Depth Dynamics via One-Bit Frequency Probing in Embedded Direct Time-of-Flight Sensing

Area of Science:

• Photonics and Sensor Technology
• Signal Processing
• Computational Imaging

Background:

• Traditional time-of-flight (ToF) sensors using single-photon avalanche diodes (SPADs) aggregate photon return times, losing crucial data for dynamic depth analysis.
• Measuring high-frequency motion and transient depth changes requires advanced signal processing that is often computationally intensive.

Purpose of the Study:

• To develop a method for transforming direct ToF sensors into depth frequency analyzers.
• To enable lightweight, on-sensor computation for measuring high-frequency motion and transient depth events.
• To extend depth dynamics analysis to time-localized detection of brief depth changes.

Main Methods:

• Replaced conventional discrete Fourier transforms (DFTs) with one-bit probing sinusoids generated via oversampled sigma-delta modulation for in-pixel frequency analysis.
• Implemented lightweight, multiplier-free, and floating-point-free frequency analysis.
• Extended analysis to Haar wavelets for time-localized detection of transient depth changes.

Main Results:

• Achieved noise performance comparable to full-resolution DFTs.
• Successfully detected sub-millimeter motions exceeding 6 kHz.
• Demonstrated localization of millisecond-scale transient depth events.
• Validated through simulations and hardware experiments.

Conclusions:

• The proposed method enables a new class of compact, motion-aware ToF sensors.
• Potential applications include industrial predictive maintenance, structural health monitoring, robotic perception, and dynamic scene understanding.
• The approach facilitates efficient, on-sensor analysis of depth dynamics.