Near-Field Channel Estimation for Extremely Large-Scale Array Communications: A Model-Based Deep Learning Approach

Zhang, Xiangyu; Wang, Zening; Zhang, Haiyang; Yang, Luxi

doi:10.1109/lcomm.2023.3245084

Cited by 21 publications

(13 citation statements)

References 41 publications

(44 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…3), which depends on both direction and range information for accurate signal processing [3]. Among the works investigating the near-field signal model, [24][25][26][27][28] consider the near-field scenario while the effect of beam-split is ignored and only mm-Wave scenario is investigated. In particular, [27] considers the wideband mm-Wave channel estimation while the authors in [28] devise a machine learning (ML)-based approach for narrowband near-field channel estimation.…”

Section: A Related Workmentioning

confidence: 99%

“…Among the works investigating the near-field signal model, [24][25][26][27][28] consider the near-field scenario while the effect of beam-split is ignored and only mm-Wave scenario is investigated. In particular, [27] considers the wideband mm-Wave channel estimation while the authors in [28] devise a machine learning (ML)-based approach for narrowband near-field channel estimation. In addition to near-field-only model, hybrid (near-and far-field) models are also present in the literature [24,26], wherein only narrowband transceiver architectures are considered.…”

Section: A Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Near-Field Terahertz Communications: Model-Based and Model-Free Channel Estimation

Elbir¹,

Shi²,

Papazafeiropoulos³

et al. 2023

Preprint

View full text Add to dashboard Cite

Terahertz (THz) band is expected to be one of the key enabling technologies of the sixth generation (6G) wireless networks because of its abundant available bandwidth and very narrow beam width. Due to high frequency operations, electrically small array apertures are employed, and the signal wavefront becomes spherical in the near-field. Therefore, near-field signal model should be considered for channel acquisition in THz systems. Unlike prior works which mostly ignore the impact of near-field beam-split (NB) and consider either narrowband scenario or far-field models, this paper introduces both a modelbased and a model-free techniques for wideband THz channel estimation in the presence of NB. The model-based approach is based on orthogonal matching pursuit (OMP) algorithm, for which we design an NB-aware dictionary. The key idea is to exploit the angular and range deviations due to the NB. We then employ the OMP algorithm, which accounts for the deviations thereby ipso facto mitigating the effect of NB. We further introduce a federated learning (FL)-based approach as a model-free solution for channel estimation in a multi-user scenario to achieve reduced complexity and training overhead.Through numerical simulations, we demonstrate the effectiveness of the proposed channel estimation techniques for wideband THz systems in comparison with the existing state-of-the-art techniques.

show abstract

Section: A Related Workmentioning

confidence: 99%

Section: A Related Workmentioning

confidence: 99%

Near-Field Terahertz Communications: Model-Based and Model-Free Channel Estimation

Elbir¹,

Shi²,

Papazafeiropoulos³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…3), which depends on both direction and range information for accurate signal processing [4]. Among the works investigating the near-field signal model, [25][26][27][28][29] consider the near-field scenario while the effect of beam-squint is ignored and only mm-Wave scenario is investigated. In particular, [28] considers the wideband mm-Wave channel estimation while the authors in [29] devise a machine learning (ML)-based approach for narrowband near-field channel estimation.…”

Section: A Related Workmentioning

confidence: 99%

“…Among the works investigating the near-field signal model, [25][26][27][28][29] consider the near-field scenario while the effect of beam-squint is ignored and only mm-Wave scenario is investigated. In particular, [28] considers the wideband mm-Wave channel estimation while the authors in [29] devise a machine learning (ML)-based approach for narrowband near-field channel estimation. In addition to nearfield-only model, hybrid (near-and far-field) models are also present in the literature [25,27], wherein only narrowband transceiver architectures are considered.…”

