Implementation of Digital Filter Functions in Edge Device Framework for IoT Application System

Currently, we are developing an edge device framework to help the setup of the edge device remotely through an Internet of Things (IoT) application server platform called SEMAR (Smart Environmental Monitoring and Analytical in Real-Time). The framework allows the connection of different sensor interfaces, standardizes data, and utilizes it through multiple output components. However, it lacks capabilities for sensor data processing at the edge device. In this paper, we implement filtering functions using digital signal processing techniques in the edge device framework. The techniques include low, high, and band-pass filters based on Butterworth and Chebyshev-I, as well as Kalman and Savitzky-Golay filters. By identifying these techniques and parameters in the edge configuration file through SEMAR, the user can utilize them without writing codes. It also provides the cascading filter to create a series of filters for data processing in sequences. For evaluations, we apply the proposed functions to the fingerprint-based indoor localization system (FILS15.4). The evaluation results confirmed the effectiveness of the implemented filtering functions in the edge device framework.