Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
May 18, 2026Cureus Journal of Computer Science.Open Access

Performance Enhancement of Supervisory Control and Data Acquisition (SCADA) IEC 60870-5-104 Intrusion Detection Using Sequence-Aware and Hybrid Deep Learning Models: A Comparative Evaluation

View Full Paper
Ask AI
Bookmark
Share

Authors

MAM. Agus Syamsul ArifinDSDeris StiawanSSSusanto Susanto

Discussion

Loading...

Member takes

Overview

Randomized trial evaluates performance of deep learning models in detecting attacks in SCADA systems, suggesting optimal cybersecurity solutions.

Key Points

  • This study compares deep learning models for effective intrusion detection in SCADA IEC 60870-5-104 networks.
  • Analyzed various deep learning architectures including BiLSTM and CNN-based hybrids.
  • Utilized a dataset from a realistic SCADA testbed comprising benign and attack traffic.
  • Evaluated model performance using accuracy, precision, recall, F1-score, and confusion matrices.
  • BiLSTM and LSTM-GRU models achieved the highest detection rates across most attack classes.
  • CNN-based hybrids performed poorly for certain attack types like ICMP flood, despite high accuracy.
  • Overall, sequence-based deep learning models showed superior effectiveness for SCADA cybersecurity.

Cite This Study

Arifin et al. (2026) studied this question.

synapsesocial.com/papers/6a0aabf55ba8ef6d83b6f8b2https://doi.org/10.7759/s44389-026-00076-0
View Full Paper
Ask AI
Bookmark
Share