pith. sign in

arxiv: 2606.18771 · v1 · pith:YFXQ5U5Bnew · submitted 2026-06-17 · 💻 cs.CR

A Predictive Neural Network Architecture for Early Detection of Low-Rate Cyberattacks

Mert Nak{\i}p This is my paper

Pith reviewed 2026-06-26 20:38 UTC · model grok-4.3

classification 💻 cs.CR
keywords low-rate DoSintrusion detectionQoS predictionneural networkIoT securitySDNDDoS detectionpredictive model
0
0 comments X

The pith

A neural network forecasts future QoS values from traffic history to detect low-rate denial-of-service attacks by spotting prediction errors.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces IDQS, a framework for early LDoS detection in IoT that first trains a recurrent neural network on past traffic to predict upcoming Quality of Service levels. A second component then measures how much the actual observed values depart from those forecasts and decides whether an attack is underway. The method targets the subtle, extended nature of low-rate attacks that slip past conventional detectors. Evaluation on the SDN-SlowRate-DDoS and CIC-IDS2017 datasets yields detection accuracies above 79 percent and 91 percent respectively, with high recall, few missed attacks, and an end-to-end runtime of 0.28 seconds.

Core claim

IDQS combines RTP-QoS, a Recurrent Trend Predictive Neural Network that learns historical traffic patterns to forecast future QoS metrics, with PDM, a Pairwise Decision Model that flags potential LDoS attacks when discrepancies arise between predicted and observed QoS. On the SDN-SlowRate-DDoS dataset it exceeds 79 percent detection accuracy and on CIC-IDS2017 it exceeds 91 percent, both with high recall and low false negatives while completing inference in 0.28 seconds.

What carries the argument

RTP-QoS, the Recurrent Trend Predictive Neural Network that forecasts QoS from historical patterns, paired with PDM, the Pairwise Decision Model that treats large prediction-observation gaps as attack indicators.

If this is right

  • LDoS attacks can be identified before they cause noticeable service degradation.
  • The approach runs fast enough for continuous monitoring on resource-limited IoT hardware.
  • High recall across varied attack patterns reduces the chance of missing ongoing low-rate assaults.
  • Performance holds on two independent public datasets, suggesting applicability beyond a single traffic source.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same prediction-plus-discrepancy logic could be applied to other measurable network quantities to catch different attack families.
  • Retraining the predictor on recent data streams might improve robustness when traffic patterns shift over time.
  • Embedding the detector inside existing SDN controllers would allow automated responses without extra sensors.

Load-bearing premise

That gaps between the network's QoS forecasts and the values actually observed are caused by attacks rather than ordinary traffic changes or quirks of the chosen datasets.

What would settle it

A collection of normal IoT traffic traces containing only routine fluctuations on which the system nevertheless generates frequent false attack alerts.

Figures

Figures reproduced from arXiv: 2606.18771 by Mert Nak{\i}p.

Figure 1
Figure 1. Figure 1: Structure of the proposed Intrusion Detection via QoS Prediction System (IDQS) [PITH_FULL_IMAGE:figures/full_fig_p007_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: QoS predicted by the modified rTPNN and the actual [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: The predicted and the actual QoS values under LDoS attacks with a single attacker creating 25 connections per second [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Boxplot of N-SMAPE calculated for three di [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figure 6
Figure 6. Figure 6: The boxplot of the performances of the PDM and FFN over the CV folds for each of Accuracy, Precision, Recall, and F1 [PITH_FULL_IMAGE:figures/full_fig_p013_6.png] view at source ↗
Figure 5
Figure 5. Figure 5: Performance of the proposed approach for the cases [PITH_FULL_IMAGE:figures/full_fig_p013_5.png] view at source ↗
Figure 7
Figure 7. Figure 7: The actual QoS values and corresponding detection results under LDoS attacks with a single attacker creating 25 connec [PITH_FULL_IMAGE:figures/full_fig_p014_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Performance of the proposed approach on the CIC [PITH_FULL_IMAGE:figures/full_fig_p015_8.png] view at source ↗
read the original abstract

