An Explainable Hybrid CNN–BiLSTM–Attention Network for Intelligent Intrusion Detection in Secure 6G Network

Prince Kumar; Neeraj Kumar Kamal

doi:10.36948/ijfmr.2026.v08i05.89354

An Explainable Hybrid CNN–BiLSTM–Attention Network for Intelligent Intrusion Detection in Secure 6G Network

Author(s)	Mr. Prince Kumar, Dr. Neeraj Kumar Kamal
Country	India
Abstract	Wireless networks and network slicing are undergoing great transformation at sixth generation (6G) ushering in unprecedented flexibility, ultra-low latency and smart provisioning of services in next-generation communication systems. But, the progress has also opened the cyber-attack surface and network slices are increasingly being targeted by increasingly sophisticated and evolving attacks. However, traditional IDSs are incapable of handling complex attacks because of their limited feature extraction ability, poor contextual learning, and lack of model interpretability, which limits their use in security-critical 6G networks. To overcome these challenges, this paper introduces a new Explainable Deep Learning (XDL) based intelligent intrusion detection system (IDS) for secure 6G network slicing, namely XHCBAN (Explainable Hybrid CNN–BiLSTM–Attention Network). The proposed model combines Convolutional Neural Networks (CNNs) to model hierarchical spatial features, Bidirectional Long Short-Term Memory (BiLSTM) networks to capture bidirectional contextual relations, and an Attention Mechanism to focus on the most discriminative security features. Moreover, to obtain transparent and interpretable predictions, SHapley Additive exPlanations (SHAP) are integrated to measure the effect of individual features on the model's predictions. The proposed framework includes detailed methodology such as Data Processing, Feature Learning, Attention guided Feature refinement, Deep Classification, and Explainability Analysis. Standard metrics such as Accuracy, Precision, Recall, F1-score, Specificity, Matthews Correlation Coefficient (MCC), and Area Under the ROC Curve (AUC) are used to evaluate model performance and compared to the conventional machine learning and deep learning baselines using SVM, Random Forest, CNN, LSTM, and BiLSTM under a consistent experimental protocol. The proposed XHCBAN framework is believed to provide better intrusion detection accuracy and performance with a strong improvement in model transparency, robustness and reliability. This work introduces a scalable, Explainable Artificial Intelligence (XAI)-enabled intrusion detection system for secure AI-powered 6G network slicing, which fosters next-generation intelligent communication infrastructures and paves the way for practical deployment of explainable cybersecurity systems.
Keywords	6G Network Slicing; Intrusion Detection System (IDS); Explainable Artificial Intelligence (XAI); SHAP; Hybrid Deep Learning; CNN–BiLSTM–Attention Network; Cybersecurity; Intelligent Network Security
Field	Computer > Network / Security
Published In	Volume 8, Issue 5, September-October 2026
Published On	2026-10-11
DOI	https://doi.org/10.36948/ijfmr.2026.v08i05.89354

View / Download PDF File

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

An Explainable Hybrid CNN–BiLSTM–Attention Network for Intelligent Intrusion Detection in Secure 6G Network

Share this