Abstract: IoT devices are getting increasingly popular. Several IoT issues highlight the necessity for IoT security. The number of assaults on these devices is growing, and most of them are minor variants of previously known attacks that can get beyond traditional firewalls. Existing systems are incompatible with IoT devices due to their low computing capability. Signature-based intrusion detection can only detect known patterns and attacks; therefore, it can't detect newer attacks with unknown patterns. Many systems also use cloud computing, which has the disadvantage of requiring constant internet access, as well as the fact that cloud services are frequently charged. The paper employed a Random Forest ML model to develop a real-time anomaly-based detection system. When a newer attack is detected that is not screened by the firewall, the anomaly-based intrusion detection system kicks in. It can handle newer/unknown assaults that signature-based systems cannot. We are also installing the IDS on a local, higher-powered device rather than using the cloud. The model developed using the IoT network traffic dataset generated by the IoT node in question.
Keywords - Intrusion Detection, Network Security, Anomaly Detection, Internet of Things.
|
DOI:
10.17148/IJARCCE.2022.11425
