We have successfully demonstrated three-dimensional (3D) ultrasound imaging via compressed sensing using piezoelectric micromachined ultrasonic transducers (pMUTs). This work reports the first experimental demonstration of compressed sensing with pMUTs, achieving 3D image reconstruction near the acoustic wavelength limit by only four receiving elements. Pseudo-random transmission signatures were encoded through frequency and delay modulation to enable high-fidelity reconstruction with a peak signal contrast of 28.7 dB. A 16-element lithium-niobate pMUT array operating at a resonant frequency of 50.4 kHz was used, with each element generating up to 120 dB sound pressure at 10 mm. Compared to the conventional scheme using beamforming approaches for 3D imaging that require a large array of transducers and complex electronics, this compressed sensing scheme significantly reduces hardware complexity, power consumption, and data acquisition. As such, this new class of pMUT-based imaging establishes a pathway toward low-cost, scalable 3D ultrasound perception for robotics, wearable electronics, autonomous navigation, and medical diagnostics.
Keywords: pMUT, Compressed sensing, Image reconstruction, 3D Imaging, Ultrasound, MEMS, Piezoelectric
b2026p0003