Highly efficient convolution computing architecture based on silicon photonic Fano resonance devices

Abstract:

Convolutional neural networks require a lot of multiplication and addition operations completed by traditional electrical multipliers, leading to high power consumption and limited speed. Here, a silicon waveguide-based wavelength division multiplexing architecture for convolution neural network is optimized with high energy efficiency Fano resonator. Coupling of T-waveguide and Micro-ring resonator generate Fano resonance with small half-width, which can significantly reduce the modulator power consumption. National Grid's insulator dataset is used to test Fano resonance modulator-based convolutional neural networks. The results show that accuracy for insulator defect recognition reaches 99.27% with much lower power consumption. Obviously, Our opti-mized photonic integration architecture for convolutional neural networks has broad potential for the artificial intelligence hardware platform.

Key words:Fano resonant, Insulator dataset, Convolutional neural networks,