AI-Driven Optimization in Mobile Industrial Wireless Networks: A Performance and Security Enhancement Framework Peter Oluwaseun Adepoju UG Scholar, Department of Computer Engineering Halic University Istanbul, Turkey peteradepoju07@gmail.com Mohammed Al-Hubaishi Assistant Professor, Department of Computer Engineering Halic University Istanbul, Turkey mohammedhbi@gmail.com Abstract—This study investigates integrating artificial intelli- gence (AI) techniques in mobile industrial wireless communi- cation networks, focusing on enhancing network performance, reliability, and security in industrial environments. Through a comprehensive analysis of industrial testbed datasets (iV2V and iV2I+), we demonstrate the effectiveness of AI-driven approaches in optimizing network operations. We implemented multiple machine learning models, including Decision Trees, Random Forests, and LightGBM, to predict quality of service (QoS) and optimize link selection. The Decision Tree Regressor achieved excellent performance, demonstrating its effectiveness for indus- trial wireless applications. We introduce a DBSCAN clustering- based approach for identifying weak signal regions, enabling proactive network optimization and coverage enhancement. The study also evaluates location fingerprinting models, achieving high accuracy in source classification through feature importance analysis. Our results emphasize the practical benefits of AI inte- gration in industrial wireless networks, particularly in predictive maintenance, real-time decision-making, and network security. The findings provide valuable information for implementing AI- driven solutions in Industry 4.0 applications. However, the study acknowledges current limitations and emphasizes the need for real-world validation in industrial operations. Index Terms—Artificial Intelligence (AI), Mobile Industrial Wireless Communication Networks, Machine Learning, Network Performance, Predictive Maintenance, Density-Based Spatial Clustering of Applications with Noise (DBSCAN) I. I NTRODUCTION Artificial intelligence (AI) is emerging as a paradigm- shifting technology in mobile industrial wireless communica- tion networks, enhancing efficiency, reliability, and security. Data analysis techniques of AI, such as machine learning (ML), optimize network performance, reduce congestion, and enable predictive maintenance by analyzing large datasets [1], [2]. Advanced intrusion detection systems, which combine misuse and anomaly detection models, are among the key ben- eficiaries of AI, significantly improving security [2]. Although a promising approach, the step towards adaptive, data-driven algorithms is still in its early stages, and further research is required [3], [4]. However, challenges such as distribution management, scal- ability, and security issues remain. Computational cost, high- quality datasets, and robustness against adversarial attacks are major challenges faced by AI models [5], [6]. Many new solu- tions are studied, adapting modern communication paradigms such as terahertz or quantum communication and AI resource allocation algorithms in 5G and beyond protocols [3], [7]. Solving these issues would result in more reliable, efficient, secure industrial wireless systems [8], [9]. As one of the most promising technologies, AI is improving industrial systems’ efficiency, reliability, and security. Using AI-driven methods such as machine learning for predictive maintenance, real-time decision-making, and resource optimization can significantly broaden the scope of the research on AI for Mobile Industrial Wireless Communication Networks [10], [11]. Fig. 1. Overview of System Structure The primary contributions of the work presented in this paper are as follows: • Data Ingestion: Loading and integrating the iV2V and iV2I+ datasets into the analysis pipeline. • Data Preprocessing: Cleaning, normalizing, and engi- neering features to prepare the data for analysis. Proceedings of the International Conference on Intelligent Computing and Control Systems (ICICCS-2025) IEEE Xplore Part Number: CFP25K74-ART; ISBN: 979-8-3315-1208-8 979-8-3315-1208-8/25/$31.00 ©2025 IEEE 499 2025 International Conference on Intelligent Computing and Control Systems (ICICCS) | 979-8-3315-1208-8/25/$31.00 ©2025 IEEE | DOI: 10.1109/ICICCS65191.2025.10984569 Authorized licensed use limited to: ULAKBIM UASL - Halic Universitesi. Downloaded on May 15,2025 at 09:14:06 UTC from IEEE Xplore. Restrictions apply.