1.
Linfu, Zou; Zhiwen, Pan; El-Hajjar, Mohammed
Graph Neural Network Aided Beamforming for Holographic Millimeter Wave MIMO Systems Journal Article
In: IEEE Transactions on Vehicular Technology, vol. 74, no. 7, pp. 10582–10595, 2025, ISSN: 1939-9359.
Abstract | Links | BibTeX | Tags: Accuracy, Array signal processing, Beamforming, Channel estimation, Downlink, Estimation, graph neural network, Graph neural networks, holographic MIMO, millimeter wave, Millimeter wave communication, OFDM, Optimization, Training
@article{linfu_graph_2025,
title = {Graph Neural Network Aided Beamforming for Holographic Millimeter Wave MIMO Systems},
author = {Zou Linfu and Pan Zhiwen and Mohammed El-Hajjar},
url = {https://ieeexplore.ieee.org/document/10896848},
doi = {10.1109/TVT.2025.3544063},
issn = {1939-9359},
year = {2025},
date = {2025-07-01},
urldate = {2025-10-08},
journal = {IEEE Transactions on Vehicular Technology},
volume = {74},
number = {7},
pages = {10582–10595},
abstract = {Holographic multiple-input multiple-output (HMIMO) systems are considered as one of the potential techniques to meet the demands of next-generation communications by replacing costly and power-hungry devices with sub-half-wavelength antenna elements. However, optimizing the beamforming matrix in the base station (BS) for HMIMO systems is challenging, given the prohibitive overhead of directly estimating the channels between the BS and the user equipment. Instead of following the traditional method of channel estimation and beamforming optimization, in this paper we employ a deep-learning technique to optimize the beamformers at the BS based on a loss function. Specifically, in this paper we introduce a graph neural network (GNN) designed to map the received pilot signals to optimized beamforming matrices and to model interactions among user equipment within the network. The simulation results show that our deep-learning method effectively maximizes the sum-rate objective while using reduced number of pilots than traditional channel estimation and beamforming optimization techniques.},
keywords = {Accuracy, Array signal processing, Beamforming, Channel estimation, Downlink, Estimation, graph neural network, Graph neural networks, holographic MIMO, millimeter wave, Millimeter wave communication, OFDM, Optimization, Training},
pubstate = {published},
tppubtype = {article}
}
Holographic multiple-input multiple-output (HMIMO) systems are considered as one of the potential techniques to meet the demands of next-generation communications by replacing costly and power-hungry devices with sub-half-wavelength antenna elements. However, optimizing the beamforming matrix in the base station (BS) for HMIMO systems is challenging, given the prohibitive overhead of directly estimating the channels between the BS and the user equipment. Instead of following the traditional method of channel estimation and beamforming optimization, in this paper we employ a deep-learning technique to optimize the beamformers at the BS based on a loss function. Specifically, in this paper we introduce a graph neural network (GNN) designed to map the received pilot signals to optimized beamforming matrices and to model interactions among user equipment within the network. The simulation results show that our deep-learning method effectively maximizes the sum-rate objective while using reduced number of pilots than traditional channel estimation and beamforming optimization techniques.
2.
Xu, Chao; Masouros, Christos; Sugiura, Shinya; Petropoulos, Periklis; Maunder, Robert G.; Yang, Lie-Liang; Haas, Harald; Hanzo, Lajos
Integrated Positioning and Communication Relying on Wireless Optical OFDM Journal Article
In: IEEE Journal on Selected Areas in Communications, vol. 43, no. 5, pp. 1721–1737, 2025, ISSN: 1558-0008.
Abstract | Links | BibTeX | Tags: Accuracy, Bandwidth, bi-static, Channel estimation, Estimation, Integrated sensing and communication, ISAC, Light emitting diodes, multipath, NLoS, non-line-of-sight, Nonlinear optics, OFDM, Optical sensors, orthogonal frequency-division multiplexing, Radar, Visible Light Communication, visible light positioning, VLC, VLP
@article{xu_integrated_2025,
title = {Integrated Positioning and Communication Relying on Wireless Optical OFDM},
author = {Chao Xu and Christos Masouros and Shinya Sugiura and Periklis Petropoulos and Robert G. Maunder and Lie-Liang Yang and Harald Haas and Lajos Hanzo},
url = {https://ieeexplore.ieee.org/abstract/document/10900727},
doi = {10.1109/JSAC.2025.3543532},
issn = {1558-0008},
year = {2025},
date = {2025-05-01},
urldate = {2025-10-08},
journal = {IEEE Journal on Selected Areas in Communications},
volume = {43},
number = {5},
pages = {1721–1737},
abstract = {Visible Light Positioning and Communication (VLPC) is a promising candidate for implementing Integrated Sensing And Communication (ISAC) in the unlicensed 400 THz to 800 THz band. The current Visible Light Positioning (VLP) systems mainly operate based on the Received Signal Strength (RSS) of the Line-of-Sight (LoS) path. However, its accuracy is degraded by interferences from Non-LoS (NLoS) paths. Furthermore, in Visible Light Communication (VLC) systems, the estimation of Channel State Information (CSI) also becomes challenging, when the optical channel becomes dispersive. Against this background, we propose a new VLPC scheme using Direct Current (DC) biased Optical Orthogonal Frequency-Division Multiplexing (VLPC-DCO-OFDM), where OFDM-based sensing is applied for the sake of improving the resolution of the estimated Channel Impulse Response (CIRs) exploited for positioning functionality. The CIRs estimated by sensing are further exploited to provide enhanced CSI for communication data detection. Moreover, we propose a hybrid Radar-RSS based solution, where the conventional RSS-aided VLP method is invoked for the sake of refining OFDM radar. Our simulation results demonstrate that the proposed VLPC-DCO-OFDM scheme – which simultaneously supports the triple functionalities of illumination, bi-static sensing and communication – is capable of achieving centimeter-level positioning accuracy and Giga-bits-per-second data rate.},
keywords = {Accuracy, Bandwidth, bi-static, Channel estimation, Estimation, Integrated sensing and communication, ISAC, Light emitting diodes, multipath, NLoS, non-line-of-sight, Nonlinear optics, OFDM, Optical sensors, orthogonal frequency-division multiplexing, Radar, Visible Light Communication, visible light positioning, VLC, VLP},
pubstate = {published},
tppubtype = {article}
}
Visible Light Positioning and Communication (VLPC) is a promising candidate for implementing Integrated Sensing And Communication (ISAC) in the unlicensed 400 THz to 800 THz band. The current Visible Light Positioning (VLP) systems mainly operate based on the Received Signal Strength (RSS) of the Line-of-Sight (LoS) path. However, its accuracy is degraded by interferences from Non-LoS (NLoS) paths. Furthermore, in Visible Light Communication (VLC) systems, the estimation of Channel State Information (CSI) also becomes challenging, when the optical channel becomes dispersive. Against this background, we propose a new VLPC scheme using Direct Current (DC) biased Optical Orthogonal Frequency-Division Multiplexing (VLPC-DCO-OFDM), where OFDM-based sensing is applied for the sake of improving the resolution of the estimated Channel Impulse Response (CIRs) exploited for positioning functionality. The CIRs estimated by sensing are further exploited to provide enhanced CSI for communication data detection. Moreover, we propose a hybrid Radar-RSS based solution, where the conventional RSS-aided VLP method is invoked for the sake of refining OFDM radar. Our simulation results demonstrate that the proposed VLPC-DCO-OFDM scheme – which simultaneously supports the triple functionalities of illumination, bi-static sensing and communication – is capable of achieving centimeter-level positioning accuracy and Giga-bits-per-second data rate.