign in

REVIEW

Angularly Sparse Channel Estimation in Dual-Wideband Tera-Hertz (THz) Hybrid MIMO Systems Relying on Bayesian Learning

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2402.12158 v1 pith:5RJMNTJI submitted 2024-02-19 eess.SP

Angularly Sparse Channel Estimation in Dual-Wideband Tera-Hertz (THz) Hybrid MIMO Systems Relying on Bayesian Learning

classification eess.SP
keywords estimationbayesianchanneldual-widebandlearningmimosystemsantenna
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X LinkedIn Reddit HN
read the original abstract

Bayesian learning aided massive antenna array based THz MIMO systems are designed for spatial-wideband and frequency-wideband scenarios, collectively termed as the dual-wideband channels. Essentially, numerous antenna modules of the THz system result in a significant delay in the transmission/ reception of signals in the time-domain across the antennas, which leads to spatial-selectivity. As a further phenomenon, the wide bandwidth of THz communication results in substantial variation of the effective angle of arrival/ departure (AoA/ AoD) with respect to the subcarrier frequency. This is termed as the beam squint effect, which renders the channel state information (CSI) estimation challenging in such systems. To address this problem, initially, a pilot-aided (PA) Bayesian learning (PA-BL) framework is derived for the estimation of the Terahertz (THz) MIMO channel that relies exclusively on the pilot beams transmitted. Since the framework designed can successfully operate in an ill-posed model, it can verifiably lead to reduced pilot transmissions in comparison to conventional methodologies. The above paradigm is subsequently extended to additionally incorporate data symbols to derive a Data-Aided (DA) BL approach that performs joint data detection and CSI estimation. We will demonstrate that it is capable of improving the dual-wideband channels estimate, despite further reducing the training overhead. The Bayesian Cramer-Rao bounds (BCRLBs) are also obtained for explicitly characterizing the lower bounds on the mean squared error (MSE) of the PA-BL and DA-BL frameworks. Our simulation results show the improved normalized MSE (NMSE) and bit-error rate (BER) performance of the proposed estimation schemes and confirm that they approach their respective BCRLB benchmarks.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.