Low-Rate Denial of Service (LDoS) attacks pose a significant challenge to IoT networks due to their subtle and prolonged nature, often evading traditional intrusion detection systems. This paper presents IDQS (Intrusion Detection via QoS Prediction), a lightweight and proactive framework for early LDoS attack detection. IDQS integrates two new key components: (i) RTP-QoS, a Recurrent Trend Predictive Neural Network that learns and forecasts future Quality of Service (QoS) based on historical traffic patterns, and (ii) PDM, a Pairwise Decision Model that evaluates discrepancies between predicted and actual QoS to identify potential attacks. Evaluated on the public SDN-SlowRate-DDoS and CIC-IDS2017 datasets, IDQS respectively achieves over 79% and 91% detection accuracy across most attack scenarios with high recall and low false negatives, while maintaining an end-to-end inference time of just 0.28 seconds. The results demonstrate the effectiveness and efficiency of IDQS for real-time deployment in resource-constrained IoT environments.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The manuscript proposes IDQS, a framework for early detection of Low-Rate Denial of Service (LDoS) attacks in IoT networks. It integrates RTP-QoS, a Recurrent Trend Predictive Neural Network that forecasts Quality of Service metrics from historical traffic patterns, and PDM, a Pairwise Decision Model that flags attacks via discrepancies between predicted and observed QoS values. On the SDN-SlowRate-DDoS and CIC-IDS2017 datasets, the method is reported to achieve detection accuracies exceeding 79% and 91% respectively, with high recall, low false negatives, and an end-to-end inference time of 0.28 seconds.

Significance. If the performance claims hold under proper validation, the work would address a recognized gap in detecting subtle, prolonged LDoS attacks that often evade signature- or anomaly-based IDS in IoT settings. The predictive-plus-comparison design is a plausible direction for proactive detection in resource-constrained environments, though its practical value rests on demonstrating that residuals are attack-specific rather than artifacts of normal traffic variance.

major comments (2)
  1. [Abstract] Abstract: The reported accuracies (over 79% and 91%) and inference time are stated without any information on the neural network training procedure, validation splits, hyperparameter choices, or controls for overfitting. This absence prevents evaluation of whether the numbers reflect genuine generalization or dataset-specific fitting.
  2. [Abstract] Abstract: The RTP-QoS + PDM pipeline is presented as separating LDoS from benign traffic via prediction discrepancies, yet no ablation, threshold robustness test, or statistical comparison of residual distributions under controlled benign variance (e.g., bursty IoT flows) is supplied. Without such evidence the central claim that discrepancies reliably indicate attacks remains unsupported.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback on our manuscript. We address each major comment below and indicate where revisions will be made to strengthen the presentation of our results.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The reported accuracies (over 79% and 91%) and inference time are stated without any information on the neural network training procedure, validation splits, hyperparameter choices, or controls for overfitting. This absence prevents evaluation of whether the numbers reflect genuine generalization or dataset-specific fitting.

    Authors: The full manuscript details the experimental protocol in Section 4, including time-series cross-validation on the two public datasets, hyperparameter selection via grid search (LSTM hidden size 64, learning rate 0.001, dropout 0.2), and early stopping based on validation loss to mitigate overfitting. The abstract is intentionally concise, but we agree that a brief reference to the validation approach would improve transparency. We will revise the abstract to include one sentence summarizing the training and validation procedure. revision: yes

  2. Referee: [Abstract] Abstract: The RTP-QoS + PDM pipeline is presented as separating LDoS from benign traffic via prediction discrepancies, yet no ablation, threshold robustness test, or statistical comparison of residual distributions under controlled benign variance (e.g., bursty IoT flows) is supplied. Without such evidence the central claim that discrepancies reliably indicate attacks remains unsupported.

    Authors: The PDM is designed to flag statistically significant deviations between predicted and observed QoS, and the reported high recall with low false negatives on both datasets provides supporting evidence. However, we acknowledge that explicit ablation of the predictive component, threshold sensitivity analysis, and direct comparison of residual distributions under controlled benign variance would further substantiate the claim. We will add these analyses (including ablation results and Kolmogorov-Smirnov tests on residuals) to the revised manuscript. revision: yes

Circularity Check

0 steps flagged

No circularity: standard predictive anomaly detection pipeline with independent training and evaluation

full rationale

The IDQS framework trains RTP-QoS on historical traffic patterns to produce QoS forecasts and then applies PDM to discrepancies with observed values. This is a conventional prediction-error anomaly detector whose training objective and detection rule are not equivalent by construction; the reported accuracies on SDN-SlowRate-DDoS and CIC-IDS2017 are external performance metrics rather than tautological outputs of the same fitted quantities. No self-citation load-bearing steps, uniqueness theorems, or ansatz smuggling appear in the provided description, and the derivation chain remains self-contained against the public datasets.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract supplies no information on free parameters, axioms, or invented entities.

