EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Blockchain-Enabled Security Framework for Cloud-Based Sensor Systems with Deep Learning-Driven Attack Classification

Naveed Ahmad iD, Yue Cao iD, William Liu iD

Sensors · 17.08.2026 · 26 (16) · 5198

DOI: 10.3390/s26165198 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 17.08.2026. Verfasst von Naveed Ahmad und 2 weiteren Autor:innen. Beteiligte Einrichtung: College of Computer & Information Sciences, Prince Sultan University, Riyadh 12435, Saudi Arabia. Die Arbeit nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Prince Sultan University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Cloud-integrated sensor and Internet of Things (IoT) systems enable scalable data storage, processing, and intelligent monitoring, but their distributed nature exposes network-flow and host-level data to unauthorized access, tampering, and cyberattacks. This study proposes a Weighted Symmetric Hashed Blockchain (WSHB) framework that integrates mutual-information-based feature weighting, deep-learning-based attack classification, AES-256-GCM authenticated encryption, cryptographic hashing, and permissioned-ledger logging. The framework was evaluated independently using HIKARI-2021 for network-based intrusion detection and ADFA-LD for host-based intrusion detection. A transparent comparative evaluation was conducted against LSTM and CNN–RNN baselines using identical data splits, preprocessing settings, input representations, hyperparameter-search budget, and repeated initialization seeds. The proposed WSHB-DNN classifier achieved macro-F1 scores of 0.6837 on HIKARI-2021 and 0.7914 on ADFA-LD, showing the strongest overall classification performance among the evaluated models. The cryptographic and permissioned-ledger components provide confidentiality protection, record-level authentication, integrity verification, and tamper-evident logging for confirmed attack-event records. These results demonstrate the potential of WSHB as a reproducible framework for attack classification and secure event logging in cloud-integrated sensor environments.

Schlagwörter

Stichwörter der Autor:innen

blockchaincloud computingcyber securitydeep learninghashingsymmetric cryptographyweighted

Autor:innen und Institutionen

  • College of Computer & Information Sciences, Prince Sultan University, Riyadh 12435, Saudi Arabia.
  • School of Cyber Science and Engineering, Wuhan University, Wuhan 430072, China.
  • School of Computing, Electrical and Applied Technologies, Unitec Institute of Technology, Auckland 1025, New Zealand.

ORCID-Profile: Naveed Ahmad, Yue Cao, William Liu

Förderung

  • Prince Sultan University

Zitieren

APA

Ahmad, N., Cao, Y., & Liu, W. (2026). A Blockchain-Enabled Security Framework for Cloud-Based Sensor Systems with Deep Learning-Driven Attack Classification. Sensors, 26(16), 5198. https://doi.org/10.3390/s26165198

Vancouver

Ahmad N, Cao Y, Liu W. A Blockchain-Enabled Security Framework for Cloud-Based Sensor Systems with Deep Learning-Driven Attack Classification. Sensors. 2026;26(16):5198. doi:10.3390/s26165198

BibTeX
@article{Ahmad2026,
  author = {Ahmad, Naveed and Cao, Yue and Liu, William},
  title = {A Blockchain-Enabled Security Framework for Cloud-Based Sensor Systems with Deep Learning-Driven Attack Classification},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {16},
  pages = {5198},
  doi = {10.3390/s26165198}
}
Kostenlos starten →