Mahalanobis City Block distance-rider Optimization Algorithm-Based Neural Network enabled intrusion detection in the Internet of Things
T. Manojkumar · Journal of Networking and Communication Systems (JNACS) · 2024
The Internet of Things (IoT) has great potential for a variety of applications ranging from healthcare automation to defensive networks.The security of IoT networks is substantially paramount to the security of underlying communication and computing infrastructure.However, owing to less constrained resources and computation capacities, the IoT network is highly prone to several attacks.Therefore, protecting IoT networks from various attacks is of vital importance.One of the significant security systems is the Intrusion Detection System (IDS).In this research, Mahalanobis City Block distance-Rider Optimization Algorithm-Based Neural Network (MCB distance-RideNN) is presented for intrusion detection in IoT.The system model of IoT is simulated initially and the communication is accomplished with each IoT node.Thereafter, data is stored in the form of features on a server.Next, the intrusion detection process is carried out by initially obtaining input log file data from the NSL-KDD dataset and thereafter, Min-Max normalization is employed to conduct data normalization.After that, feature selection is conducted based on the Mahalanobis City Block (MCB) distance.However, MCB distance is newly presented by integrating Mahalanobis distance and City Block distance.Lastly, intrusion detection is carried out employing RideNN.In addition, MCB distance-RideNN attained 91.185% accuracy, 92.574% True Negative Rate (TNR), and 90.412% of True Positive Rate (TPR).