Securing IoT-Enabled Smart Cities and Detecting Cyber Attacks in Smart Homes for a Greener Future

Smart homes, as integral components of smart cities, are increasingly targeted by various cyberattacks such as Distributed Denial of Service (DDoS), botnet-based intrusions, and network scanning and flooding attacks. These threats can compromise privacy, disrupt services, and even cause physical harm by manipulating connected devices such as smart gas systems or door locks. This paper presents a convolutional neural network (CNN) based model to detect these attacks. CNN is used to extract the local and global features of the input traffic for efficient detection of attacks. The proposed model was tested on a Kaggle dataset with ten different attack types. In the dataset, there are 7062606 samples. Results show that the proposed model has an accuracy 99% for detecting most of the attacks. The proposed model outperforms traditional machine learning models like logistic regression, decision tree, random forest, support vector machine, and K-nearest neighbors.