High-speed single-photon sources for large-scale practical quantum computers

Photonic quantum computers with photon qubits promise to realize large-scale practical quantum computers with its long coherent time. Multi-dimension cluster states with entangled photon pairs realize large-scale measurement-based quantum computers (MBQC).

Single photon source is essential. Especially, electrically-driven, deterministic high-speed generation of indistinguishable photons is crucial. To realize all these properties simultaneously, we develop quantum-dot vertical microcavities by implementing resonant tunneling injection and polarization into quantum-dot-embedded high-Q cavity. Metalens to efficiently couple generated photons into optical fibers is under development. We also develop high-speed generation of Bell state with high-speed single-photon sources to realize multi-dimension cluster state.

Photonic Nanodevices for Trapped-Ion Quantum Computers

Trapped ions are a promising platform for implementing quantum computers because of their exceptional quantum properties. One of the most active recent research trends is the amalgamation of optical functionalities into compact chip-based configurations to facilitate the scalability of these systems. Beyond merely shining light onto individual ions for manipulation, the ability to finely tailor optical attributes, such as polarization states, assumes paramount importance for enabling universal quantum operations. The on-chip integration of such optical functionalities, particularly within the near-ultraviolet to visible wavelength range corresponding to the typical transition wavelengths of ions, poses a significant challenge. Our objective is to embody on-chip ion trap devices capable
of executing sophisticated quantum computations by harnessing state-of-the-art optical
technologies/physics such as topological photonics.

Active photonic cavities for Dynamics AI

Analyzing dynamics of data is essential for software-driven automobiles and drones or large-scale infrastructures such as data networks and energy networks to detect anomalies and security risks instantaneously. Reservoir computing realizes quick risk detection by real-time training of data dynamics with small amount of data. Physical reservoir computing employs physical system as reservoir which is connected to CPU for weighting and realizes AI hardware for CPU accelerator.

Photonic reservoir computing employing photonics as physical system enables high-speed and energy-efficient data analysis. Continuous-media physical reservoir realizes scalable and universal reservoir computing. We develop active photonic cavities implemented by high-Q 2D photonic cavity and active layer with quantum wells to realize continuous-media photonic reservoir for CPU accelerator.