Journal of Cybersecurity and Information Management JCIM 2690-6775 2769-7851 10.54216/JCIM https://www.americaspg.com/journals/show/1748 2019 2019 American University in the Emirates, Dubai, UAE Ahmed N. Al Al-Masri LTU University of Technology, Sweden Hamam Mokayed The Internet of Things (IoT) has transformed the way we live and work, with billions of interconnected devices continuously exchanging data. However, the increasing adoption of IoT devices has also made them an attractive target for cybercriminals. Botnets, a network of compromised devices that can be remotely controlled by attackers, are one of the most significant threats to IoT networks. Traditional security solutions are insufficient to combat this threat, as they often rely on signature-based detection methods that can be easily bypassed by attackers. This work proposes an applied deep learning-based approach to secure IoT networks against botnet attacks, based on residual learning architecture that combine convolutional neural network to analyze device behavior and identify abnormal activity patterns that may indicate botnet infection. Our approach is evaluated on real-world BotNet dataset and achieved a high detection rate of botnet activity, outperforming traditional detection methods. The empirical findings show that ours can be used as a tool for developing more advanced and adaptive security solutions to safeguard the IoT galaxy. 2021 2021 17 27 10.54216/JCIM.050102 https://www.americaspg.com/articleinfo/2/show/1748