Toward Realistic Wi-Fi Fault Diagnosis: A Multi-Modal Benchmark

Fengxiao Tang; Haobo Deng; Junjian Zhang; Ming Zhao; Nei Kato; Xinxin Li

arxiv: 2605.22008 · v1 · pith:SZPOWVIEnew · submitted 2026-05-21 · 💻 cs.NI

Toward Realistic Wi-Fi Fault Diagnosis: A Multi-Modal Benchmark

Junjian Zhang , Haobo Deng , Xinxin Li , Ming Zhao , Fengxiao Tang , Nei Kato This is my paper

Pith reviewed 2026-05-22 03:10 UTC · model grok-4.3

classification 💻 cs.NI

keywords Wi-Fi fault diagnosismulti-modal datasetnetwork testbedfault injection systembenchmarkingwireless networksLLM diagnosiscross-layer observations

0 comments

The pith

A real-world Wi-Fi testbed and multi-modal dataset with over 10,000 fault samples benchmarks diagnosis methods in heterogeneous environments.

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

The paper constructs a campus-based Wi-Fi testbed equipped with an automated fault injection system to generate realistic fault conditions. It collects a large dataset that includes multiple types of operational data from different wireless scenarios. This resource supports a unified benchmark for various diagnosis tasks and methods, including traditional and LLM-based approaches. Results highlight the difficulty of using heterogeneous data effectively and point to key considerations for advancing the field.

Core claim

By deploying a real-world Wi-Fi testbed in campus environments with automated fault injection, the authors collect over 10,000 multi-modal fault samples capturing heterogeneous cross-layer observations, establishing the first public benchmark for Wi-Fi fault diagnosis and showing that existing approaches struggle to leverage the diversity of data.

What carries the argument

The multi-modal Wi-Fi fault dataset that jointly captures heterogeneous cross-layer operational observations across diverse wireless scenarios.

If this is right

Existing diagnosis approaches have difficulty effectively leveraging heterogeneous operational data.
LLM-based approaches can be assessed using a reasoning-oriented framework for consistency with network conditions.
Several important considerations emerge for designing future multi-modal Wi-Fi diagnosis systems.
The dataset enables systematic evaluation spanning multiple tasks, modalities, and paradigms.

Where Pith is reading between the lines

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

Similar testbeds could be developed for other wireless technologies to create comparable benchmarks.
The findings may encourage the development of new models that better integrate cross-layer data for fault detection.
Public release of the dataset allows independent verification and extension by the research community.

Load-bearing premise

The automated fault injection system and campus deployment generate fault patterns and environmental heterogeneity that match those found in typical practical Wi-Fi networks.

What would settle it

A comparison study showing that diagnosis models trained on this dataset perform significantly differently when tested on faults from non-campus Wi-Fi networks would challenge the representativeness of the benchmark.

read the original abstract

Intelligent network operation and maintenance systems in modern networks continuously generate large volumes of multi-modal operational data. However, Wi-Fi fault diagnosis under heterogeneous operational environments remains insufficiently understood. We build a real-world Wi-Fi testbed deployed in campus working environments with an automated fault injection system, and collect a multi-modal Wi-Fi fault dataset containing over 10,000 fault samples across diverse wireless scenarios. To the best of our knowledge, this is among the first publicly available datasets jointly capturing heterogeneous cross-layer operational observations for Wi-Fi fault diagnosis. Based on this dataset, we establish a unified benchmark spanning multiple diagnosis tasks, operational modalities, and representative diagnosis paradigms. Experimental results indicate that effectively leveraging heterogeneous operational data remains challenging for existing diagnosis approaches. We further evaluate emerging LLM-based approaches and develop a reasoningoriented evaluation framework to assess the consistency between generated diagnostic analyses and actual network conditions. Our findings suggest several important considerations for future multi-modal Wi-Fi diagnosis.

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.

