CLEO 2023 Junnosuke Kokubun


CLEOPR 2023 Participation Report

7th May - 12th May, Convention Center, San Jose, USA

Junnosuke Kokubu, 1st year master's student

About CLEO-PR2023

This meeting will be held in conjunction with the APS (American Physical Society), OSA (The Optical Society), IEEE
Conference on Laser and Electro-Optic, an international conference sponsored by the Photonics Society.
As you can see, it is a very large conference that brings together researchers in the fields of lasers and electro-optics.
Some prominent researchers were also present. 2022 will be a hybrid of in-person and online sessions.
In 2023, however, the conference will be held only in person. Many people gathered at the venue, and it was a face-to-face conference.
The venue was lively, as is typical of the event. Thorlabs and other sponsors exhibited their booths.
exhibited companies from many countries, including Japan.

2. Presentation by the presenter

Title: Small Spectrometer With Extremely High -Resolution (0.07 nm)
Realized With an Improved Reconstruction Algorithm
Presenter:Junnosuke Kokubu
Affiliation:Keio University
Presentation number:STh4G.7(Thu, May 11)

The development of a spectrometer using a photonic crystal waveguide with randomness was presented.
In this study, we investigated the structure of a photonic crystal waveguide with a stepwise chirped structure for spectroscopy.
The resolution limit determined by the manufacturing process is overcome by using random localization caused by manufacturing errors.
The localization in photonic crystals is known to be very sensitive to wavelength. Localization in photonic crystals is known to be very wavelength sensitive.
While the resolution limit of the meter structure is 1.5~1.7 nm, the resolution limit for the case where localized light in the waveguide is taken into account is
The resolution can be improved to 0.08 nm in the case of a single sample.

3. presentations attended

Title: Computational Spectroscopy via Large Aperture Meta-Optics
Presenter:Johannes Froech
Affiliation:University of Washington
Presentation No.: STh4G.5 (Thu, May 11)

This is a study on the development of an ultra-compact spectrometer using metasurfaces and a CCD camera. In this study
In a study conducted by the University of Tokyo, silicon nitride metasurfaces were fabricated on 5-mm-diameter silicon substrates.
A chip made of silicon nitride metasurface was used. The light transmitted from the chip substrate side is formed into wavelength-dependent rays by the silicon nitride metasurface and passed through a lens to the CCD camera
The system reaches In this study, the wavelength dependence of the PSF (Point Spread Function) of the system is calculated in advance.
The spectral characteristics are recorded, and for inputs with unknown spectra, spectra are calculated based on the prior data.
is reconstructed. In this study, we use the L0, L1, and L2 (Tikhonov) regularization method to reconstruct the spectra
It is confirmed that both discrete and continuous spectra can be reconstructed.
The wavelength resolution obtained was 3.5 nm. A unique feature of this study is that the chip and the camera
The point is that it is possible to change the operating wavelength range as a spectrometer according to the distance between the
The system is shown in Fig. 1. As a future prospect, the system in this study can be used as a hyperspectral camera.
The following is a list of the possible extensions to the

Title:Inverse-designed linear coherent photonic networks for high-resolution
spectral reconstruction
Presenter:Yuan Li
Affiliation:The Chinese University of Hong Kong
Presentation number: STh4G.1 (Thu, May 11)

This is a study of creating an on-chip spectrometer in a CMOS-compatible process.
Using a network with a large number of optical elements of three types (litter, router, and mirror), the division of
Optical operation is implemented. The overall element size is 88 × 300 square micrometers, with a 25-channel
Spectral reconstruction is performed on the basis of the tunnel outputs. The wavelength-fractionalization is evaluated by the autocorrelation function
The value is 0.27 nm. The reconstruction method used in this study is a pseudo-inverse matrix method with a pre
The results are further refined by a simulated annealing method.
It is a hybrid method that exploits the advantages of each algorithm.