An IoT-based Deep Belief Network in Intrusion Detection System using Butterfly Optimization

Le résumé fourni par la source

Rapid technological advancements in the communication and internet industries have led to a huge increase in network scale and data volume. Network security is having a hard time identifying attacks as new threats are being created. Additionally, you can’t overlook the possibility of unauthorized users trying to attack the network in various methods. Numerous technologies, such as an intrusion detection system, can be utilized to protect the network from potential threats (IDS). The collecting and dissemination of data by Internet-connected devices is referred to as the “Internet of Things” (IoT). There are now more privacy and security problems due to the proliferation of Internet connections and the introduction of new technologies like the Internet of Things (IoT). IoT incursions are becoming increasingly complex and numerous as a result of the proliferation of connected devices. Companies are boosting their investment in research and development in order to better detect these threats. In this research, Deep Belief Network (DBN) and Butterfly Optimization Algorithm (BOA) Intrusion Detection Systems (IDS) are proposed for detecting Io T network assaults. In order to minimize information leakage on the test data, the UNSW-NB15 dataset was normalized using the Absolute Maximum Scaling (AMS) approach in the first stage of this study process. This dataset contains a mixture of current assaults and typical network traffic, categorized into nine different attack types. The next step was to use Principal Component Analysis to reduce the dimensionality (PCA). Last but not least, results showed that DBN-BOA was superior to other IoT intrusion detection systems in terms of accuracy and speed.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.