ISSN: 2182-2069 (printed) / ISSN: 2182-2077 (online)
Improving Wireless Network Intrusion Detection with Deep Feature Framework
The rapid adoption of Internet-of-Things (IoT) applications has raised concerns about the security of IoT communication systems. The escalating number of harmful attacks poses severe threats, leading to network disruptions and system failures. In this research, we propose a novel solution called the Frog Leap-based Hyper-parameter Tuned Deep Neural (FLbHTDN) model, designed to effectively detect intrusions in IoT communication networks.To validate the effectiveness of the FLbHTDN model, we employ the Nsl-Kdd dataset and conduct comprehensive data preprocessing to remove errors and extract features related to various attack classes. The FLbHTDN model is implemented on the Java platform, and its performance is compared against existing intrusion detection techniques.Our findings demonstrate that the FLbHTDN model outperforms state-of-the-art methods, exhibiting superior attack prediction scores and significantly reduced processing times. This underscores the significance of advanced intrusion detection methods in securing IoT communication systems.The key contributions of this research lie in the development of the FLbHTDN model, which provides a robust defense against evolving cyber threats, ensuring a safer and more reliable IoT ecosystem. As the IoT landscape continues to evolve, the FLbHTDN model sets a strong foundation for proactive security measures, offering a promising avenue for safeguarding IoT networks from malicious activities.