Secure intrusion detection system for IOT sensor data using LR-ECC and FY-SFL-DLNN algorithm

Abstract

Identification and prevention of intrusions is the primary challenge in high-speed networks like IoT. Many Intrusion Detection Systems (IDSs) have been developed already by researchers; but, they are not fully capable of protecting the network. Conventional methods do not optimise their learning models and ignore feature reduction. This negatively impacts the accuracy, efficiency, computation time, and security of the conventional IDS. Hence, this paper proposed a Secure IDS by utilising LR-ECC and FY-SFL-DLNN. Initially, the input dataset is preprocessed and the features are extracted from the data. Utilising the EK-LDA, the features are decreased. After that, utilising the BM-KMA, the data separation occurs. Lastly, by employing FY-SFL-DLNN, data classification is performed. Thus, the IoT sensor data is safely transmitted by utilising the LR-ECC algorithm after detection. Grounded on the experimental outcomes, the proposed work’s performance is analogised with the conventional techniques for validating the proposed system’s effectiveness.