Unsupervised Machine Learning-Based User Clustering in THz-NOMA Systems - Paper Abstract

Aug 4, 2026

Upload by Eliana

This paper applies unsupervised machine learning clustering to non-orthogonal multiple access (NOMA) assisted terahertz (THz) networks, using K-Means, agglomerative hierarchical clustering (AHC), and DBSCAN. The primary goal is to cluster secondary users without requiring prior knowledge of the number of clusters, while avoiding degradation of primary-user performance. Results indicate that AHC and DBSCAN provide higher system throughput and connectivity than K-Means, where cluster count is selected adaptively and automatically. The study evaluates performance versus complexity for THz-NOMA operation.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages6

Reading Time15 min

AccessFree to download

Abstract

Introduction

System Model and Problem Formulation

Spatial and Beamforming Model

The University of Manchester Research Unsupervised Machine Learning-Based User Clustering in THz-NOMA Systems DOI: 10.1109/LWC.2023.3262788 Document Version Accepted author manuscript Link to publication record in Manchester Research Explorer Citation for published version (APA): Lin, Y. , Wang, K. , & Ding, Z. (2023) . Unsupervised Machine Learning-Based User Clustering in THz-NOMA Systems. IEEE Wireless Communications Letters, 12(7), 1130-1134 . [https://doi.org/10.1109/LWC.2023.3262788](https://doi.org/10.1109/LWC.2023.3262788) Published in: IEEE Wireless Communications Letters Citing this paper Please note that where the full-text provided on Manchester Research Explorer is the Author Accepted Manuscript or Proof version this may differ from the final Published version. If citing, it is advised that you check and use the publisher's definitive version. General rights Copyright and moral rights for the publications made accessible in the Research Explorer are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. Takedown policy If you believe that this document breaches copyright please refer to the University of Manchester’s Takedown Procedures [[http://man.ac.uk/04Y6Bo](http://man.ac.uk/04Y6Bo)] or [contact openresearch@manchester.ac.uk](contact openresearch@manchester.ac.uk) providing relevant details, so we can investigate your claim. Download date:03 . Aug. 2026 Unsupervised Machine Learning-Based User Clustering in THz-NOMA Systems Yushen Lin, Student Member, IEEE, Kaidi Wang, Member, IEEE, Zhiguo Ding, Fellow, IEEE Abstract—In this paper, different unsupervised machine learning (ML)-based user clustering algorithms, including K-Means, agglomerative hierarchical clustering (AHC), and density-based spatial clustering of applications with noise (DBSCAN) are applied in non-orthogonal multiple access (NOMA) assisted terahertz (THz) networks. The key contribution of this paper is to design ML-based approaches to ensure that the secondary users can be clustered without knowing the number of clusters and degrading the performance of the primary users. The studies carried out in the paper show that the proposed schemes based on AHC and DBSCAN can achieve superior performance on system throughput and connectivity compared to the traditional clustering strategy, i.e., K-means, where the number of clusters is determined in an adaptive and automatic manner. Index Terms—Machine learning (ML), non-orthogonal multiple access (NOMA), and user clustering. I. INTRODUCTION With the fifth-generation (5G) communication successfully standardized and deployed globally, the research for sixthgeneration (6G) wireless communication recently attracts significant attention as the current 5G communication networks can no longer meet all the requirements of tremendous traffic growth envisioned in the future wireless networks, i.e., ultra-high throughput, ultra-massive connectivity, and ultralow latency, etc [1] . Non-orthogonal multiple access (NOMA) combined with beamforming (BF) and terahertz (THz) technologies is seen as a promising solution to meet the growing demand. The THz spectrum offers a vast amount of bandwidth for communication, and NOMA can boost spectral efficiency by enabling multiple users to share the same bandwidth resources [2], [3] . This paper differs from existing works on THz-NOMA [4]–[7] in that it focuses on the application of machine learning(ML)-based clustering techniques for legacy THzNOMA systems, where beams have been pre-configured to serve primary users. The main contributions of this work include: i) investigating how existing beams can be utilized to serve secondary users by applying unsupervised ML-based user clustering algorithms, specifically agglomerative hierarchical clustering (AHC) and density-based spatial clustering of app

FAQs

What clustering algorithms are used for user grouping in THz-NOMA networks?

The paper evaluates K-Means, agglomerative hierarchical clustering (AHC), and DBSCAN for unsupervised user clustering in THz-NOMA systems.

How does the proposed approach handle the unknown number of clusters?

It designs unsupervised ML-based clustering schemes so secondary users can be grouped without knowing the number of clusters in advance.

Which methods show better performance than K-Means and what metrics improve?

AHC and DBSCAN outperform the traditional K-Means strategy, achieving superior system throughput and connectivity while maintaining primary-user performance.

Aug 4, 2026

Upload by Eliana