<![CDATA[Indonesia is located in the most seismically active region in the world, creating an urgent need for an affordable earthquake monitoring system. This study presents a real-time earthquake event detection system based on an MPU6050 MEMS (Micro-Electro-Mechanical System) accelerometer with adaptive drift correction integrated with an ESP32 microcontroller and Internet of Things (IoT)-based data transmission. The system acquires three-axis ground acceleration in real time, adaptive bias compensation, drift suppression using zero-velocity update (ZUPT), and event detection based on an adaptive threshold relative to a local noise baseline. Key seismic parameters, including Peak Ground Acceleration (PGA) and Peak Ground Velocity (PGV), are calculated in real time, while frequency domain characterization is performed offline using Fast Fourier Transform (FFT) analysis. Long-term stationary measurements over approximately 60 minutes showed stable noise characteristics, with root mean square (RMS) noise levels of 0.0067 m/s², 0.0039 m/s², and 0.0027 m/s² in the X, Y, and Z axes, respectively. The system demonstrated excellent baseline stability, with a PGA variation of only 0.7% and a measured drift rate of 0.00088 m/s² per hour. Comparative validation against a smartphone-based reference sensor demonstrated a 100% event detection rate, and a PGA estimation error of ±2%, with linearity exceeding a coefficient of determination of 0.99.]]>
Copyrights © 2026
