Design and Evaluation of a Scalable and Reconfigurable Multi-Platform System for Acoustic Imaging

Area of Science:

• Acoustic imaging
• Sensor technology
• Signal processing

Background:

• Micro-Electro-Mechanical Systems (MEMS) microphones offer a cost-effective solution for acoustic sensing.
• Scalable and multi-platform frameworks are needed for advanced signal acquisition and processing.

Purpose of the Study:

• To propose a scalable, multi-platform framework for acoustic image generation using MEMS microphone arrays.
• To evaluate the performance and processing times of the framework across different subsystems.
• To demonstrate the feasibility of the system for applications such as biometric identification.

Main Methods:

• Acoustic characterization of MEMS sensors and beam pattern analysis of an 8x8 planar array module.
• Development of a flexible framework integrating FPGA, embedded processor, desktop computer, and GPU.
• Evaluation of signal processing and wideband beamforming (FFT) algorithm processing times on each subsystem.

Main Results:

• Three distinct framework configurations were proposed based on subsystem utilization and algorithm sharing.
• Acoustic images were successfully generated from sound reflections, demonstrating system feasibility.
• Processing times were analyzed across different computational subsystems.

Conclusions:

• The proposed framework provides a low-cost, scalable solution for acoustic imaging with MEMS microphones.
• The system is suitable for various applications, including biometric identification through acoustic reflection analysis.
• The flexible architecture allows for optimization based on specific processing needs and available hardware.