Partial coherence enhances parallelized photonic computing

Partial coherence enhances parallelized photonic computing

Nature

10.1038/s41586-024-07590-y

http://dx.doi.org/10.1038/s41586-024-07590-y

Bowei Dong, Frank Brückerhoff-Plückelmann, Lennart Meyer, Jelle Dijkstra, Ivonne Bente, Daniel Wendland, Akhil Varri, Samarth Aggarwal, Nikolaos Farmakidis, Mengyun Wang, Guoce Yang, June Sang Lee, Yuhan He, Emmanuel Gooskens, Dim-Lee Kwong, Peter Bienstman, Wolfram H. P. Pernice, Harish Bhaskaran

silicon photonics, photonic computing, artificial intelligence

31 July 2024

Dong, B., Brückerhoff-Plückelmann, F., Meyer, L., Dijkstra, J., Bente, I., Wendland, D., Varri, A., Aggarwal, S., Farmakidis, N., Wang, M., Yang, G., Lee, J. S., He, Y., Gooskens, E., Kwong, D.-L., Bienstman, P., Pernice, W. H. P., & Bhaskaran, H. (2024). Partial coherence enhances parallelized photonic computing. Nature, 632(8023), 55–62. https://doi.org/10.1038/s41586-024-07590-y

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).

Attribution 4.0 International (CC BY 4.0)

This research is supported by core funding from: A*STAR Institute of Microelectronics
Grant Reference no. : SC24/22-107100

This research is supported by the Singapore A*STAR International Fellowship (AIF).

https://oar.a-star.edu.sg/communities-collections/articles/20728

0028-0836
1476-4687

Institute of Microelectronics