GNN-based Online Beamforming Design for HAPS-Assisted NTN

Lavanya S S Anjapuli 1 , Animesh Yadav 1 , and Halim Yanikomeroglu 2 · 2026

Read GNN-based Online Beamforming Design for HAPS-Assisted NTN by Lavanya S S Anjapuli 1 , Animesh Yadav 1 , and Halim Yanikomeroglu 2 (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-00244.txt.

Introduction

Anjapuli, Yadav, and Yanikomeroglu add a high sky relay to help users at a cell's edge. Edge users in towns face weak links from loss, shade, and clash with next cells. A high-altitude platform can catch their data and pass it on. Clear sight lines from that height make the hop strong.

Ground posts send edge data up to the sky post for relay. The team frames a beam-shape task that lifts energy thrift at both posts. The task is bent in shape, so they build a graph net to solve it live. The net reads the web's form as part of its math. Sims show clear gains for edge users. The worst-off fifth of users gain most in thrift terms.

Catalog record

CREATORLavanya S S Anjapuli 1 , Animesh Yadav 1 , and Halim Yanikomeroglu 2
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE37 KB
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2606.00244

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

GNN-based Online Beamforming Design for HAPS-Assisted NTN Lavanya S S Anjapuli 1 , Animesh Yadav 1 , and Halim Yanikomeroglu 2 Abstract In terrestrial networks, especially in urban areas, cell-edge users often face significant capacity limitations due to high path loss, shadowing, and inter-cell interference (ICI). This paper proposes integrating a high-altitude platform station (HAPS) into terrestrial networks, where terrestrial base stations (BS) can alleviate these issues by relaying data intended for cell-edge users via HAPS, thereby leveraging line-of-sight (LoS) links. We formulate an energy-efficiency (EE) maximization problem to jointly design beamforming vectors at the BS and HAPS with the goal of improving cell-edge user performance. Since the resulting problem is non-convex, we develop an online optimization framework based on a graph neural networks (GNN), which effectively captures the network topology. Numerical results show that the proposed HAPS-assisted architecture improves network performance, particularly by increasing the 5th-percentile EE, thereby enhancing service for cell-edge users. I Introduction Multi-cell network architectures form the backbone of cellular wireless systems and are continued to remain fundamental in beyond fifth-generation (5G) and sixth-generation (6G) networks to support dense connectivity, high data rates, and wide-area coverage through massive multiple input multiple output (MIMO), advanced beamforming, and dynamic interference management [ 1 ] . However, inter-cell interference, shadowing, and non-line-of-sight (NLoS) propagation significantly degrade the performance of cell-edge users, limiting spectral and EE [ 6 ] . HAPS have therefore emerged as a promising solution to these challenges by offering large-scale coverage with a high probability of line-of-sight (LoS) air-to-ground links, thereby effectively mitigating shadowing effects [ 12 ] . Furthermore, due to the presence of LoS air-to-ground links and the availability of wide bandwidth, millimeter-wave (mmWave) communication is well suited for HAPS-assisted networks. The LoS links help mmWave signals compensating severe path loss but more can be achieved by employing large antenna elements to achieve beamforming gain [ 20 ] . Implementing fully digital beamforming with large antenna arrays is often impractical, as it requires one radio-frequency (RF) chain per antenna element, resulting in high hardware cost, power consumption, and baseband processing complexity [ 17 ] . Hybrid beamforming offers a practical alternative by combining analog RF beam steering with low-dimensional digital precoding, thereby providing a favorable trade-off between performance and hardware efficiency. Nevertheless, hybrid architectures introduce additional challenges in beamformer optimization under constant-modulus constraints and dynamic resource allocation, motivating the development of advanced design methodologies [ 24 , 2 ] . Iterative optimization-based algorithms for hybrid beamforming design include alternating optimization methods [ 29 , 18 ] , adaptive cross-entropy (ACE) approaches [ 8 ] , and stochastic beamforming techniques [ 5 ] . Recently, learning-based methods have gained traction for their ability to reduce the computational burden of iterative optimization [ 25 , 3 , 30 , 23 ] . For instance, the deep neural network (DNN)-based hybrid beamforming framework in [ 25 ] reduces training complexity in multi-user mmWave multiple-input multiple-outpur (MIMO) systems. [ 3 ] develops a convolutional neural networks (CNN) architecture through supervised training to jointly learn analog and digital beamforming weights for improved energy efficiency