DaaS: Dew Computing as a Service for Intelligent Intrusion Detection in Edge-of-Things Ecosystem

Edge of Things (EoT) enables the seamless transfer services, storage, and data processing from cloud layer to edge devices in a large-scale distributed Internet (IoT) ecosystems (e.g., Industrial systems). This transition raises privacy security concerns EoT paradigm at different layers. Intrusion detection systems (IDSs) are implemented protect underlying resources attackers. However, current IDSs not intelligent enough control false alarms, which significantly lower reliability add analysis burden on IDSs. In this article, we present Dew Computing as Service (DaaS) for intrusion ecosystems. DaaS, deep learning-based classifier is used design an alarm filtration mechanism. mechanism, accuracy improved (or sustained) by using belief networks. past, cloud-based techniques have been applied offloading tasks, increases middle communication delay. Here, introduce dew computing features that smart reduction system. when experimented simulated environment, reflects response time process ecosystem. The revamped DBN model achieved classification up 95%. Moreover, it depicts 60% improvement latency 35% workload servers compared IDS.