Data Reduction Approach Based on Fog Computing in IoT Environment

This paper investigates a data processing model for a real experimental environment in which data is collected from several IoT devices on an edge server where a clustering-based data reduction model is implemented. Then, only representative data is transmitted to a cloud-hosted service to avoid high bandwidth consumption and the storage space at the cloud. In our model, the subtractive clustering algorithm is employed for the first time for streamed IoT data with high efficiency. Developed services show the real impact of data reduction technique at the fog node on enhancing overall system performance. High accuracy and reduction rate have been obtained through visualizing data before and after reduction.