Anomaly Detection framework for efficient sensing in healthcare IoT systems

Cloud computing and IoT are designed to assist the development of intelligent applications. A novel anomaly detection system could resolve the service composition problem in the dynamic IoT and cloud environment. Thus a framework for monitoring real-time anomalies is suggested to monitor abnormalities in IoT systems. This paper discusses an anomaly detection system for recognizing flaws in an IoT environment. The recommended system utilizes an Entropy Generalized Discriminant Analysis (E-GDA) technique to eliminate the redundant features and the Deep Learning Vector Quantization Correlation-based Mayfly Algorithm (DLVQ-CMA) for the classification of IoT data. Data packets channelized by IoT devices must be analyzed, operation rules must be learned, and management personnel should be reminded when the device performs an abnormal operation to guarantee the safety of the control process.