Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM68794
Volume:
10 |
11 |
Nov 2026
Pages:
1-9
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
computer science
Sri Ramakrishna College of Arts and Science for women, Coimbatore
Assistant professor ,School of computer science
Sri Ramakrishna College of Arts and Science for women, Coimbatore
Modern communication networks support cloud services, enterprise applications, Internet of Things (IoT) deployments and latency-sensitive digital services. Their scale and dynamic behaviour make manual monitoring and rule-based fault management increasingly difficult. Network faults such as link degradation, congestion, routing instability, device failures and abnormal traffic can propagate across interconnected components and cause service disruption. This article presents a deep learning-oriented framework for intelligent network management and automated fault detection. The proposed framework combines continuous telemetry collection, data preprocessing, temporal and spatial feature learning, anomaly classification, fault localisation and controlled remediation. A hybrid model design is discussed in which a temporal neural network, such as a Long Short-Term Memory (LSTM) network, learns changes in time-series performance indicators, while a one-dimensional Convolutional Neural Network (1D-CNN) extracts local patterns from multivariate telemetry windows. An optional autoencoder supports anomaly detection when labelled fault examples are limited. The framework is designed to integrate with network management systems through telemetry interfaces and an event-driven decision layer. The article also outlines a reproducible evaluation methodology using public network intrusion and traffic datasets alongside a separately collected operational telemetry dataset. Evaluation measures include precision, recall, F1-score, false alarm rate, detection latency, localisation accuracy and resource overhead. No experimental performance values are claimed because the framework is proposed for future implementation and empirical validation. The discussion identifies key deployment challenges, including data imbalance, concept drift, explainability, privacy, model robustness and safe automation. The proposed approach provides a structured foundation for moving from reactive monitoring toward predictive, data-driven and human-supervised network operations.
KEYWORDS
This article proposed a deep learning-based framework for intelligent network management and automated fault detection. The framework combines telemetry collection, data preparation, CNN and LSTM feature learning, autoencoder-based anomaly scoring, topology-aware event correlation, fault localisation and policy-governed response. It emphasises that operational network management requires more than high classification accuracy: detection latency, false alarm burden, localisation quality, resource overhead, explainability and safe recovery must also be assessed. The research methodology outlines how public traffic datasets can be complemented with controlled operational telemetry and injected infrastructure faults. It also recommends chronological evaluation, device- or scenario-held-out testing, baseline comparison and ablation analysis. Since no implementation or empirical experiment is reported here, the framework should be treated as a research proposal requiring validation. Future work should implement the architecture in a representative testbed, publish dataset and preprocessing details, compare model families under consistent conditions and evaluate human-supervised automation. Such work can help establish whether deep learning can provide reliable, scalable and accountable support for next-generation network operations.
[1] F. M. Anis, M. Alabdullatif, S. Aljbli, and M. Hammoudeh, “A Survey on the Applications of Deep Learning in Network Intrusion Detection Systems to Enhance Network Security,” IEEE Access, vol. 13, pp. 185357–185373, 2025. doi: 10.1109/ACCESS.2025.3624952. [2] V. G. da Silva Ruffo, D. M. B. Lent, M. Komarchesqui, V. F. Schiavon, M. V. O. de Assis, L. F. Carvalho, and M. L. Proença Jr., “Anomaly and intrusion detection using deep learning for software-defined networks: A survey,” Expert Systems with Applications, vol. 256, article 124982, 2024. doi: 10.1016/j.eswa.2024.124982. [3] “Unknown, Atypical and Polymorphic Network Intrusion Detection: A Systematic Survey,” IEEE Transactions on Network and Service Management, vol. 21, no. 1, pp. 1190–1212, 2024. doi: 10.1109/TNSM.2023.3298533. [4] International Telecommunication Union, “Recommendation ITU-T Y.3177: Architectural framework for artificial intelligence-based network automation for resource and fault management in future networks including IMT-2020,” Feb. 2021. Available: https://www.itu.int/rec/T-REC-Y.3177-202102-I [5] International Telecommunication Union, “Recommendation ITU-T Y.3061: Autonomous networks—Architecture framework,” Dec. 2023. Available: https://www.itu.int/rec/T-REC-Y.3061-202312-I [6] A. Momand, S. U. Jan, and N. Ramzan, “A Systematic and Comprehensive Survey of Recent Advances in Intrusion Detection Systems Using Machine Learning: Deep Learning, Datasets, and Attack Taxonomy,” Journal of Sensors, vol. 2023, article 6048087, 2023. doi: 10.1155/2023/6048087. [7] Z. Xu, Y. Wu, S. Wang, J. Gao, T. Qiu, Z. Wang, H. Wan, and X. Zhao, “Deep Learning-based Intrusion Detection Systems: A Survey,” arXiv:2504.07839, 2025. Available: https://arxiv.org/abs/2504.07839. [8] “Network anomaly detection and security defense technology based on machine learning: A review,” Computers & Electrical Engineering, vol. 119, article 109581, 2024. doi: 10.1016/j.compeleceng.2024.109581.
Publication Date
11/11/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
1
Total views since publication
PDF Downloads
5
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...