Performance Comparison and Analysis of Machine Learning and Deep Learning Models for Network Intrusion Detection In IoT-Edge Frameworks

The Internet of Things (IoT) has fundamentally changed the approach of technology utilization in day-to-day life. With a drastic increase in the use of sensors, a huge volume of data is generated, and it becomes cumbersome for central repositories to handle it. To address this issue, an edge device layer is introduced in IoT to lessen cloud computing and storage load by exploiting the resources close to the data source. However, this infrastructure added various security concerns. Therefore, detecting intrusions in IoT edge devices became a primary goal. Recent advancements in intrusion detection in IoT Edge require Machine learning (ML) and deep learning (DL) algorithms to identify the intrusions. The effectiveness of ML/DL models for intrusion detection in IoT edge devices is shown in this paper. Three prominent threats are identified for intrusion detection. The outcomes of various algorithms are compared using different performance metrics and analyzed for their applicability in IoT edge devices.