Fault Prediction Using Svm And Annon Iot Environment With Heterogeneous Sensing Data Fusion

Abstract

This study proposes a novel approach for fault prediction within Internet of Things (IoT) environments leveraging a SVM and ANN framework. The IoT ecosystem is characterized by its heterogeneity and diverse sensing data sources, posing challenges for fault detection and prediction. In this research, these challenges are addressed by integrating heterogeneous sensing data through fusion techniques and employing a SVM and ANN model for fault prediction. The architecture incorporates to handle uncertainties inherent in IoT data streams, enhancing the robustness and accuracy of fault prediction. Through extensive experimentation and evaluation, our approach demonstrates promising results achieving 92% accuracy in predicting faults within IoT environments.