Desk Editor's Note private letter to a colleague

This paper releases a new multi-modal Wi-Fi fault dataset from a campus testbed with automated injection, which fills a gap but rests on unvalidated claims of realism.

read the letter

The one or two things to know: this work releases a public multi-modal Wi-Fi fault dataset collected from a campus testbed using automated fault injection, with over 10,000 samples, and sets up benchmarks for diagnosis tasks including LLM-based methods. That is the core offering. They do a good job moving past synthetic or single-modality data by capturing cross-layer observations across diverse scenarios and showing that existing approaches struggle with the heterogeneity. The addition of a reasoning-oriented framework to check LLM diagnostic consistency is a practical step that highlights real challenges without overclaiming success. The dataset release itself is the clearest positive, since public resources like this are still uncommon in network operations work. The soft spots are around representativeness. The automated injection and campus environment are presented as realistic, yet there is no indication of quantitative matching to production traces from commercial or enterprise networks, such as fault frequency or interference patterns. If the full paper lacks those checks, the benchmark could embed testbed artifacts that limit its value for practical diagnosis research. This is a moderate issue for a data paper rather than a fatal one, but it directly affects how much the community can trust the results outside the specific setup. The citation pattern looks standard and the focus stays on the data and experiments rather than ungrounded modeling. This paper is for researchers in network management and multi-modal ML for wireless systems who need a shared testbed to compare methods. A reader working on fault diagnosis algorithms would get concrete value from the released samples and baselines. It deserves a serious referee because the dataset contribution is tangible and can be evaluated on its own terms. I recommend sending it for peer review.

Referee Report

2 major / 2 minor

Summary. The paper constructs a real-world Wi-Fi testbed in campus working environments using an automated fault injection system to collect a multi-modal dataset exceeding 10,000 fault samples across diverse wireless scenarios. It positions this as one of the first public datasets jointly capturing heterogeneous cross-layer observations for Wi-Fi fault diagnosis. The work then defines a unified benchmark covering multiple diagnosis tasks, modalities, and paradigms; evaluates conventional and LLM-based diagnosis approaches; introduces a reasoning-oriented framework to check consistency between LLM-generated analyses and actual conditions; and outlines considerations for future multi-modal diagnosis research.

Significance. If the collected faults and environmental heterogeneity prove representative, the public dataset and benchmark would fill a notable gap in Wi-Fi fault diagnosis research by enabling reproducible study of cross-layer, multi-modal data under realistic conditions. The explicit evaluation of emerging LLM-based methods together with a consistency-checking framework is a constructive addition that could inform how such models are assessed in network operations.

major comments (2)

[Testbed and data collection] Testbed and data collection section: The central claim that the automated fault injection system and campus deployment produce representative fault patterns and cross-layer heterogeneity is not supported by any quantitative comparison (e.g., statistical matching of fault-type frequencies, duration distributions, or interference signatures) to logs from commercial or enterprise Wi-Fi networks. This validation is load-bearing for the benchmark's claimed realism and generalizability.
[Benchmark and experimental evaluation] Benchmark and experimental evaluation section: The unified benchmark results indicating that existing approaches struggle with heterogeneous data are presented without sufficient detail on the exact cross-validation protocol or the definition of 'ground-truth' labels for the >10,000 samples, making it difficult to assess whether the reported challenges are intrinsic or artifacts of the testbed labeling process.

minor comments (2)

[Abstract] Abstract: The statement 'among the first publicly available datasets' would be strengthened by citing any prior Wi-Fi fault datasets and clarifying the precise novelty in joint cross-layer capture.
[Dataset description] The manuscript would benefit from an explicit table summarizing the modalities, fault types, and scenario diversity to improve readability of the dataset description.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed feedback. We address each major comment below, indicating planned revisions to strengthen the manuscript where appropriate.

read point-by-point responses

Referee: [Testbed and data collection] Testbed and data collection section: The central claim that the automated fault injection system and campus deployment produce representative fault patterns and cross-layer heterogeneity is not supported by any quantitative comparison (e.g., statistical matching of fault-type frequencies, duration distributions, or interference signatures) to logs from commercial or enterprise Wi-Fi networks. This validation is load-bearing for the benchmark's claimed realism and generalizability.

