A Comparative Analysis of IoT Malware Detection Using CNN and Deep Learning

The emergence of Internet of Things devices has significantly increased malware attacks on IoT devices. Therefore, there is a growing need for efficient and reliable malware detection mechanisms that can identify and prevent such attacks. This paper proposes a deep learning methodology for IoT malware detection using two popular convolutional neural networks (CNNs): CNN and VGG16. We evaluated the performance of two deep learning models, convolutional neural network (CNN) and VGG16, for IoT malware detection. We used a standard image dataset of IoT malware samples to train and test the models. Our experimentations showed that both models can achieve high accuracy in detecting IoT malware, with VGG16 performing slightly better than CNN. We conducted an examination of the impact of various hyperparameters, such as batch size, learning rate, and optimizer, on the performance of these models. We found that optimizing these hyperparameters can remarkably improve the detection accuracy of Neural Networks. Our findings suggest that deep learning algorithms can be effectively used for IoT malware detection. The results of this study can be useful for developing more potent and efficient malware detection systems for IoT devices.