AI ENABLED FIREWALL USING TWO METRICS ATRS AND RAA | IJCT Volume 12 – Issue 5 | IJCT-V12I5P60

Traditional firewalls struggle against advanced cyber threats such as zero-day attacks, encrypted malicious traffic, and polymorphic malware. This study proposes an AI-powered firewall utilizing Adaptive Threat Response Score (ATRS) and Risk Adaptive Access (RAA) to enhance threat detection and adaptive access control. The system employs machine learning classifiers trained on benchmark datasets like NSL-KDD and CICIDS2017, combined with explainable AI techniques for interpretable decision-making. Virtual lab experiments simulating attacker-defender scenarios demonstrate detection accuracy exceeding 99.5%, minimal false positives, and effective real-time risk assessment. The results highlight the potential of AI- enabled firewalls to provide robust, transparent, and adaptive network security.

In the future, the AI-powered firewall can be tested in real organizational or cloud networks to check how well it handles heavy and encrypted data traffic. It can be improved by adding continuous learning so that it automatically updates itself against new and advanced cyberattacks. The system can also be optimized to run faster and use fewer resources, making it suitable for large-scale use. By linking it with other security systems and analyzing user behavior, it can provide stronger, smarter, and more adaptive network protection.

[1]Ahmad, W. (2024). Enhancing network defense systems through intelligent security frameworks: A study on next-generation protective mechanisms [Master’s thesis, Tampere University]. [2]Ahmadi, S. (2023). Evaluation of AI-assisted security systems for modern network protection. International Journal of Computing Studies, December, 1–14. [3]Alsaqour, R., Motmi, A., & Abdelhaq, M. (2021). Assessing deployment strategies for network protection tools. Journal of Network Security and Data Protection, April. [4]Fatima, Z., Hussain, R., Dilshad, A., Shaikh, M. S., & Laghari, A. A.(2025). Transparent decision support in cybersecurity: Explainable intelligent security modules. Cybersecurity and Privacy Review (Wiley), 8(3), April. [5]Hakim, A., Aini, N., Khan, S. B. A., & Faizal, M. (2024). Integrated approaches to digital security using monitoring, encryption, and layered access controls. July. [6]He, X. (2020). Strategies for maintaining robust network defenses using protective technologies. Journal of Physics: Conference Series – MACE 2020, October. [7]Krishna Prakasha, K., Reddy, Y. R. M., Pallakonda, A., Amar, R. D., & Mrudula, P. S. (2025). Optimized intrusion prevention in smart networks with privacy-centric and decoy-based methods. Wiley Cybersecurity Access Journal, July. [8]Malik, A., Arshid, K., Noonari, N., & Munir, R. (2025). Predictive threat mitigation in computer networks using intelligent analytics. Scholars Journal of Engineering and Technology, June a. . [9]Patel, U. (2024). Modernizing digital protection via advanced network security mechanisms. International Journal of Engineering and Technology Research, August. [10]Phour, A., & Khurana, D. (2025). Hybrid intelligent algorithms for complex threat recognition in next-gen systems. Journal of Innovative Science and Technology Engineering, July. [11]Purvant, S., & Sowmya, K. S. (2024). A technical review of contemporary network defense strategies and practices. International Journal of Research Publications, January. [12]Soureya, Y. G., Ngossaha, J. M., Djomadji, E. M. D., Amougou, N., Tsakou, S. B., & Ndjodo, M. F. (2025). Designing adaptive security monitoring frameworks using sustainable methodological approaches. Journal of Computer Systems and Communications, July. [13]Wang, P. (2022). Impact of protective technologies on secure network infrastructure design. Frontiers in Computing and Intelligent Networks, August. [14]Nabi, A. U., Ahmed, M., & Abro, A. A. (2022). Classification and performance assessment of network security tools and functionalities. International Journal of Advanced Computing Systems, July. [15] Kumar, P., Rani, K., & Kumar, P. (2023). Evaluating efficiency and challenges of network defense systems. International Journal of Contemporary Science, August. [16]Ahmad, W. (2025). Dynamic intelligent network defense for real- time threat response. International Research Journal on Advanced Engineering, 3(3), March. [17]Subrahmanyam, V., & Prasad, M. V. S. (2025). Adaptive defense strategies in modern networks using intelligent threat detection. International Journal of Science and Research, 14(9), September a. . [18]Mingze, M. (2024). Analysis of network logs and threat detection using feature selection techniques. IET Research Journal, May. [19]Akamai Technologies. (2025). Advanced protection solutions for AI-integrated applications. Press Release, April [20]IRJA Communications. (2025). Layered security strategies for next-generation network protection. World Journal of Advanced Research Reviews. [21]IJCert Editorial Team. (2025). Frameworks for adaptive monitoring, packet screening, and anomaly recognition. International Journal of Computer Engineering and Research Technologies, March. [22]Mingze, M., et al. (2024). Cloud security enhancement using generative intelligent systems in modern networks. Online Scientific Research Journal, December. [23]Hossain, R., & Malik, S. (2025). Transparent evaluation of cybersecurity systems using explainable intelligent models. Wiley Cybersecurity and Privacy, October. [24]Fortinet. (2024). Emerging trends in digital threat protection through intelligent solutions. White Paper, December. [25]SNS Insider. (2023). Market trends and future projections for next- generation network protection. Market Analysis Report. [26]Akamai Technologies. (2025). Akamai Firewall for AI Enables Secure AI Applications with Advanced Threat Protection. Press Release, April. [27]Palo Alto Networks. (2023). Next-Generation Firewalls: Securing Modern Networks with AI and Deep Learning. Industry Whitepaper. [28]Mingze, M., et al. (2025). Machine learning-inspired intrusion detection system for IoT environments. Elsevier, October. [29]Check Point. (2025). AI Security Report—Trends and Threats in Next-Generation Firewall Technologies. Industry Analysis. ScienceDirect Editorial Team. (2025). A comprehensive review of explainable AI in cybersecurity: Models, limits, applications. ScienceDirect, October.

Call for Paper (Hub) Fast Paper Publication IJCT Indexing & Databases Submit Manuscript Call for Paper: Sep–Oct 2025 Call for Paper: Publish Research