Enhancing 5G and IoT network security
Nishanth Gadey, Sagar Dhanraj Pande, Aditya Khamparia · 2024
The Internet of Things (IoT) is expanding quickly, and their incorporation into daily life has made it possible for unprecedented connectedness and creativity. Simultaneously, the rapid expansion of 5G networks brings the promise of enhanced communication. However, this transformative landscape is accompanied by a pressing array of security challenges, demanding innovative solutions. In response, this research places paramount importance on fortifying the security of IoT networks while extending its protective umbrella to encompass 5G networks, all through a comprehensive multi-model deep learning Intrusion detection approach for attack classification. This approach leverages the extensive CICIoT2023 dataset to provide a robust security framework. Harnessing the capabilities of Artificial Neural Networks (ANNs), Recurrent Neural Networks (RNNs), Deep Neural Networks (DNNs), and strategically integrating Random Forest to extract the top 40 features, our research ensures a holistic understanding of network traffic patterns. Subsequently, we apply ANNs, DNNs, RNNs, and our proposed model, achieving an impressive maximum accuracy of 98.45%. This research emphasizes the effectiveness of deep learning in safeguarding IoT networks and extending this protection to 5G networks. This multifaceted approach not only addresses the intricate security requirements of IoT networks but also seamlessly applies these principles to secure the evolving landscape of 5G communications. By prioritizing IoT security and bridging the domains of IoT and 5G, this study aids in the continuous mission to protect the interconnected future of smart devices and high-speed communication networks.