A Deep Intrusion Detection System in Lambda Architecture Based on Edge Cloud Computing for IoT

IoT devices enable a massive amount of data to be aggregated and analyzed for anomaly detection. The nature of heterogeneous devices introduces the challenge of collecting and handling these massive datasets to perform data analyses to discover cyber attacks in near real-time. However, the traditional IDS cannot deal with such a problem due to scalability limitations and insufficient storage and processing capabilities. Moreover, issues such as network bandwidth, real-time support, and security limit the power of cloud server-based IDSs. This paper presents an edge-cloud deep IDS model in Lambda architecture for IoT security to address these challenges. Our system decreases the training phase's time compared to traditional machine learning algorithms and increases the accuracy of true positive detected attacks. Furthermore, neural network layers lead deep learning to achieve better performance and flexibility compared to conventional machine learning. Our solution enables the detection of suspicious activities in real-time and allows to classify them by analyzing historical data in a batch process.