Related Experiment Video
Updated: Sep 11, 2025

05:57
Characterization of SiN Integrated Optical Phased Arrays on a Wafer-Scale Test Station
Published on: April 1, 2020
8.1K
Hybrid integrated silicon nitride optical phased array with phase calibration for two-dimensional beam steering
Optics Express
|August 13, 2025
Summary
Researchers developed a compact, high-performance hybrid optical phased array for versatile applications. This silicon photonics platform enables efficient, low-power, two-dimensional beam steering using advanced integrated technologies and deep reinforcement learning for calibration.
Area of Science:
- Photonics and Optical Engineering
- Integrated Optics
- Silicon Photonics
Background:
- Optical phased arrays (OPAs) are crucial for beam steering in applications like space communication and sensing.
- Existing OPA architectures often face challenges with size, power consumption, and integration complexity.
- Heterostructure silicon photonics offers a promising avenue for miniaturized and cost-effective optical systems.
Purpose of the Study:
- To introduce and validate a novel hybrid, integrated two-dimensional multi-layer optical phased array architecture.
- To demonstrate the key enabling technologies for compact, low-loss, and high-performance OPA systems.
- To showcase the application of deep reinforcement learning for efficient phase calibration and beam steering.
Main Methods:
- Development of a hybrid integrated silicon photonics platform.
- Implementation of efficient multi-level conical waveguides and low-scattering interlayer couplers.
- Design of compact cross waveguides and sector-structured grating antennas.
- Application of deep reinforcement learning for OPA phase calibration.
- Experimental demonstration of two-dimensional beam steering using thermal-optic (TO) phase control.
Main Results:
- Achieved a compact, low-loss, high-performance hybrid integrated two-dimensional OPA.
- Demonstrated key technologies including advanced waveguides, couplers, and grating antennas.
- Validated the superiority of deep reinforcement learning for OPA phase calibration.
- Successfully achieved two-dimensional beam steering with a wide field-of-view using the silicon photonics platform.
Conclusions:
- The proposed hybrid integrated OPA architecture offers significant advantages in terms of size, performance, and versatility.
- The developed platform enables cost-effective, miniaturized, and widely applicable heterostructure silicon photonics.
- This technology holds strong potential for enabling low-power, high-speed beam steering for diverse applications.

