AI/ML-Enhanced Security Monitoring for 5G-Enabled Big Data Sensor Networks

• Main Authors: Da Silva, Stefania E. Iza, Rodriguez, Demostenes Z., Rosa, Renata L., Adasme, Pablo, Saadi, Muhammad
• Format: Conference Proceeding
• Language: English
• Subjects: 5G network; Accuracy; Analytical models; Artificial Intelligence; Big Data; Data mining; Data models; Long short term memory; Machine Learning; Monitoring; Scalability; Security; Security Monitoring; Sensor Networks; Spatial databases; Throughput
• ISSN: 1847-358X
• DOI: 10.23919/SoftCOM62040.2024.10721857

The integration of 5G technology with sensor network systems has enabled extensive connectivity and data exchange between cyber and physical entities. However, this connectivity brings significant challenges in managing vast amounts of data, including issues related to data duplication, scalability, and security. In this paper, we propose an advanced AI/MLbased algorithm for enhancing the security of 5 G -enabled sensor networks within a big data environment. Our approach leverages both geographical and temporal data to improve the extraction of critical information from multidimensional time-series data, thereby accurately modeling and analyzing attack patterns. The proposed method achieved an overall detection accuracy of \mathbf{9 3. 3 \%}, demonstrating a good performance in detecting cyber threats. Additionally, the model exhibited scalability by maintaining high throughput with increasing numbers of sensor nodes. The results ensure the safety, reliability, and scalability of 5G-enabled sensor networks in real-world scenarios.