Improved Resolution and Cost Performance of Low-Cost MEMS Seismic Sensor through Parallel Acquisition

Area of Science:

• Geophysics
• Seismology
• Sensor Technology

Background:

• Effective earthquake monitoring relies on dense, high-performance seismic sensor networks for rapid reporting and early warning.
• Existing seismic sensors often face limitations in resolution and cost, impacting network effectiveness.

Purpose of the Study:

• To propose a method for improving seismic sensor resolution using multiple low-cost, low-resolution digital MEMS accelerometers.
• To develop a cost-effective MEMS seismic sensor with enhanced data processing and storage capabilities.

Main Methods:

• Utilizing the correlation average method with multiple MEMS accelerometers to increase seismic data resolution.
• Developing an embedded system with ARM and Linux, featuring a large-capacity SD card for long-term data storage (approx. 12 months).
• Implementing real-time seismic data processing algorithms for automatic event identification and ground motion parameter calculation.

Main Results:

• Reduced RMS noise from 0.096 cm/s² (single sensor) to 0.034 cm/s² (eight sensors) in the 0.1-20 Hz bandwidth.
• Achieved a dynamic range exceeding 90 dB, with an amplitude-frequency response of DC-80 Hz (-3 dB).
• Demonstrated improved signal-to-noise ratio (SNR) for detecting earthquakes (M2.2-M5.1, <200 km) compared to previous methods.

Conclusions:

• The correlation average method effectively enhances the resolution of low-cost MEMS accelerometers for earthquake monitoring.
• The developed cost-effective MEMS seismic sensor offers improved performance, long-term data storage, and ease of installation.
• This approach provides a viable solution for increasing seismic network density and improving earthquake early warning systems.