Authors: We acknowledge that the manuscript lacks a direct quantitative statistical comparison to proprietary commercial or enterprise Wi-Fi logs, which are generally inaccessible. The testbed was deployed in live campus environments with real user traffic and typical interference, and fault types were chosen from commonly reported Wi-Fi issues in the literature. In revision we will add an expanded discussion of fault selection criteria, qualitative alignment with public Wi-Fi studies, and an explicit limitations statement on generalizability to large-scale enterprise settings. This clarifies the basis for realism without overstating the evidence. revision: partial
Referee: [Benchmark and experimental evaluation] Benchmark and experimental evaluation section: The unified benchmark results indicating that existing approaches struggle with heterogeneous data are presented without sufficient detail on the exact cross-validation protocol or the definition of 'ground-truth' labels for the >10,000 samples, making it difficult to assess whether the reported challenges are intrinsic or artifacts of the testbed labeling process.

Authors: We agree that greater detail is required for reproducibility and to rule out labeling artifacts. Ground-truth labels were assigned according to the specific faults injected by the automated system, with verification performed on a random subset via manual inspection of the multi-modal traces. In the revised manuscript we will describe the exact cross-validation protocol (including split ratios and any stratification) and provide a step-by-step account of how ground-truth was generated and validated. These additions will allow readers to better distinguish intrinsic difficulties from testbed-specific factors. revision: yes

Circularity Check

0 steps flagged

Empirical data collection and benchmarking with no circular derivation chain

full rationale

The paper centers on building a physical testbed, injecting faults, and releasing a new multi-modal dataset of over 10,000 samples. No equations, fitted parameters, or predictions are presented that reduce by construction to prior inputs. Claims of being 'among the first' are empirical assertions about dataset novelty rather than self-referential derivations. The work is self-contained as an independent empirical contribution and does not rely on load-bearing self-citations or ansatzes for its core results.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

This is an empirical dataset and benchmarking study; no mathematical free parameters, axioms, or invented entities are introduced.

pith-pipeline@v0.9.0 · 5699 in / 965 out tokens · 40449 ms · 2026-05-22T03:10:57.831354+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

We build a real-world Wi-Fi testbed deployed in campus working environments with an automated fault injection system, and collect a multi-modal Wi-Fi fault dataset containing over 10,000 fault samples across diverse wireless scenarios.
IndisputableMonolith/Foundation/ArithmeticFromLogic.lean LogicNat induction and recovery unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

We further develop a structured reasoning evaluation framework that leverages LLMs to evaluate the consistency between generated fault analyses and underlying network conditions.

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

24 extracted references · 24 canonical work pages

[1]

Towards automatic network fault local- ization in real time using probabilistic inference,

A. Johnsson and C. Meirosu, “Towards automatic network fault local- ization in real time using probabilistic inference,” in2013 IFIP/IEEE In- ternational Symposium on Integrated Network Management (IM 2013). IEEE, 2013, pp. 1393–1398

work page 2013
[2]

Service-triggered failure identifica- tion/localization through monitoring of multiple parameters,

M. Ruiz, F. Fresi, A. P. Vela, G. Meloni, N. Sambo, F. Cugini, L. Poti, L. Velasco, and P. Castoldi, “Service-triggered failure identifica- tion/localization through monitoring of multiple parameters,” inECOC 2016; 42nd European Conference on Optical Communication. VDE, 2016, pp. 1–3. 11

work page 2016
[3]

Fault detection in wireless sensor networks through svm classifier,

S. Zidi, T. Moulahi, and B. Alaya, “Fault detection in wireless sensor networks through svm classifier,”IEEE Sensors Journal, vol. 18, no. 1, pp. 340–347, 2017

work page 2017
[4]

Detecting application-level failures in component-based internet services,

E. Kiciman and A. Fox, “Detecting application-level failures in component-based internet services,”IEEE transactions on neural net- works, vol. 16, no. 5, pp. 1027–1041, 2005

work page 2005
[5]

Failure prediction using machine learning and time series in optical network,

Z. Wang, M. Zhang, D. Wang, C. Song, M. Liu, J. Li, L. Lou, and Z. Liu, “Failure prediction using machine learning and time series in optical network,”Optics express, vol. 25, no. 16, pp. 18 553–18 565, 2017

work page 2017
[6]

