An Innovative SALO-IDT-Based Intrusion Detection Model for Increasing the Security of IoT Networks

The Internet of Technology (IoT) is one of the emerging and popular networking field extensively used in many real-time applications. But, it is more prone of the security vulnerabilities attacks, which down the performance of entire networking systems. Hence, developing an efficient intrusion detection framework is one of the highly demanding and crucial tasks mainly focused by many researchers. In the conventional works, various optimization and classification techniques are deployed to design and develop an accurate intrusion detection framework for securing IoT networks. Still, it faced the problem to ensure the complete security of IoT systems, due to the challenges of reduced accuracy, increased time consumption, and complex system modeling. Therefore, the proposed intends to implement a hybrid Salp-swarm Ant Lion Optimization (SALO) incorporated Improved Decision Tree (IDT) model for IoT security. Here, the data preprocessing is performed to enhance the quality of input network datasets based on the operations of normalization, discretization, and integration. The hybrid SALO technique is used to optimally choose the parameters for training the IDT classifier, which produces the label for identifying the intrusions. To validate the results of this approach, various performance parameters are estimated for the different network attack datasets.