pith-pipeline@v0.9.1-grok · 5711 in / 1010 out tokens · 30516 ms · 2026-06-26T20:38:04.949726+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

58 extracted references · 6 canonical work pages

  1. [1]

    Al-Turjman, Artificial intelligence in IoT, V ol

    F. Al-Turjman, Artificial intelligence in IoT, V ol. 13, Springer, 2019

    2019

  2. [2]

    Baccour, N

    E. Baccour, N. Mhaisen, A. A. Abdel- latif, A. Erbad, A. Mohamed, M. Hamdi, 16 M. Guizani, Pervasive ai for iot applications: A survey on resource-efficient distributed arti- ficial intelligence, IEEE Communications Sur- veys & Tutorials 24 (4) (2022) 2366–2418

    2022

  3. [3]

    K. G. Manoharan, J. A. Nehru, S. Balasubra- manian, Artificial intelligence and iot, Springer 10 (2021) 978–981

    2021

  4. [4]

    Ericsson, Ericsson Mobility Report (Nov. 2019). [link]. URLhttps://www.ericsson.com/en/ mobi-lity-report

    2019

  5. [5]

    URLhttps://www.hp.com/us-en/ hp-news/press-release.html?id= 1744676

    Hp study reveals 70 percent of Internet of Things devices vulnerable to attack. URLhttps://www.hp.com/us-en/ hp-news/press-release.html?id= 1744676

  6. [6]

    Gelenbe, M

    E. Gelenbe, M. Nakıp, T. Czachórski, Improv- ing massive access to IoT gateways, Perfor- mance Evaluation 157 (2022) 102308

    2022

  7. [7]

    Rodoplu, M

    V . Rodoplu, M. Nakıp, D. T. Eliiyi, C. Güzeli¸ s, A multiscale algorithm for joint forecasting– scheduling to solve the massive access problem of iot, IEEE Internet of Things Journal 7 (9) (2020) 8572–8589

    2020

  8. [8]

    Yoachimik, DDoS attack trends for 2022 Q2, accessed: 2023-03-09 (June 2022)

    O. Yoachimik, DDoS attack trends for 2022 Q2, accessed: 2023-03-09 (June 2022). URLhttps://blog.cloudflare.com/ ddos-attack-trends-for-2022-q2/

    2022

  9. [9]

    URLhttps://www.extrahop.com/ re-sources/attacks/dos/

    ExtraHop, Denial of Service attack: definition, examples, and prevention, accessed: 2023-03- 09 (January 2022). URLhttps://www.extrahop.com/ re-sources/attacks/dos/

    2023

  10. [11]

    Nasereddin, M

    M. Nasereddin, M. Nakıp, E. Gelenbe, Mea- surement based evaluation and mitigation of flood attacks on a lan test-bed, in: 2023 IEEE 48th Conference on Local Computer Net- works (LCN), 2023, pp. 1–4.doi:10.1109/ LCN58197.2023.10223358

    arXiv 2023

  11. [12]

    Kiner, T

    E. Kiner, T. April, Google mitigated the largest DDoS attack to date, peaking above 398 million rps, accessed: 2023-11-20 (October 2023). URLhttps://cloud.google.com/ blog/products/identity-security/ google-cloud-mitigated-largest-ddos-attack-peaking-above-398-million-rps

    2023

  12. [13]

    IEEE Access8, 62011–62031 (2020) https://doi.org/10.1109/ACCESS.2020

    W. Zhijun, L. Wenjing, L. Liang, Y . Meng, Low-rate dos attacks, detection, defense, and challenges: A survey, IEEE Access 8 (2020) 43920–43943.doi:10.1109/ACCESS.2020. 2976609

    work page doi:10.1109/access.2020 2020

  13. [14]

    URLhttps://www.cloudflare

    What is a low and slow attack?, accessed: 2024-03-02. URLhttps://www.cloudflare. com/en-gb/learning/ddos/ ddos-low-and-slow-attack/

    2024

  14. [15]

    N. M. Yungaicela-Naula, C. Vargas-Rosales, J. A. Pérez-Díaz, E. Jacob, C. Martinez- Cagnazzo, Sdn-slowrate-ddos dataset (2023). doi:10.21227/amrt-8y98. URLhttps://dx.doi.org/10.21227/ amrt-8y98

    work page doi:10.21227/amrt-8y98 2023

  15. [16]