A new approach for clustering alarm sequences in mobile operators,

S. Sozuer, C. Etemoglu, and E. Zeydan, “A new approach for clustering alarm sequences in mobile operators,” inNOMS 2016-2016 IEEE/IFIP Network Operations and Management Symposium. IEEE, 2016, pp. 1055–1060

work page 2016
[7]

A fault diagnosis method for wireless sensor network nodes based on a belief rule base with adaptive attribute weights,

K.-X. Shi, S.-M. Li, G.-W. Sun, Z.-C. Feng, and W. He, “A fault diagnosis method for wireless sensor network nodes based on a belief rule base with adaptive attribute weights,”Scientific Reports, vol. 14, no. 1, p. 4038, 2024

work page 2024
[8]

Automatic fault detection and diagnosis in cellular networks and beyond 5g: Intelligent network management,

A. K. Sangaiah, S. Rezaei, A. Javadpour, F. Miri, W. Zhang, and D. Wang, “Automatic fault detection and diagnosis in cellular networks and beyond 5g: Intelligent network management,”Algorithms, vol. 15, no. 11, p. 432, 2022

work page 2022
[9]

Ciciov2024: Advancing realistic ids approaches against dos and spoofing attack in iov can bus,

E. C. P. Neto, H. Taslimasa, S. Dadkhah, S. Iqbal, P. Xiong, T. Rahman, and A. A. Ghorbani, “Ciciov2024: Advancing realistic ids approaches against dos and spoofing attack in iov can bus,” Internet of Things, vol. 26, p. 101209, 2024. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S2542660524001501

work page 2024
[10]

Ciciomt2024: A benchmark dataset for multi-protocol security assessment in iomt,

S. Dadkhah, E. C. P. Neto, R. Ferreira, R. C. Molokwu, S. Sadeghi, and A. A. Ghorbani, “Ciciomt2024: A benchmark dataset for multi-protocol security assessment in iomt,”Internet of Things, vol. 28, p. 101351, 2024

work page 2024
[11]

Ciciot2023: A real-time dataset and benchmark for large-scale attacks in iot environment,

E. C. P. Neto, S. Dadkhah, R. Ferreira, A. Zohourian, R. Lu, and A. A. Ghorbani, “Ciciot2023: A real-time dataset and benchmark for large-scale attacks in iot environment,”Sensors, vol. 23, no. 13, 2023. [Online]. Available: https://www.mdpi.com/1424-8220/23/13/5941

work page 2023
[12]

Towards the development of a realistic multi- dimensional iot profiling dataset,

S. Dadkhah, H. Mahdikhani, P. K. Danso, A. Zohourian, K. A. Truong, and A. A. Ghorbani, “Towards the development of a realistic multi- dimensional iot profiling dataset,” in2022 19th Annual International Conference on Privacy, Security & Trust (PST), 2022, pp. 1–11

work page 2022
[13]

Ton iot telemetry dataset: A new generation dataset of iot and iiot for data-driven intrusion detection systems,

A. Alsaedi, N. Moustafa, Z. Tari, A. Mahmood, and A. Anwar, “Ton iot telemetry dataset: A new generation dataset of iot and iiot for data-driven intrusion detection systems,”Ieee Access, vol. 8, pp. 165 130–165 150, 2020

work page 2020
[14]

Graph neural network-based fault diagnosis: a review,

Z. Chen, J. Xu, C. Alippi, S. X. Ding, Y . Shardt, T. Peng, and C. Yang, “Graph neural network-based fault diagnosis: a review,”arXiv preprint arXiv:2111.08185, 2021

work page arXiv 2021
[15]

Open benchmarks for assessment of process monitoring and fault diagnosis techniques: A review and critical analysis,

A. Melo, M. M. C ˆamara, N. Clavijo, and J. C. Pinto, “Open benchmarks for assessment of process monitoring and fault diagnosis techniques: A review and critical analysis,”Computers & Chemical Engineering, vol. 165, p. 107964, 2022

work page 2022
[16]

Domain generalization for cross-domain fault diagnosis: An application-oriented perspective and a benchmark study,

C. Zhao, E. Zio, and W. Shen, “Domain generalization for cross-domain fault diagnosis: An application-oriented perspective and a benchmark study,”Reliability Engineering & System Safety, vol. 245, p. 109964, 2024. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0951832024000395

work page 2024
[17]

