Method EWMA (Exponentially Weighted Moving Average) as a Filter to Fine and Remove Noise on Time Series Data

Abstract

The Moving Average method is the most common filter in Digital Signal Processing, the Moving Average method is optimal for reducing unstable value fluctuations due to interference by maintaining accurate readings. Therefore it is often used for signal coding in the context of time (Smith, 1999). The kind of moving average method is the Exponentially Weighted Moving Average (EWMA) which is often applied to a time sequence of random variables, by calculating the weighted average of the sequence by applying a weight that decreases geometrically with the length of observation (Perry, 2011). this method can be applied to tools/instruments that use data continuously in a certain time so that it will require a filter so that the fluctuations in the data are not extreme which will result in changes in the output response being softer so as not to damage the device on the output. This study uses the EWMA method to eliminate noise in the processing of sensor readings. The results of the study, the system is able to monitor realtime ph and temperature data. In fish farming, measuring and monitoring water temperature and ph is important, if it is not regularly measured and monitored it can cause various problems such as pH and temperature that are not optimal for fish growth and health. The research method to be carried out is R&D, EWMA will be applied to a multi-sensor circuit then processed by a data logger device to then be displayed on serial data on its output.