in mmWave massive MIMO. Similarly, [ 30 ] presents a CNN model that optimizes hybrid beamforming weights in a multiple-input single-output (MISO) system using an signal-to-noise ratio (SNR) as a loss function. Lastly, [ 23 ] proposes an end-to-end CNN framework that integrates channel state information (CSI) feedback and hybrid beamforming weight design for mmWave MIMO systems, aimed at maximizing spectral efficiency. While the aforementioned learning-based approaches significantly reduce computational complexity, they typically operate under fixed input–output dimensions and lack explicit modeling of network topology and interference relationships. As a result, their ability to generalize is limited when the number of users or base stations changes. Moreover, practical wireless networks experience time-varying channels due to dynamic propagation environments, which can lead to mismatches between training data and real-time channel distributions. Online optimization frameworks mitigate this issue by continuously updating model parameters using real-time channel observations [ 4 ] . Motivated by these insights, we propose an online graph neural networks (GNN)-based hybrid beamforming framework for energy-efficient HAPS-assisted multi-cell NOMA networks. GNN address these limitations by modeling wireless systems as graphs and explicitly capturing interference relationships, enabling scalable and topology-aware optimization [ 22 ] . The contributions of this paper are summarized as follow: • A GNN-based online optimization technique is developed to jointly optimize hybrid beamforming weights at both terrestrial BS and HAPS by maximizing the EE of the network. • Comprehensive performance evaluations is provided by comparing the proposed approach with DNN, CNN and ACE-based techniques demonstrating the superiority of the proposed technique. The remainder of this paper is organized as follows: Section II describes the system model, while Section III formulates the EE optimization problem. The solution to this problem is presented in Section IV. Section V outlines the simulation setup and discusses the results, and Section VI concludes the paper. Fig. 1 : System architecture of the network considered. II System Model In this paper, we consider a multi-cell network in which each cell is served by a BS equipped with a uniform two-dimensional planar array consisting of M_{b}=M_{bh}M_{bv} antenna elements where M_{bh} and M_{bv} are the number of elements along the horizontal x -axis and vertical z -axis, respectively. A full-duplex (FD) high-altitude platform station (HAPS), capable of simultaneous transmission and reception over the same frequency band, is deployed to support cell-edge UE that experience severe inter-cell interference, while near UE are directly served by their associated BS. Specifically, the FD-HAPS simultaneously receives signals from multiple BS and transmits data to edge UE over the same time–frequency resources. The HAPS is also equipped with a UPA comprising M_{h}=M_{hx}M_{hy} antenna elements arranged along the horizontal x and y axes.To enhance spectral efficiency, power-domain non-orthogonal multiple access (NOMA) is employed at the transmitting nodes to multiplex near and edge UE signals over the same time–frequency resources, while SIC is applied at the receivers to decode the superimposed signals.The entire architecture is illustrated in Fig. 1 . All terrestrial links in the considered multi-cell network operate in the millimeter-wave (mmWave) frequency band, typically in the range of 24–100 GHz, to exploit the large available bandwidth for high data rate transmission [ 1 , 20 , 7 ] . In this work, we consider 28 GHz as the operating frequency for all transmissions. II-A Channel Model Because of the use of mmWave band frequencies, which undergo limited scattering, all the terrestrial links, such as BS-to-UE, are modeled using a sparse geometric multipath framework [ 19 ] . Considering S dominant scatterers, the channel from BS b to UE u can be represented as \displaystyle\mathbf{h}_{b,b_{u}}=\sqrt{\frac{1}{S}}\sum_{k=1}^{S}g^{k}_{b,b_{u}}\mathbf{r}_{t}(\theta^{D,k}_{b,b_{u}},\varphi^{D,k}_{b,b_{u}})\sqrt{G_{b}\big(\theta^{D,k}_{b,b_{u}},\varphi^{D,k}_{b,u}\big)}, (1) where g^{k}_{b,b_{u}} represents the complex channel gain of the path k , which also includes path loss, PL = \left(\frac{4\pi df_{c}}{c}\right)^{2} with f_{c} is the carrier frequency, c=3\times 10^{8}\ \text{m/s} is the speed of light in free space and d is the distance from BS to UE [ 16 ] . The angles \theta^{D,k}_{b,u} , \theta\in[-\pi/2,\pi/2] and \varphi^{D,k}_{b,u} , \varphi\in[0,2\pi] are the elevation and azimuth angles of departure, respectively. The term G_{b}\big(\theta^{D,k}_{b,b_{u}},\varphi^{D,k}_{b,b_{u}}\big) represents the antenna gain of BS b and \mathbf{r}_{t}(\theta^{D,k}_{b,u},\varphi^{D,k}_{b,u}) , for the path k , represents the transmit

Read the complete text