Towards a standard benchmarking framework for domain adaptation in intelligent fault diagnosis,

M. M. Farag, “Towards a standard benchmarking framework for domain adaptation in intelligent fault diagnosis,”IEEE Access, vol. 13, pp. 24 426–24 453, 2025

work page 2025
[18]

Information exposure from consumer iot devices: A multidimensional, network-informed measurement approach,

J. Ren, D. J. Dubois, D. Choffnes, A. M. Mandalari, R. Kolcun, and H. Haddadi, “Information exposure from consumer iot devices: A multidimensional, network-informed measurement approach,” in Proceedings of the Internet Measurement Conference, ser. IMC ’19. New York, NY , USA: Association for Computing Machinery, 2019, p. 267–279. [Online]. Available: http...

work page doi:10.1145/3355369.3355577 2019
[19]

A scheme for generating a dataset for anomalous activity detection in iot networks,

I. Ullah and Q. H. Mahmoud, “A scheme for generating a dataset for anomalous activity detection in iot networks,” inCanadian conference on artificial intelligence. Springer, 2020, pp. 508–520

work page 2020
[20]

Towards the development of realistic botnet dataset in the internet of things for network forensic analytics: Bot-iot dataset,

N. Koroniotis, N. Moustafa, E. Sitnikova, and B. Turnbull, “Towards the development of realistic botnet dataset in the internet of things for network forensic analytics: Bot-iot dataset,”Future Generation Computer Systems, vol. 100, pp. 779–796, 2019

work page 2019
[21]

An efficient self attention-based 1d-cnn-lstm network for iot attack detection and identi- fication using network traffic,

T. Sasi, A. H. Lashkari, R. Lu, P. Xiong, and S. Iqbal, “An efficient self attention-based 1d-cnn-lstm network for iot attack detection and identi- fication using network traffic,”Journal of Information and Intelligence, 2024

work page 2024
[22]

Benchmarking the benchmark—comparing synthetic and real-world network ids datasets,

S. Layeghy, M. Gallagher, and M. Portmann, “Benchmarking the benchmark—comparing synthetic and real-world network ids datasets,” Journal of Information Security and Applications, vol. 80, p. 103689, 2024

work page 2024
[23]

Device identification and anomaly detection in iot environments,

M. Rabbani, J. Gui, F. Nejati, Z. Zhou, A. Kaniyamattam, M. Mirani, G. Piya, I. Opushnyev, R. Lu, and A. A. Ghorbani, “Device identification and anomaly detection in iot environments,”IEEE Internet of Things Journal, vol. 12, no. 10, pp. 13 625–13 643, 2024

work page 2024
[24]

Care to compare: A real-world benchmark dataset for early fault detection in wind turbine data,

C. G ¨uck, C. M. A. Roelofs, and S. Faulstich, “Care to compare: A real-world benchmark dataset for early fault detection in wind turbine data,”Data, vol. 9, no. 12, 2024. [Online]. Available: https://www.mdpi.com/2306-5729/9/12/138 Junjian Zhang(Graduate Student Member, IEEE) received the B.S. degree in Electrical Engineering and Automation from Hunan Un...

work page 2024

[1] [1]

Towards automatic network fault local- ization in real time using probabilistic inference,

A. Johnsson and C. Meirosu, “Towards automatic network fault local- ization in real time using probabilistic inference,” in2013 IFIP/IEEE In- ternational Symposium on Integrated Network Management (IM 2013). IEEE, 2013, pp. 1393–1398

work page 2013

[2] [2]

Service-triggered failure identifica- tion/localization through monitoring of multiple parameters,

M. Ruiz, F. Fresi, A. P. Vela, G. Meloni, N. Sambo, F. Cugini, L. Poti, L. Velasco, and P. Castoldi, “Service-triggered failure identifica- tion/localization through monitoring of multiple parameters,” inECOC 2016; 42nd European Conference on Optical Communication. VDE, 2016, pp. 1–3. 11

work page 2016

[3] [3]

Fault detection in wireless sensor networks through svm classifier,

S. Zidi, T. Moulahi, and B. Alaya, “Fault detection in wireless sensor networks through svm classifier,”IEEE Sensors Journal, vol. 18, no. 1, pp. 340–347, 2017

work page 2017

[4] [4]

