A SiN-Si dual-layer optical phased array (OPA) chip is designed and fabricated. It combines the low loss of SiN with the excellent modulation performance of Si, which improves the performance of Si single-layer OPA. A novel optical antenna and an improved phase modulation method are also proposed, and a two-dimensional scanning range of 96 ° × 14 ° is achieved, which makes the OPA chip more practical.

A novel optical antenna for optical phased arrays is proposed and simulated. A high-contrast grating structure is used to achieve extremely efficient emission. The emission efficiency is as high as 93.94% at 1.55 μm, which exceeds 50% in a range of wavelength from 1.48 μm to 1.62 μm. The antenna can achieve a perfect grating lobe suppression with background suppression of 28.4 dB when the phase difference between adjacent waveguides is 0. A 16-wire optical phased array can easily achieve a scan range of ± 22.8° × 20.2° with a beam width of 2.4° × 2.5°, by employing the optical antenna proposed.

Optical phased array can be widely used in many fields benefiting from its superior performance. We designed and fabricated chirped grating antennas and uniform grating antennas for the optical phased array. The effective aperture of the chirped grating antennas is about twice that of the uniform grating antennas. The chirped grating optical phased array can receive the reflected signal of the object at a distance of 100 m, while the uniform grating optical phased array can only receive 50 m under the same conditions. Additionally, a ranging distance of 25 m is achieved when two chirped grating optical phased arrays are set as the transmitter and receiver.

All-solid-state beam scanning chip is ideal for next-generation LiDAR due to its reliability and small size. Here we propose a focal plane array chip for two-dimensional scanning using field-of-view splicing technology on silicon photonics platform. The chip has two rotationally symmetric structures, each including a 1 × 64 antenna array accompanied by a 1 × 64 micro-ring optical switch array. We demonstrate a two-dimensional scanning equivalent to an 8-line LiDAR with a field-of-view of 82° × 32°, a beam divergence angle of 0.07° × 0.07°, and a background suppression ratio of over 20 dB. Our chip works in such a simple way that only one optical switch needs to be turned on each time the beam is emitted. And the chip is scalable that a larger range of two-dimensional scanning can be achieved when using more antennas for field-of-view splicing and cascading more optical switches.

Light detection and ranging (LiDAR) is a widely utilized technology for extracting information from the outside world in fields such as automotive, robotics, and aerospace. Optical phased array (OPA) is a promising solution for LiDAR technology, although its application is limited by loss and alias-free steering range. In this paper, we propose a dual-layer antenna that achieves a peak directionality of over 92%, thereby mitigating antenna loss and enhancing power efficiency. Based on this antenna, we design and fabricate a 256-channel non-uniform OPA that achieves 150° alias-free steering.

We propose the photonic architecture based on microring resonator (MRR) arrays for binary convolutional neural networks (BCNNs) accelerated computing. The MRR crossbar unit is used for computing weight {−1, 1} and the single MRR is for input {0, 1}. The computing parallelism is improved through wavelength division multiplexing. The photonic BCNN achieves 97.29% classification accuracy on the MNIST test set which is only 1.94% lower than the accuracy of the 32-bit neural network, and saves $32\times $ at memory usage. We analyze effects of input and weight encoding errors on the photonic BCNN. When the input or weight error rate is less than 0.01%, the test accuracy remains unchanged. We evaluate the performance of the photonic BCNN architecture considering optical loss, inter-channel crosstalk, operation frequency and device power consumption. The energy efficiency of the designed photonic BCNN architecture is 1.72 pJ/MAC, which is $4.80\times $ and $61.32\times $ better than the 8-bit and 16-bit architecture respectively. The photonic BCNN is promising to be used for edge computing.