    Sharafaldin, A

    I. Sharafaldin, A. H. Lashkari, A. A. Ghorbani, et al., Toward generating a new intrusion de- tection dataset and intrusion traffic characteri- zation., ICISSp 1 (2018) 108–116

    2018

  16. [17]

    Bovenzi, G

    G. Bovenzi, G. Aceto, D. Ciuonzo, A. Montieri, V . Persico, A. Pescapé, Network anomaly de- tection methods in IoT environments via deep learning: A fair comparison of performance and robustness, Computers & Security 128 (2023) 103167

    2023

  17. [18]

    IoT-23: A labeled dataset with malicious and benign IoT network traffic,

    S. Garcia, A. Parmisano, M. J. Erquiaga, IoT- 23: A labeled dataset with malicious and be- nign IoT network traffic (May 2021).doi: 10.5281/zenodo.4743746. URLhttps://doi.org/10.5281/zenodo. 4743746 17

    work page doi:10.5281/zenodo.4743746 2021

  18. [19]

    URLhttps://www.kaggle.com/ymirsky/ network-attack-dataset-kitsune

    Kitsune Network Attack Dataset (August 2020). URLhttps://www.kaggle.com/ymirsky/ network-attack-dataset-kitsune

    2020

  19. [20]

    Nakip, E

    M. Nakip, E. Gelenbe, Mirai botnet attack detection with auto-associative dense random neural network, in: 2021 IEEE Global Commu- nications Conference (GLOBECOM), IEEE, 2021, pp. 01–06

    2021

  20. [21]

    Nakip, E

    M. Nakip, E. Gelenbe, Botnet attack detection with incremental online learning, in: Security in Computer and Information Sciences: Second International Symposium, EuroCybersec 2021, Nice, France, October 25–26, 2021, Revised Selected Papers, Springer, 2022, pp. 51–60

    2021

  21. [22]

    C. S. Htwe, Y . M. Thant, M. M. S. Thwin, Botnets attack detection using machine learn- ing approach for IoT environment, in: Journal of Physics: Conference Series, V ol. 1646, IOP Publishing, 2020, p. 012101

    2020

  22. [23]

    Sriram, R

    S. Sriram, R. Vinayakumar, M. Alazab, K. So- man, Network flow based IoT botnet attack detection using deep learning, in: IEEE IN- FOCOM 2020-IEEE Conference on Computer Communications Workshops (INFOCOM WK- SHPS), IEEE, 2020, pp. 189–194

    2020

  23. [24]

    Y . N. Soe, Y . Feng, P. I. Santosa, R. Hartanto, K. Sakurai, Machine learning-based IoT-botnet attack detection with sequential architecture, Sensors 20 (16) (2020) 4372

    2020

  24. [25]

    Nakıp, E

    M. Nakıp, E. Gelenbe, Online self-supervised deep learning for intrusion detection systems, IEEE Transactions on Information Forensics and Security (2024)

    2024

  25. [26]

    Polat, M

    H. Polat, M. Türko ˘glu, O. Polat, A. ¸ Sengür, A novel approach for accurate detection of the ddos attacks in sdn-based scada systems based on deep recurrent neural networks, Expert Sys- tems with Applications 197 (2022) 116748

    2022

  26. [27]

    Polat, S

    O. Polat, S. Oyucu, M. Türko ˘glu, H. Polat, A. Aksoz, F. Yardımcı, Hybrid ai-powered real- time distributed denial of service detection and traffic monitoring for software-defined-based vehicular ad hoc networks: A new paradigm for securing intelligent transportation networks, Applied Sciences 14 (22) (2024) 10501

    2024

  27. [28]

    A. Jain, R. Bagoria, P. Arora, An intelligent zero-day attack detection system using unsu- pervised machine learning for enhancing cy- ber security, Knowledge-Based Systems (2025) 113833

    2025

  28. [29]

    Macia-Fernandez, J

    G. Macia-Fernandez, J. E. Diaz-Verdejo, P. Garcia-Teodoro, Mathematical model for low-rate dos attacks against application servers, IEEE Transactions on Information Forensics and Security 4 (3) (2009) 519–529.doi:10. 1109/TIFS.2009.2024719

    arXiv 2009

  29. [30]

