Explainable Intrusion Detection for Cyber Defences in the Internet of Things: Opportunities and Solutions
The DOI matches its Crossref record; 191 of the 192 references it cites were found in a registry or an index, and 1 are not checked yet.
Abstract
The field of Explainable Artificial Intelligence (XAI) has garnered considerable research attention in recent years, aiming to provide interpretability and confidence to the inner workings of state-of-the-art deep learning models. However, XAI-enhanced cybersecurity measures in the Internet of Things (IoT) and its sub-domains, require further investigation to provide effective discovery of attack surfaces, their corresponding vectors, and interpretable justification of model outputs. Cyber defence involves operations conducted in the cybersecurity field supporting mission objectives to identify and prevent cyberattacks using various tools and techniques, including intrusion detection systems (IDS), threat intelligence and hunting, and intrusion prevention. In cyber defence, especially anomaly-based IDS, the emerging applications of deep learning models require the interpretation of the models’ architecture and the explanation of models’ prediction to examine how cyberattacks would occur. This paper presents a comprehensive review of XAI techniques for anomaly-based intrusion detection in IoT networks. Firstly, we review IDSs focusing on anomaly-based detection techniques in IoT and how XAI models can augment them to provide trust and confidence in their detections. Secondly, we review AI models, including machine learning (ML) and deep learning (DL), for anomaly detection applications and IoT ecosystems. Moreover, we discuss DL’s ability to effectively learn from large-scale IoT datasets, accomplishing high performances in discovering and interpreting security events. Thirdly, we demonstrate recent research on the intersection of XAI, anomaly-based IDS and IoT. Finally, we discuss the current challenges and solutions of XAI for security applications in the cyber defence perspective of IoT networks, revealing future research directions. By analysing our findings, new cybersecurity applications that require XAI models emerge, assisting decision-makers in understanding and explaining security events in compromised IoT networks.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“By integrating XAI techniques, these models can highlight which features such as unusual packet sizes, protocol usage, or temporal patterns triggered an anomaly flag, thereby aiding analysts in decision-making and improving system resilience (Moustafa et al., 2023).”
From Explainable AI: enhancing decision-making in the detection of cyber threats (2026) · cites this paper as Moustafa et al., 2023
In the citing paper
“In cybersecurity practice, XAI has become critical not only for intrusion detection systems, where it helps reduce false positives and improve analyst trust, but also in IoT and IIoT networks, where it highlights abnormal device behavior, and in forensic and compliance contexts, where transparency is necessary to meet regulatory demands such as the EU AI Act (Nwakanma et al., 2023;Moustafa et al.,”
From Explainable AI: enhancing decision-making in the detection of cyber threats (2026) · cites this paper as Moustafa et al., 2023
In the citing paper
“Likewise, Moustafa et al. (2023) stress that this problem is particularly acute in IoTfocused domains, where inconsistent assessment undermines confidence in proposed solutions.”
From Explainable AI: enhancing decision-making in the detection of cyber threats (2026) · cites this paper as Moustafa et al. (2023)
In the citing paper
“Moustafa et al. (2023) explicitly highlight that “most current XAI methods produce results that only make sense to researchers, ” emphasizing that while the academic community may appreciate the novelty of these methods, they remain inaccessible to those working on the frontlines of cybersecurity defence.”
From Explainable AI: enhancing decision-making in the detection of cyber threats (2026) · cites this paper as Moustafa et al. (2023)
In the citing paper
“While partial dependence plots result in a significantly biased computed feature effect, ALEs provide a faster and unbiased substitute for visualisation [138,139].”
From Interpreting Black-Box Models: A Review on Explainable Artificial Intelligence (2024) · cites this paper as [138]
Citations
Cites
Breiman, L. (2001) · Machine Learning · doi:10.1023/a:1010933404324 · Cited by 133,538 · cited 2×
“The Permutation Feature Importance (PFI) was first introduced in 2001 by Breiman [144] and measures the increase in prediction error of a fitted model after permuting the feature’s values.”
BackgroundA Unified Approach to Interpreting Model Predictions
Lundberg, S., & Lee, S.-I. (2017) · arXiv · doi:10.48550/arxiv.1705.07874 · Cited by 23,707
Gauriat, C.-M., Pencolé, Y., Ribot, P., & Brouillet, G. (2024) · HAL (Le Centre pour la Communication Scientifique Directe) · doi:10.4230/oasics.dx.2024.27 · Cited by 12,586
Arrieta, A. B., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., García, S., Gil-López, S., Molina, D., Benjamins, R., et al. (2019) · arXiv · doi:10.48550/arxiv.1910.10045 · Cited by 11,321
Barredo Arrieta, A., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., Garcia, S., Gil-Lopez, S., Molina, D., Benjamins, R., et al. (2020) · Information Fusion · doi:10.1016/j.inffus.2019.12.012 · Cited by 10,423
17. A Value for n-Person Games
Shapley, L. S. (1953) · Contributions to the Theory of Games (AM-28), Volume II · doi:10.1515/9781400881970-018 · Cited by 6,950
“Why Should I Trust You?”: Explaining the Predictions of Any Classifier
Ribeiro, M., Singh, S., & Guestrin, C. (2016) · Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations · doi:10.18653/v1/n16-3020 · Cited by 6,682
A Comprehensive Survey on Transfer Learning
Zhuang, F., Qi, Z., Duan, K., Xi, D., Zhu, Y., Zhu, H., Xiong, H., & He, Q. (2021) · Proceedings of the IEEE · doi:10.1109/jproc.2020.3004555 · Cited by 6,678 · cited 2×
“Semi-supervised learning is helpful in many scenarios when data collecting is expensive, time-consuming or even unrealistic [112].”
BackgroundMethodPeeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI)
Adadi, A., & Berrada, M. (2018) · IEEE Access · doi:10.1109/access.2018.2870052 · Cited by 6,294
Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps
Simonyan, K., Vedaldi, A., & Zisserman, A. (2013) · arXiv · doi:10.48550/arxiv.1312.6034 · Cited by 5,370
Research works printed but not linked: 28
- Morgan, 2020, Cybercrime to Cost the World $10.5 Trillion Annually by 2025
- 2021, Australian cyber security centre annual cyber threat report 2020–21
- Das, 2018, Direct costs associated with cybersecurity incidents costs Australian businesses $29 billion per annum
- 2020, U.S. government to spend over $18 billion on cybersecurity—Atlas VPN
- De Luca, 2021, Explainable artificial intelligence for workflow verification in visual IoT/robotics programming language environment, J. Artif. Intell. Technol., 1, 21
- 2021, Explainable and trustworthy artifcial intelligence
- 2021, Explainable AI for healthcare
- 2021, Theoretic foundation, criticism, and application trend of explainable AI
- 2020, Explainable artificial intelligence
- 2020, Interpretable & explainable machine vision
- 2019, Interpreting, explaining and visualizing deep learning
- 2019, Explainable deep learning for efficient and robust pattern recognition
- 2019, Explainable artificial intelligence (XAI)
- 2019, Interpretating and explaining visual artifcial intelligence models
- 2020, Explainable artificial intelligence (XAI)
- 2020, ACM conference on fairness, accountability, and transparency
- 2019, What happens when self-driving cars kill people
- Martin, 2021, Cyber kill chain
- Singh, 2014, Computer network and information security, Comput. Netw. Inf. Security, 8, 41
- Anderson, Evading machine learning malware detection, 9
- Goodfellow, 2016, Deep Learning, 1
- Raschka, 2017, Python Machine Learning—Second Edition
- 10.5555/3045118.3045336 · doi:10.5555/3045118.3045336
- 10.2307/2699986 · doi:10.2307/2699986
- 2016, Fog Computing and the Internet of Things: Extend the Cloud to Where the Things Are
- 2020, 2020 unit 42 IoT threat report
- Amershi, 2020, Research collection: Research supporting Responsible Ai
- 2021, Responsible AI practices
Cited by
Items 301–325 of 325
Gharkan, M., Al-Janabi, M. I. H., & Salim, O. (2026) · Al-Noor Journal of Engineering Management and Computer Science · doi:10.71229/e60hg537
Bhumireddi, S. (2025) · NORMA
PQC-Secured Intrusion Detection for Quantum Networks: A Two-Layer Defense Architecture
Aydeger, A., & Zeydan, E. (2026) · Conference on Local Computer Networks · doi:10.1109/lcn67947.2026.11660772
A hybrid CNN–LSTM framework for cross-domain anomaly detection in cyber-physical IoT networksPreprint
Inuwa, M. M., Daş, R., & Özdem, M. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.7435053
Hybrid Approach for Intrusion Detection in IoT Networks
Awa, M. N., Sall, M., Njoya, A. N., Nouria, F. M. K., & Attiogbe, A. (2026) · EAI/Springer Innovations in Communication and Computing · doi:10.1007/978-3-032-08820-8_15
Γονιδάκης, Κ. Δ. (2026) · University of Thessaly Institutional Repository (University of Thessaly)
An enhanced proactive social defense mechanism for Fork attack in IoT network using QHC and BFCC
Aditya, Y., & Jain, P. (2026) · Scientific Reports · doi:10.1038/s41598-026-68725-x
Iqbal, A., & Asghar, S. (2026) · Journal of Information Security and Applications · doi:10.1016/j.jisa.2026.104644
Hadi, S. Z. A., Issa, S. S., & Котенко, А. П. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-39711-9_6
Explainable intrusion detection for the internet of things: a systematic literature review
Souto, A. M., Sobrinho, Á., da Silva, L. D., da Silva, L. D., Ceccarelli, A., & Perkusich, Â. (2026) · Internet of Things · doi:10.1016/j.iot.2026.102107
Explainable AI and TinyML for Trustworthy and Efficient IoT: A Survey
Al-Junaid, A. F., Al-Qahtani, M. M., & Alhefdi, M. (2026) · IEEE Access · doi:10.1109/access.2026.3736702
“These works focus heavily on hardware-software co-design [5], the creation of taxonomies for applications ranging from smart farming to anomaly detection [35], and the specific challenges associated with deploying microcontroller units (MCUs) in healthcare wearables [38].”
