Abstract
The increasing popularity of the Internet of Things (IoT) has significantly impacted our daily lives in the past few years. On one hand, it brings convenience, simplicity, and efficiency for us; on the other hand, the devices are susceptible to various cyber-attacks due to the lack of solid security mechanisms and hardware security support. In this paper, we present IMIDS, an intelligent intrusion detection system (IDS) to protect IoT devices. IMIDS’s core is a lightweight convolutional neural network model to classify multiple cyber threats. To mitigate the training data shortage issue, we also propose an attack data generator powered by a conditional generative adversarial network. In the experiment, we demonstrate that IMIDS could detect nine cyber-attack types (e.g., backdoors, shellcode, worms) with an average F-measure of 97.22% and outperforms its competitors. Furthermore, IMIDS’s detection performance is notably improved after being further trained by the data generated by our attack data generator. These results demonstrate that IMIDS can be a practical IDS for the IoT scenario.
Highlights
Recent years have witnessed the proliferation of the Internet of Things, aiming to bring every physical object into the digital world, resulting in billions of IoT devices connected to the Internet
We present IMIDS, an intrusion detection system powered by a convolutional neural network (CNN) model that can differentiate normal and abnormal activities but identify the type of cyber-attack hidden behind these activities
This paper introduces an effective intrusion detection system powered by a CNN model, namely IMIDS, and a novel attack data generator leveraging a conditional generative adversarial network
Summary
Recent years have witnessed the proliferation of the Internet of Things, aiming to bring every physical object into the digital world, resulting in billions of IoT devices connected to the Internet. The motivation of this work arises from a basic observation: it is non-trivial to enhance the attack detection quality while retaining the model’s simplicity This observation is described in several research papers about IDS, in which authors tend to introduce novel attack detection models to increase the detection accuracy regardless of their complexity, but we believe that high-quality training data play a critical role in achieving an effective model. The main technical goal of this paper is to detect various cyber-attacks accurately by providing an efficient IDS along with an artificial method to generate valuable training data. We present IMIDS, an intrusion detection system powered by a CNN model that can differentiate normal and abnormal activities but identify the type of cyber-attack hidden behind these activities.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
