Implementation of a Virtual Microphone Array to Obtain High Resolution Acoustic Images

Area of Science:

• Acoustics
• Signal Processing
• Sensor Technology

Background:

• Digital MEMS microphones and FPGA systems enable cost-effective sensor arrays.
• Scaling to thousands of sensors presents significant implementation challenges.
• Virtual arrays offer a potential solution for large-scale acoustic sensing.

Purpose of the Study:

• To analyze the implementation and performance of a large virtual microphone array.
• To evaluate the beampattern, focusing capacity, and imaging capabilities of the virtual array.
• To demonstrate the feasibility of high-resolution acoustic imaging with virtual arrays.

Main Methods:

• Implemented a virtual array of 6400 MEMS microphones by repositioning an 8x8 physical array.
• Utilized a 2D positioning system to achieve a 1x1 m² array spatial aperture.
• Applied SOund Detection And Ranging (SODAR) principles for analysis.
• Analyzed beampattern and focusing using beamforming algorithms with spherical wave assumptions.
• Acquired acoustic images of a mannequin at various frequencies and ranges.

Main Results:

• Successfully implemented and characterized an 80x80 virtual MEMS microphone array.
• Demonstrated effective focusing capabilities consistent with SODAR principles.
• Achieved high angular resolution in acoustic imaging.
• Successfully identified distinct body parts of the mannequin in acoustic images.

Conclusions:

• Virtual microphone arrays are a viable method for creating large-scale acoustic sensing systems.
• This approach enables high-resolution acoustic imaging suitable for object identification.
• The findings support the use of virtual arrays in applications requiring detailed acoustic visualization.