ABSTRACT
Advancements in optical coherence control1-5 have unlocked many cutting-edge applications, including long-haul communication, light detection and ranging (LiDAR) and optical coherence tomography6-8. Prevailing wisdom suggests that using more coherent light sources leads to enhanced system performance and device functionalities9-11. Our study introduces a photonic convolutional processing system that takes advantage of partially coherent light to boost computing parallelism without substantially sacrificing accuracy, potentially enabling larger-size photonic tensor cores. The reduction of the degree of coherence optimizes bandwidth use in the photonic convolutional processing system. This breakthrough challenges the traditional belief that coherence is essential or even advantageous in integrated photonic accelerators, thereby enabling the use of light sources with less rigorous feedback control and thermal-management requirements for high-throughput photonic computing. Here we demonstrate such a system in two photonic platforms for computing applications: a photonic tensor core using phase-change-material photonic memories that delivers parallel convolution operations to classify the gaits of ten patients with Parkinson's disease with 92.2% accuracy (92.7% theoretically) and a silicon photonic tensor core with embedded electro-absorption modulators (EAMs) to facilitate 0.108 tera operations per second (TOPS) convolutional processing for classifying the Modified National Institute of Standards and Technology (MNIST) handwritten digits dataset with 92.4% accuracy (95.0% theoretically).
Subject(s)
Neural Networks, Computer , Optics and Photonics , Photons , Tomography, Optical Coherence , Humans , Optics and Photonics/instrumentation , Optics and Photonics/methods , Parkinson Disease/diagnosis , Parkinson Disease/physiopathology , Silicon/chemistry , Tomography, Optical Coherence/instrumentation , Tomography, Optical Coherence/methods , Gait/physiology , Datasets as Topic , Sensitivity and SpecificityABSTRACT
The operation of photonic devices often relies on modulation of their refractive index. While the sub-bandgap index change through bound-electron optical nonlinearity offers a faster response than utilizing free carriers with an overbandgap pump, optical switching often suffers from inefficiency. Here, we use a recently observed metasurface based on mirror-induced optical bound states in the continuum, to enable superior modulation characteristics. We achieve a pulsewidth-limited switching time of 100 fs, reflectance change of 22%, remarkably low energy consumption of 255 µJ/cm2, and an enhancement of modulation contrast by a factor of 440 compared to unpatterned silicon. Additionally, the narrow photonic resonance facilitates the detection of the dispersive nondegenerate two-photon nonlinearity, allowing tunable pump and probe excitation. These findings are explained by a two-band theoretical model for the dispersive nonlinear index. The demonstrated efficient and rapid switching holds immense potential for applications, including quantum photonics, sensing, and metrology.