Detecting application-level failures in component-based internet services,

E. Kiciman and A. Fox, “Detecting application-level failures in component-based internet services,”IEEE transactions on neural net- works, vol. 16, no. 5, pp. 1027–1041, 2005

work page 2005

[5] [5]

Failure prediction using machine learning and time series in optical network,

Z. Wang, M. Zhang, D. Wang, C. Song, M. Liu, J. Li, L. Lou, and Z. Liu, “Failure prediction using machine learning and time series in optical network,”Optics express, vol. 25, no. 16, pp. 18 553–18 565, 2017

work page 2017

[6] [6]

A new approach for clustering alarm sequences in mobile operators,

S. Sozuer, C. Etemoglu, and E. Zeydan, “A new approach for clustering alarm sequences in mobile operators,” inNOMS 2016-2016 IEEE/IFIP Network Operations and Management Symposium. IEEE, 2016, pp. 1055–1060

work page 2016

[7] [7]

A fault diagnosis method for wireless sensor network nodes based on a belief rule base with adaptive attribute weights,

K.-X. Shi, S.-M. Li, G.-W. Sun, Z.-C. Feng, and W. He, “A fault diagnosis method for wireless sensor network nodes based on a belief rule base with adaptive attribute weights,”Scientific Reports, vol. 14, no. 1, p. 4038, 2024

work page 2024

[8] [8]

Automatic fault detection and diagnosis in cellular networks and beyond 5g: Intelligent network management,

A. K. Sangaiah, S. Rezaei, A. Javadpour, F. Miri, W. Zhang, and D. Wang, “Automatic fault detection and diagnosis in cellular networks and beyond 5g: Intelligent network management,”Algorithms, vol. 15, no. 11, p. 432, 2022

work page 2022

[9] [9]

Ciciov2024: Advancing realistic ids approaches against dos and spoofing attack in iov can bus,

E. C. P. Neto, H. Taslimasa, S. Dadkhah, S. Iqbal, P. Xiong, T. Rahman, and A. A. Ghorbani, “Ciciov2024: Advancing realistic ids approaches against dos and spoofing attack in iov can bus,” Internet of Things, vol. 26, p. 101209, 2024. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S2542660524001501

work page 2024

[10] [10]

Ciciomt2024: A benchmark dataset for multi-protocol security assessment in iomt,

S. Dadkhah, E. C. P. Neto, R. Ferreira, R. C. Molokwu, S. Sadeghi, and A. A. Ghorbani, “Ciciomt2024: A benchmark dataset for multi-protocol security assessment in iomt,”Internet of Things, vol. 28, p. 101351, 2024

work page 2024

[11] [11]

Ciciot2023: A real-time dataset and benchmark for large-scale attacks in iot environment,

E. C. P. Neto, S. Dadkhah, R. Ferreira, A. Zohourian, R. Lu, and A. A. Ghorbani, “Ciciot2023: A real-time dataset and benchmark for large-scale attacks in iot environment,”Sensors, vol. 23, no. 13, 2023. [Online]. Available: https://www.mdpi.com/1424-8220/23/13/5941

work page 2023

[12] [12]

Towards the development of a realistic multi- dimensional iot profiling dataset,

S. Dadkhah, H. Mahdikhani, P. K. Danso, A. Zohourian, K. A. Truong, and A. A. Ghorbani, “Towards the development of a realistic multi- dimensional iot profiling dataset,” in2022 19th Annual International Conference on Privacy, Security & Trust (PST), 2022, pp. 1–11

work page 2022

[13] [13]

Ton iot telemetry dataset: A new generation dataset of iot and iiot for data-driven intrusion detection systems,

A. Alsaedi, N. Moustafa, Z. Tari, A. Mahmood, and A. Anwar, “Ton iot telemetry dataset: A new generation dataset of iot and iiot for data-driven intrusion detection systems,”Ieee Access, vol. 8, pp. 165 130–165 150, 2020

work page 2020

[14] [14]

