EUVIMEDEUVIMED entdecken
Crossref

Enhanced Cobra Optimization for Lightweight Intrusion Detection in Edge-IoT Networks

Huda Ali Mahdi

Journal of Al-Farabi for Engineering Sciences · 02.06.2026 · 5 (1) · 108-121

DOI: 10.59746/7q2q6r19 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Al-Farabi for Engineering Sciences, online erschienen am 02.06.2026. Verfasst von Huda Ali Mahdi.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The rapid proliferation of Internet of Things (IoT) ecosystems has led to the generation of massive, heterogeneous, and noisy data streams from distributed sensors, posing significant challenges for intrusion detection. Conventional machine learning models, such as Support Vector Machines (SVM) and k-Nearest Neighbors (k-NN), rely on handcrafted and static feature representations, limiting their ability to capture complex and dynamic IoT attack patterns. To address these limitations, this paper proposes an Enhanced Cobra Optimization Algorithm (E-COA) integrated within a hybrid edge–cloud deployment framework for efficient and real-time IoT intrusion detection. The proposed E-COA introduces a non-linear adaptive hunting probability to dynamically balance exploration and exploitation, along with a composite fitness function that jointly optimizes classification accuracy, false positive rate (FPR), and feature sparsity. Experiments are conducted on the MQTT-IoT-IDS2020 dataset, which consists of 71,341 labeled network traffic samples across five classes. The proposed model achieves a classification accuracy of 97.2%, a macro-averaged F1-score of 0.971, and a false positive rate of 2.2%. It also demonstrates superior performance compared to SVM, Random Forest, 1D-CNN, PSO, and standard COA models. Furthermore, the optimized model maintains a compact size of approximately 75 KB and achieves an average inference latency of 4.3 ms on edge-simulated hardware. These results indicate that the proposed framework effectively balances detection accuracy, computational efficiency, and suitability for deployment in resource-constrained edge-IoT environments.

Zitieren

APA

Mahdi, H. A. (2026). Enhanced Cobra Optimization for Lightweight Intrusion Detection in Edge-IoT Networks. Journal of Al-Farabi for Engineering Sciences, 5(1), 108-121. https://doi.org/10.59746/7q2q6r19

Vancouver

Mahdi HA. Enhanced Cobra Optimization for Lightweight Intrusion Detection in Edge-IoT Networks. Journal of Al-Farabi for Engineering Sciences. 2026;5(1):108-121. doi:10.59746/7q2q6r19

BibTeX
@article{Mahdi2026,
  author = {Mahdi, Huda Ali},
  title = {Enhanced Cobra Optimization for Lightweight Intrusion Detection in Edge-IoT Networks},
  journal = {Journal of Al-Farabi for Engineering Sciences},
  year = {2026},
  volume = {5},
  number = {1},
  pages = {108--121},
  doi = {10.59746/7q2q6r19}
}
Kostenlos starten →