Alalawi, K. (2026) · IEEE Access · doi:10.1109/access.2026.3738040
“Intrusion detection for IoT environments has attracted sustained research attention over the past decade [1], [4], and machine learning studies have reported high detection performance alongside methodological problems that remain unresolved.”
InfluentialSingh, S. P., & Ojha, D. (2026) · Journal of Mechanical and Construction Engineering (JMCE)
Al-Wrafif, S. A.-D. T., Alduais, N. A. M., Al-Quadhi, A., Mostafa, S. A., Hlayel, M., & Elhefnawy, A. M. M. M. (2026) · International Conference on Advancements in Smart, Secure and Intelligent Computing (ASSIC) · doi:10.1109/assic70037.2026.11716784
“…fact that many of these devices utilize a small amount of processing power and do not have much in terms of security measures built into their design, they are susceptible to a number of different forms of cyberattacks (e.g., DDoS, ransomware, scanning, injection, backdoor access, and XSS) [1][2].”
Amosa, B. M. G., Fabiyi, A. O., Etudaiye, A. I., & Akporherhe, O. A. (2026) · British Journal of Contemporary Research · doi:10.67693/bjcr-hww3n6jf
CORAL-IDS: A Calibrated Risk Layer for Explainable, Open-world Intrusion Detection
Khaled, S. H., Hasan, R. H., Mohan, A. T., & Khioon, A. S. A. A. (2026) · International journal of intelligent engineering and systems · doi:10.22266/ijies2026.1130.82
Bajaj, A. (2026) · International Conference on Smart Electronics and Communication (ICOSEC) · doi:10.1109/icosec70158.2026.11735698
“This framework integrated machine learning and deep learning with interpretability for decision-making, enhancing transparency of attacks, cybersecurity trust and threat analysis, uncovering research challenges, and highlighting future opportunities for IoT cyber defense [18].”
Olaymi, S. E. Z. (2026) · Neural Computing and Applications · doi:10.1007/s00521-026-12597-9
Lu, T., & Xia, Z. (2026) · Discover Artificial Intelligence · doi:10.1007/s44163-026-02638-6
“…threat detection depends on network logs, endpoint data, user behavior data, IoT signals, system alerts, cloud activity records, and threat intelligence sources, which are commonly used in cybersecurity analytics, intrusion detection, anomaly detection, and IoT-based security systems [12, 25, 26].”
InfluentialTanwar, G., Kumawat, S., Kumari, P., Prasad, M., Rai, A. K., & Agrawal, A. (2026) · Genetics and Molecular Research · doi:10.4238/2axf3427
Ayoub, R., Haider, U., & Creutzburg, R. M. (2026) · International Conference on Computer Science and Engineering (UBMK) · doi:10.1109/ubmk71847.2026.11745391
“Moustafa et al. [17] survey XAI-enhanced IDS and identify explanation stability and consistency across input types as critical open challenges, the specific gaps this paper addresses.”
Journeyprint: Detecting Malicious Agentic AI by the Behavioral Fingerprint of Its Web Journey
Nacshon, L., Garzon, J., Bell, N., & Anderson, K. (2026) · International Conference on Cybersecurity and AI-Based Systems (Cyber-AI) · doi:10.1109/cyber-ai70573.2026.11755167
Moreno, J. F. C., Rizzardi, A., Sicari, S., & Coen‐Porisini, A. (2026) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2026.116653
Ogu, R. E. (2026) · Journal of Computing and Data Technology · doi:10.71426/jcdt.v2.i2.pp207-215
AĞGÜL, B. (2026) · Frontiers in Big Data · doi:10.3389/fdata.2026.1934235
Metadata
- Authors
- Moustafa, N., Koroniotis, N., Keshk, M., Zomaya, A. Y., & Tari, Z.
- Affiliations
- UNSW Sydney (AU)
- The University of Sydney (AU)
- RMIT University (AU)
- Venue
- IEEE Communications Surveys & Tutorials
- Published
- 2023
- Volume
- 25
- Issue
- 3
- Pages
- 1775-1807
- ISSN
- 1553-877X (ISSN-L)
- 2373-745X
- OpenAlex
- W4378365142
- Cited by
- 266
- References
- 187
- Citations by year
- 2026: 70 · 2025: 110 · 2024: 71 · 2023: 15
- FWCI
- 29.55
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE Communications Surveys & Tutorials (publishedVersion)
- Figshare (submittedVersion)
- Funders
- Australian Research Council’s Discovery Early Career Researcher Award (DE230100116)
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › Network Security and Intrusion Detection
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning
- Keywords
- explainable AI, intrusion detection, Internet of Things security, anomaly-based intrusion detection, cyber defense, interpretability, deep learning anomaly detection, attack detection
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.