    Z. Wu, L. Zhang, M. Yue, Low-rate dos attacks detection based on network multifractal, IEEE Transactions on Dependable and Secure Com- puting 13 (5) (2016) 559–567.doi:10.1109/ TDSC.2015.2443807

    arXiv 2016

  30. [31]

    D. Tang, S. Zhang, J. Chen, X. Wang, The detection of low-rate dos attacks using the sadbscan algorithm, Information Sciences 565 (2021) 229–247

    2021

  31. [32]

    Aiello, E

    M. Aiello, E. Cambiaso, M. Mongelli, G. Pa- paleo, An on-line intrusion detection approach to identify low-rate dos attacks, in: 2014 In- ternational Carnahan Conference on Security Technology (ICCST), 2014, pp. 1–6.doi: 10.1109/CCST.2014.6987039

    work page doi:10.1109/ccst.2014.6987039 2014

  32. [33]

    A. Reed, L. S. Dooley, S. K. Mostefaoui, A reliable real-time slow dos detection frame- work for resource-constrained iot networks, in: 2021 IEEE Global Communications Confer- ence (GLOBECOM), 2021, pp. 1–6.doi: 10.1109/GLOBECOM46510.2021.9685612

    work page doi:10.1109/globecom46510.2021.9685612 2021

  33. [34]

    L. Liu, H. Wang, Z. Wu, M. Yue, The detec- tion method of low-rate dos attack based on multi-feature fusion, Digital Communications and Networks 6 (4) (2020) 504–513

    2020

  34. [35]

    Cheng, J

    H. Cheng, J. Liu, T. Xu, B. Ren, J. Mao, W. Zhang, Machine learning based low-rate 18 ddos attack detection for sdn enabled iot net- works, International Journal of Sensor Net- works 34 (1) (2020) 56–69

    2020

  35. [36]

    C. Xu, J. Shen, X. Du, Low-rate dos attack detection method based on hybrid deep neural networks, Journal of Information Security and Applications 60 (2021) 102879

    2021

  36. [37]

    H. S. Ilango, M. Ma, R. Su, Low rate dos at- tack detection in iot - sdn using deep learning, in: 2021 IEEE International Conferences on Internet of Things (iThings) and IEEE Green Computing & Communications (GreenCom) and IEEE Cyber, Physical & Social Comput- ing (CPSCom) and IEEE Smart Data (Smart- Data) and IEEE Congress on Cybermatics (Cy- bermatics),...

    2021

  37. [38]

    H. S. Ilango, M. Ma, R. Su, A feedforward– convolutional neural network to detect low-rate dos in iot, Engineering Applications of Artifi- cial Intelligence 114 (2022) 105059

    2022

  38. [39]

    Vengathattil, S

    S. Vengathattil, S. M. Shaffi, Advanced network security through predictive intelli- gence: Machine learning approaches for proac- tive threat detection—an experimental study, Methodology 2 (3)

  39. [40]

    S. E. Maoudj, A. Belghiat, A deep learning- based approach with two-step minority classes prediction for intrusion detection in internet of things networks, Knowledge-Based Systems 312 (2025) 113143

    2025

  40. [41]

    S. S. Akhi, F. Shakil, S. K. Dey, M. I. Tusher, F. Kamruzzaman, S. S. Jamee, S. A. Tisha, N. Rahman, Enhancing banking cybersecurity: An ensemble-based predictive machine learn- ing approach, Am. J. Eng. Technol. 7 (03) (2025) 88–97

    2025

  41. [42]

    Huang, X

    Z. Huang, X. Huang, K. Xue, J. Han, L. Chen, Q. Sun, J. Lu, Defending against link-flooding attacks with adversary interest prediction and grouped online load balancing, IEEE Transac- tions on Networking 33 (3) (2025) 1056–1069

    2025

  42. [43]

    Al-Quayed, Z

    F. Al-Quayed, Z. Ahmad, M. Humayun, A sit- uation based predictive approach for cyberse- curity intrusion detection and prevention us- ing machine learning and deep learning algo- rithms in wireless sensor networks of industry 4.0, Ieee Access 12 (2024) 34800–34819

    2024

  43. [44]

    V . D. M. Rios, P. R. Inácio, D. Magoni, M. M. Freire, Detection and mitigation of low-rate denial-of-service attacks: A survey, IEEE Ac- cess 10 (2022) 76648–76668

    2022

  44. [45]

