The exponential growth of Internet of Things (IoT) devices underscores the need for robust security measures against cyber-attacks. Extensive research in the IoT security community has centered on effective traffic detection models, with a particular focus on anomaly intrusion detection systems (AIDS). This paper specifically addresses the preprocessing stage for IoT datasets and feature selection approaches to reduce the complexity of the data. The goal is to develop an efficient AIDS that strikes a balance between high accuracy and low detection time. To achieve this goal, we propose a hybrid feature selection approach that combines filter and wrapper methods. This approach is integrated into a two-level anomaly intrusion detection system. At level 1, our approach classifies network packets into normal or attack, with level 2 further classifying the attack to determine its specific category. One critical aspect we consider is the imbalance in these datasets, which is addressed using the Synthetic Minority Over-sampling Technique (SMOTE). To evaluate how the selected features affect the performance of the machine learning model across different algorithms, namely Decision Tree, Random Forest, Gaussian Naive Bayes, and k-Nearest Neighbor, we employ benchmark datasets: BoT-IoT, TON-IoT, and CIC-DDoS2019. Evaluation metrics encompass detection accuracy, precision, recall, and F1-score. Results indicate that the decision tree achieves high detection accuracy, ranging between 99.82 and 100%, with short detection times ranging between 0.02 and 0.15 s, outperforming existing AIDS architectures for IoT networks and establishing its superiority in achieving both accuracy and efficient detection times.
In the face of growing Internet of Things (IoT) security challenges, traditional Intrusion Detection Systems (IDSs) fall short due to IoT devices’ unique characteristics and constraints. This paper presents an effective, lightweight detection model that strengthens IoT security by addressing the high dimensionality of IoT data. This model merges an asymmetric stacked autoencoder with a Deep Neural Network (DNN), applying one-class learning. It achieves a high detection rate with minimal false positives in a short time. Compared with state-of-the-art approaches based on the BoT-IoT dataset, it shows a higher detection rate of up to 96.27% in 0.27 s. Also, the model achieves an accuracy of 99.99%, precision of 99.21%, and f1 score of 97.69%. These results demonstrate the effectiveness and significance of the proposed model, confirming its potential for reliable deployment in real IoT security problems.
The proliferation of smart home devices, while enhancing convenience, introduces new security vulnerabilities within the ever-expanding Internet of Things. These resource-constrained devices generate complex data, further complicated by limited labeled data, hindering traditional supervised learning approaches. To address these challenges, this research proposes a novel Anomaly Intrusion Detection System framework for smart home environments. This framework leverages the distributed processing capabilities of the edge-fog-cloud architecture, where a one-class asymmetric stacked autoencoder plays a crucial role. This deep learning model performs efficient dimensionality reduction, transforming complex, high-dimensional data into a lower-dimensional representation and facilitating efficient processing on resource-constrained smart home devices. The proposed model utilizes unsupervised learning to detect anomalies and suspicious activities without requiring extensive labeled data, making it well-suited for dynamic IoT environments. The proposed framework is evaluated using two real-world datasets, BoT-IoT and IoT-23, demonstrating exceptional performance. It achieved a high accuracy of 99.9970% and 97.7284%, with a precision of 99.9987% and 97.5610%, recall of 99.9983% and 97.5401%, and F1-scores of 99.9985% and 97.5505%, respectively. The model also exhibited a low false positive rate of 0.10% and 0.02%, along with a high Area Under the ROC Curve of 99.7559% and 98.3235%. Additionally, the Matthews Correlation Coefficient values of 0.8881 and 0.9543 indicate a balanced classification between normal and attack traffic. The framework demonstrated efficient processing times, with training durations of 9.9349 and 237.2206 seconds and significantly faster testing times of 0.1936 and 0.4504 seconds for the two datasets. Preprocessing times were also minimal at 55.8496 and 0.0973 seconds, confirming the model’s suitability for real-time anomaly detection. These results validate the effectiveness and scalability of the proposed model-for enhancing security in smart home IoT networks.