Section: A Related Workmentioning

confidence: 99%

Near-Field Terahertz Communications: Model-Based and Model-Free Channel Estimation

et al. 2023

View full text Add to dashboard Cite

Terahertz (THz) band is expected to be one of the key enabling technologies of the sixth generation (6G) wireless networks because of its abundant available bandwidth and very narrow beamwidth. Due to high frequency operations, electrically small array apertures are employed, and the signal wavefront becomes spherical in the near-field. Therefore, near-field signal model should be considered for channel acquisition in THz systems. Unlike prior works which mostly ignore the impact of near-field beam-squint (NB) and consider either narrowband scenario or far-field models, this paper introduces both a model-based and a model-free techniques for wideband THz channel estimation in the presence of NB. The model-based approach is based on orthogonal matching pursuit (OMP) algorithm, for which we design an NB-aware dictionary. The key idea is to exploit the angular and range deviations due to the NB. We then employ the OMP algorithm, which accounts for the deviations thereby ipso facto mitigating the effect of NB. We further introduce a federated learning (FL)-based approach as a model-free solution for channel estimation in a multi-user scenario to achieve reduced complexity and training overhead. Through numerical simulations, we demonstrate the effectiveness of the proposed channel estimation techniques for wideband THz systems in comparison with the existing state-of-the-art techniques.INDEX TERMS Beamsquint, channel estimation, federated learning, machine learning, near-field, orthogonal matching pursuit, sparse recovery, terahertz.

show abstract

“…In a distinct approach, a model-based deep learning framework for near-field channel estimation of XL-MIMO systems was showcased in [255]. The spherical-wave propagation channel model was applied in building a spatial gridding based sparse dictionary, such that the channel estimation problem was attributed to a CS problem that was solved using the learning iterative shrinkage and thresholding algorithm (LISTA).…”

Section: A Hmimo Channel Estimationmentioning

confidence: 99%

Holographic MIMO Communications: Theoretical Foundations, Enabling Technologies, and Future Directions

Gong¹,

Gavriilidis²,

Ji³

et al. 2023

Preprint

View full text Add to dashboard Cite

<p>Future wireless systems are envisioned to create an endogenously holography-capable, intelligent, and programmable radio propagation environment, that will offer unprecedented capabilities for high spectral and energy efficiency, low latency, and massive connectivity. A potential and promising technology for supporting the expected extreme requirements of the sixth-generation (6G) communication systems is the concept of the holographic multiple-input multiple-output (HMIMO), which will actualize holographic radios with reasonable power consumption and fabrication cost. The HMIMO is facilitated by ultra-thin, extremely large, and nearly continuous surfaces that incorporate reconfigurable and sub-wavelength-spaced antennas and/or metamaterials. Such surfaces comprising dense electromagnetic (EM) excited elements are capable of recording and manipulating impinging fields with utmost flexibility and precision, as well as with reduced cost and power consumption, thereby shaping arbitrary-intended EM waves with high energy efficiency. The powerful EM processing capability of HMIMO opens up the possibility of wireless communications of holographic imaging level, paving the way for signal processing techniques realized in the EM-domain, possibly in conjunction with their digital-domain counterparts. However, in spite of the significant potential, the studies on HMIMO communications are still at an initial stage, its fundamental limits remain to be unveiled, and a certain number of critical technical challenges need to be addressed. In this survey, we present a comprehensive overview of the latest advances in the HMIMO communications paradigm, with a special focus on their physical aspects, their theoretical foundations, as well as the enabling technologies for HMIMO systems. We also compare the HMIMO with existing multi-antenna technologies, especially the massive MIMO, present various promising synergies of HMIMO with current and future candidate technologies, and provide an extensive list of research challenges and open directions for future HMIMO-empowered wireless applications.</p>

show abstract

Near-Field Channel Estimation for Extremely Large-Scale Array Communications: A Model-Based Deep Learning Approach

Cited by 21 publications

References 41 publications

Near-Field Terahertz Communications: Model-Based and Model-Free Channel Estimation

Near-Field Terahertz Communications: Model-Based and Model-Free Channel Estimation

Near-Field Terahertz Communications: Model-Based and Model-Free Channel Estimation

Holographic MIMO Communications: Theoretical Foundations, Enabling Technologies, and Future Directions

Contact Info

Product

Resources

About