Experimental comparison of EMA and Kalman filters improves accuracy in microcontroller-based temperature systems, indicating better measurement results.
Key Points
The aim is to compare the performance of Exponential Moving Average and Kalman filters in embedded temperature measurement systems.
Compared EMA and Kalman filters using an LM35 temperature sensor and STM32F4 microcontroller.
Evaluated filtering accuracy, response time, and computational load on the same analog data set.
Conducted experiments to assess the impact of filtering techniques on measurement quality.
Kalman filter demonstrated superior accuracy compared to the EMA filter.
EMA filter showed lower computational load and faster response times.
Effective filtering significantly reduced noise in temperature measurements, enhancing overall system performance.