A comprehensive review of ensemble based and emerging techniques for DDoS detection: taxonomy, challenges, and research roadmap
DYNA · 01.11.2026 · 101 (6) · 614-620
DOI: 10.52152/d11786 ↗Kurz gefasst
Fachartikel in der Zeitschrift DYNA, online erschienen am 01.11.2026. Verfasst von Sultan Shutyan S Albalawi und 2 weiteren Autor:innen.
Abstract (Originalsprache)
DDoS attacks are becoming more advanced and volumetric in current network scenarios, which consist of software-defined networking (SDN), Internet of Things (IoT), cloud and edge infrastructures. Ensemble learning approaches are being utilized with increased regularity to improve precision, harden systems against attacks, and dynamically adjust to the evolving cybersecurity landscape. This systematic literature review (SLR) focuses on analyzing the research conducted on ensemble learning-based DDoS detection in the period from 2021 to 2025. Eighty-nine peer-reviewed research studies were selected and analyzed after going through a meticulous review process. Publication trends, deployment scenarios, datasets, ensemble models, performance evaluations, challenges and future directions are discussed in this review. The current paper aims to provide a comparative study of the structural composition of algorithms used for ensemble-based DDoS detection. The categories included are datasets used, feature extraction and evaluation methods. Our classification allows for systematic and rank-based objectives and methodologies to be analyzed for a better understanding of current research. The analysis revealed that SDN and IoT were the most focused-on deployment scenarios and that most current research utilizes augmentation, stacking and hybrid technology. An evidence-based taxonomy is presented, which categorizes all existing ensemble learning-based approaches into six categories: Bagging-Based, Boosting-Based, Stacking-Based, Hybrid, Optimization-Driven, and Emerging Intelligent Ensembles. Other notable challenges include data dependency, scalability, time complexity, model interpretability, privacy preservation and robustness to adversarial and unknown attacks. Finally, this SLR presents an evidence-based research agenda for more adaptive, interpretable, scalable, and privacy-preserving ensemble learning-based DDoS defense systems. Keywords: Distributed Denial of Service, DDoS, Ensemble Learning, Machine Learning, DDoS Detection, Network Security, Intrusion Detection Systems, Cyber Threat Detection, Intelligent Security Systems, Classification Algorithms.
Zitieren
APA
Albalawi, S. S. S., Idris, M. Y. I., & Wahab, A. W. B. A. (2026). A comprehensive review of ensemble based and emerging techniques for DDoS detection: taxonomy, challenges, and research roadmap. DYNA, 101(6), 614-620. https://doi.org/10.52152/d11786
Vancouver
Albalawi SSS, Idris MYI, Wahab AWBA. A comprehensive review of ensemble based and emerging techniques for DDoS detection: taxonomy, challenges, and research roadmap. DYNA. 2026;101(6):614-620. doi:10.52152/d11786
BibTeX
@article{Albalawi2026,
author = {Albalawi, Sultan Shutyan S and Idris, Mohd Yamani Idna and Wahab, Ainuddin Wahid Bin Abdul},
title = {A comprehensive review of ensemble based and emerging techniques for DDoS detection: taxonomy, challenges, and research roadmap},
journal = {DYNA},
year = {2026},
volume = {101},
number = {6},
pages = {614--620},
doi = {10.52152/d11786}
}
EUVIMED