Graph neural network-based fault diagnosis: a review,

Z. Chen, J. Xu, C. Alippi, S. X. Ding, Y . Shardt, T. Peng, and C. Yang, “Graph neural network-based fault diagnosis: a review,”arXiv preprint arXiv:2111.08185, 2021

work page arXiv 2021

[15] [15]

Open benchmarks for assessment of process monitoring and fault diagnosis techniques: A review and critical analysis,

A. Melo, M. M. C ˆamara, N. Clavijo, and J. C. Pinto, “Open benchmarks for assessment of process monitoring and fault diagnosis techniques: A review and critical analysis,”Computers & Chemical Engineering, vol. 165, p. 107964, 2022

work page 2022

[16] [16]

Domain generalization for cross-domain fault diagnosis: An application-oriented perspective and a benchmark study,

C. Zhao, E. Zio, and W. Shen, “Domain generalization for cross-domain fault diagnosis: An application-oriented perspective and a benchmark study,”Reliability Engineering & System Safety, vol. 245, p. 109964, 2024. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0951832024000395

work page 2024

[17] [17]

Towards a standard benchmarking framework for domain adaptation in intelligent fault diagnosis,

M. M. Farag, “Towards a standard benchmarking framework for domain adaptation in intelligent fault diagnosis,”IEEE Access, vol. 13, pp. 24 426–24 453, 2025

work page 2025

[18] [18]

Information exposure from consumer iot devices: A multidimensional, network-informed measurement approach,

J. Ren, D. J. Dubois, D. Choffnes, A. M. Mandalari, R. Kolcun, and H. Haddadi, “Information exposure from consumer iot devices: A multidimensional, network-informed measurement approach,” in Proceedings of the Internet Measurement Conference, ser. IMC ’19. New York, NY , USA: Association for Computing Machinery, 2019, p. 267–279. [Online]. Available: http...

work page doi:10.1145/3355369.3355577 2019

[19] [19]

A scheme for generating a dataset for anomalous activity detection in iot networks,

I. Ullah and Q. H. Mahmoud, “A scheme for generating a dataset for anomalous activity detection in iot networks,” inCanadian conference on artificial intelligence. Springer, 2020, pp. 508–520

work page 2020

[20] [20]

Towards the development of realistic botnet dataset in the internet of things for network forensic analytics: Bot-iot dataset,

N. Koroniotis, N. Moustafa, E. Sitnikova, and B. Turnbull, “Towards the development of realistic botnet dataset in the internet of things for network forensic analytics: Bot-iot dataset,”Future Generation Computer Systems, vol. 100, pp. 779–796, 2019

work page 2019

[21] [21]

An efficient self attention-based 1d-cnn-lstm network for iot attack detection and identi- fication using network traffic,

T. Sasi, A. H. Lashkari, R. Lu, P. Xiong, and S. Iqbal, “An efficient self attention-based 1d-cnn-lstm network for iot attack detection and identi- fication using network traffic,”Journal of Information and Intelligence, 2024

work page 2024

[22] [22]

Benchmarking the benchmark—comparing synthetic and real-world network ids datasets,

S. Layeghy, M. Gallagher, and M. Portmann, “Benchmarking the benchmark—comparing synthetic and real-world network ids datasets,” Journal of Information Security and Applications, vol. 80, p. 103689, 2024

work page 2024

[23] [23]

Device identification and anomaly detection in iot environments,

M. Rabbani, J. Gui, F. Nejati, Z. Zhou, A. Kaniyamattam, M. Mirani, G. Piya, I. Opushnyev, R. Lu, and A. A. Ghorbani, “Device identification and anomaly detection in iot environments,”IEEE Internet of Things Journal, vol. 12, no. 10, pp. 13 625–13 643, 2024

work page 2024

[24] [24]

Care to compare: A real-world benchmark dataset for early fault detection in wind turbine data,

C. G ¨uck, C. M. A. Roelofs, and S. Faulstich, “Care to compare: A real-world benchmark dataset for early fault detection in wind turbine data,”Data, vol. 9, no. 12, 2024. [Online]. Available: https://www.mdpi.com/2306-5729/9/12/138 Junjian Zhang(Graduate Student Member, IEEE) received the B.S. degree in Electrical Engineering and Automation from Hunan Un...

work page 2024