EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Enhancing IoT and cloud Infrastructures Security with an AI-driven Hybrid Intrusion Detection System IDS for Encrypted Next Generation Network Traffic Using Deep Learning

Muhammad Waleed Khawar, Atif Ali, Fatima Zafar, Asad Khalid Khan, Samra Shaheen, Beenish Iftikhar

The Asian Bulletin of Big Data Management · 31.03.2026 · 6 (1) · 535-566

DOI: 10.62019/2akvtc64 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Asian Bulletin of Big Data Management, online erschienen am 31.03.2026. Verfasst von Muhammad Waleed Khawar und 5 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Next generation networks (NGNs) have brought in new opportunities and new security challenges with the advent of 5G, edge computing, IoT and SDN. The signature and anomaly-based traditional intrusion detection/p prevention (IDS) technology cannot cope with the volume, variety and evolving nature of the threats that can be found in the NGN. Artificial Intelligence (AI) has proven to be a game-changer for modern IDPS.In turn, Artificial Intelligence (AI) has proved to be a game-changer for modern IDPS. It highlights some of the classical machine learning techniques like decision trees, support vector machines, random forest, and then discusses deep learning architectures like convolutional neural networks (CNNs), recurrent neural networks (RNNs), long short-term memory networks (LSTMs) and autoencoders. It also elaborates on the hybrid approaches that blend ensemble learning, federated learning and meta-learning, and tailored applications for SDN, IoT, edge and cloud, and 5G. The paper identifies some of the challenges at play today, including paucity of data, generalisation gap, computational complexity, adversarial robustness, explainability, and privacy. The future directions are towards making the dataset of NGN realistic, lightweight and accurate architectures, frameworks ensuring the privacy and integrated detection and prevention mechanisms. As a whole, AI-driven IDPS have great promise in helping to safeguard NGNs, but achieving this will need progress in accuracy, efficiency, explainability, and resilience. The high rate of adoption of encrypted network communication has hugely diminished the quality of the traditional Intrusion Detection Systems (IDS), which are based on payload examination to identify bad actions. This paper will suggest a hybrid IDS based on deep learning, which combines the Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks to identify intrusions in encrypted traffic based on statistical and behavioral features. The evaluation of the proposed model is conducted on the basis of the CICIDS2017 dataset, where preprocessing techniques like data cleaning, normalization, and selection of features are used. As shown in the experiment, the hybrid CNN-LSTM model is effective in capturing spatial and temporal patterns of network traffic and has a better detection performance than the traditional machine learning models. The model attains good generalization and has a reliable classification. Nevertheless, the higher complexity of the model in terms of computation necessitates optimization in resource-constrained environments. In general, the proposed solution is solid and smart for intrusion detection in contemporary encrypted network systems.

Zitieren

APA

Khawar, M. W., Ali, A., Zafar, F., Khan, A. K., Shaheen, S., & Iftikhar, B. (2026). Enhancing IoT and cloud Infrastructures Security with an AI-driven Hybrid Intrusion Detection System IDS for Encrypted Next Generation Network Traffic Using Deep Learning. The Asian Bulletin of Big Data Management, 6(1), 535-566. https://doi.org/10.62019/2akvtc64

Vancouver

Khawar MW, Ali A, Zafar F, Khan AK, Shaheen S, Iftikhar B. Enhancing IoT and cloud Infrastructures Security with an AI-driven Hybrid Intrusion Detection System IDS for Encrypted Next Generation Network Traffic Using Deep Learning. The Asian Bulletin of Big Data Management. 2026;6(1):535-566. doi:10.62019/2akvtc64

BibTeX
@article{Khawar2026,
  author = {Khawar, Muhammad Waleed and Ali, Atif and Zafar, Fatima and Khan, Asad Khalid and Shaheen, Samra and Iftikhar, Beenish},
  title = {Enhancing IoT and cloud Infrastructures Security with an AI-driven Hybrid Intrusion Detection System IDS for Encrypted Next Generation Network Traffic Using Deep Learning},
  journal = {The Asian Bulletin of Big Data Management},
  year = {2026},
  volume = {6},
  number = {1},
  pages = {535--566},
  doi = {10.62019/2akvtc64}
}
Kostenlos starten →