MUMBAI, India, Oct. 12 -- Intellectual Property India has published a patent application (202641118974 A) filed by Sreyas Institute Of Engineering And Technology on October 05, 2026, for Quantum Dot Photonic Integrated Circuit For High-Speed, Low-Power Optical Neural Network Computing..
Inventors include Dr N Murali Krishna; Feroz Mohammad; and R Uthira Devi.
The application for the patent was published on October 09, 2026, under issue no. 41/2026.
Abstract: ABSTRACT [0017] A quantum dot photonic integrated circuit is disclosed for performing high-speed and low-power optical neural network computations. The circuit integrates arrays of quantum-dot light sources, wavelength-selective microring resonators for weighting, quantum-dot-based nonlinear activation elements, and silicon photonic waveguides on a single heterogeneous platform. Light generated by the quantum-dot sources propagates through a multi-layer mesh of tunable resonators that implement matrix multiplications optically, while the nonlinear elements provide activation functions without external electrical power. The architecture achieves computational densities exceeding several tera-operations per square millimeter while consuming femtojoule-scale energy per operation. Fabrication follows a foundry- compatible heterogeneous integration sequence that bonds quantum-dot epitaxial layers onto silicon photonic waveguides, followed by lithographic definition of resonators, phase shifters, and detectors. The resulting device supports real-time inference for classification and regression tasks at clock rates above twenty gigahertz, offering a compact, energy-efficient solution for edge artificial-intelligence systems.
Disclaimer: Curated by HT Syndication.
