In the past few decades, great efforts were devoted to improving radar's functionality, precision, and response time, of which the key is to generate, control and process a wideband signal with high speed. Therefore, StarLight holds promise for realizing low energy consumption hardware accelerators to address the incoming challenges of data-intensive artificial intelligence (AI) applications.Īs the only method for all-weather, all-time and long-distance target detection and recognition, radar has been intensively studied since it was invented, and is considered as an essential sensor for future intelligent society. We also implemented a simulation on the Iris dataset classification and achieved an inference accuracy of 96%, which is entirely consistent with the classification accuracy on a 64-bit computer. As a proof of concept, a 4×4×4 optical computing unit featuring 4-wavelength and 4-mode is simulated with 10 Gbps, 15 Gbps and 20 Gbps data rates. Furthermore, we design an on-chip wavelength and mode hybrid multiplexing module and scheme to increase the computational parallelism.

Specifically, StarLight uses nanophotonic non-volatile memory and passive microring resonators (MRs) to form a photonic dot-produce engine, achieving optical ‘in-memory’ multiplication operation with near-zero power consumption during the inference phase.

This paper proposes StarLight, a low-power consumption and high inference throughput photonic artificial neural network (ANN) accelerator featuring the photonic ‘in-memory’ computing and hybrid mode-wavelength division multiplexing (MDM-WDM) technologies.

Digital optical converter