    Vedula, P

    V . Vedula, P. Lama, R. V . Boppana, L. A. Trejo, On the detection of low-rate denial of service attacks at transport and application layers, Elec- tronics 10 (17) (2021) 2105

    2021

  45. [46]

    D. Tang, X. Wang, Y . Yan, D. Zhang, H. Zhao, Adms: An online attack detection and mitiga- tion system for ldos attacks via sdn, Computer Communications 181 (2022) 454–471

    2022

  46. [47]

    Hajimaghsoodi, R

    M. Hajimaghsoodi, R. Jalili, Rad: A statis- tical mechanism based on behavioral analysis for ddos attack countermeasure, IEEE Transac- tions on Information Forensics and Security 17 (2022) 2732–2745

    2022

  47. [48]

    Çakmak, The impact of denial-of-service attacks and queue management algorithms on cellular networks, Journal of Intelligent Sys- tems: Theory and Applications 7 (1) (2024) 1– 13

    M. Çakmak, The impact of denial-of-service attacks and queue management algorithms on cellular networks, Journal of Intelligent Sys- tems: Theory and Applications 7 (1) (2024) 1– 13

    2024

  48. [49]

    D. V . Hernandez, Y .-K. Lai, H. T. Ignatius, Real-time ddos detection in high-speed net- works: A deep learning approach with multi- variate time series, Electronics 14 (13) (2025) 2673

    2025

  49. [50]

    de Miranda Rios, Detection of stealthy dis- tributed denial of service attacks using artifi- cial intelligence methods, Ph.D

    V . de Miranda Rios, Detection of stealthy dis- tributed denial of service attacks using artifi- cial intelligence methods, Ph.D. thesis, Univer- sidade da Beira Interior (2025)

    2025

  50. [51]

    C.-C. Chiu, T. He, Stealthy dgos attack: de- grading of service under the watch of network tomography, IEEE/ACM Transactions on Net- working 29 (3) (2021) 1294–1307. 19

    2021

  51. [52]

    Yousef, Unveiling the stealthy threat: Low- rate denial of service (ldos) attacks, in: Key Is- sues in Network Protocols and Security, Inte- chOpen, 2024

    D. Yousef, Unveiling the stealthy threat: Low- rate denial of service (ldos) attacks, in: Key Is- sues in Network Protocols and Security, Inte- chOpen, 2024

    2024

  52. [53]

    Y . Wu, Y . Ru, Z. Lin, C. Liu, T. Xue, X. Zhao, J. Chen, Research on cyber attacks and de- fensive measures of power communication net- work, IEEE internet of things journal 10 (9) (2022) 7613–7635

    2022

  53. [54]

    Moreira de Souza Berretta, P

    F. Moreira de Souza Berretta, P. C. Oprime, J. E. Sordan, Cyberattacks and the risk to pro- duction and service systems–an analysis from the perspective of quality management, The TQM Journal (2025)

    2025

  54. [55]

    T. A. Nguyen, D. Min, E. Choi, J.-W. Lee, De- pendability and security quantification of an in- ternet of medical things infrastructure based on cloud-fog-edge continuum for healthcare moni- toring using hierarchical models, IEEE Internet of Things Journal 8 (21) (2021) 15704–15748

    2021

  55. [56]

    Baldini, Detection of cybersecurity spoofing attacks in vehicular networks with recurrence quantification analysis, Computer Communica- tions 191 (2022) 486–499

    G. Baldini, Detection of cybersecurity spoofing attacks in vehicular networks with recurrence quantification analysis, Computer Communica- tions 191 (2022) 486–499

    2022

  56. [57]

    Yakin, M

    N. Yakin, M. Zhitkov, A. Chernikov, P. Pe- pelyaev, Security threats and service degrada- tion detection in lorawan networks, in: 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology (USBEREIT), IEEE, 2021, pp. 0455–0458

    2021

  57. [58]

    Sawas, H

    A. Sawas, H. E. Farag, Real-time detection of stealthy iot-based cyber-attacks on power distribution systems: A novel anomaly pre- diction approach, Electric Power Systems Re- search 223 (2023) 109496

    2023

  58. [59]

    Kfoury, Jorge Crichigno, and Elias Bou-Harb

    M. Nakip, C. Güzeli¸ s, O. Yildiz, Recur- rent trend predictive neural network for multi- sensor fire detection, IEEE Access 9 (2021) 84204–84216.doi:10.1109/ACCESS.2021. 3087736. 20

    work page doi:10.1109/access.2021 2021