Use of skew rays in multimode fibers to generate speckle field with nonzero vorticity

M. A. Bolshtyansky; A. Yu. Savchenko; B. Ya. Zel’dovich

doi:10.1364/ol.24.000433

Fourier-Domain Modal Delay Measurements for Multimode Fibers Optimized for the 850-nm Band in a Local Area Network

IEICE Transactions on Communications ◽

10.1587/transcom.e96.b.2840 ◽

2013 ◽

Vol E96.B (11) ◽

pp. 2840-2844 ◽

Cited By ~ 1

Author(s):

Chan-Young KIM ◽

Tae-Jung AHN

Keyword(s):

Local Area Network ◽

Local Area ◽

Fourier Domain ◽

Area Network ◽

Multimode Fibers

Download Full-text

Integrated Double-Sided Random Microlens Array Used for Laser Beam Homogenization

Micromachines ◽

10.3390/mi12060673 ◽

2021 ◽

Vol 12 (6) ◽

pp. 673

Author(s):

Wei Yuan ◽

Cheng Xu ◽

Li Xue ◽

Hui Pang ◽

Axiu Cao ◽

...

Keyword(s):

Laser Beam ◽

Microlens Array ◽

Energy Utilization ◽

Utilization Rate ◽

Periodic Lattice ◽

Lattice Distribution ◽

Speckle Field ◽

High Laser ◽

Alignment Problem ◽

In Series

Double microlens arrays (MLAs) in series can be used to divide and superpose laser beam so as to achieve a homogenized spot. However, for laser beam homogenization with high coherence, the periodic lattice distribution in the homogenized spot will be generated due to the periodicity of the traditional MLA, which greatly reduces the uniformity of the homogenized spot. To solve this problem, a monolithic and highly integrated double-sided random microlens array (D-rMLA) is proposed for the purpose of achieving laser beam homogenization. The periodicity of the MLA is disturbed by the closely arranged microlens structures with random apertures. And the random speckle field is achieved to improve the uniformity of the homogenized spot by the superposition of the divided sub-beams. In addition, the double-sided exposure technique is proposed to prepare the rMLA on both sides of the same substrate with high precision alignment to form an integrated D-rMLA structure, which avoids the strict alignment problem in the installation process of traditional discrete MLAs. Then the laser beam homogenization experiments have been carried out by using the prepared D-rMLA structure. The laser beam homogenized spots of different wavelengths have been tested, including the wavelengths of 650 nm (R), 532 nm (G), and 405 nm (B). The experimental results show that the uniformity of the RGB homogenized spots is about 91%, 89%, and 90%. And the energy utilization rate is about 89%, 87%, 86%, respectively. Hence, the prepared structure has high laser beam homogenization ability and energy utilization rate, which is suitable for wide wavelength regime.

Download Full-text

Image reconstruction through a multimode fiber with a simple neural network architecture

Scientific Reports ◽

10.1038/s41598-020-79646-8 ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Changyan Zhu ◽

Eng Aik Chan ◽

You Wang ◽

Weina Peng ◽

Ruixiang Guo ◽

...

Keyword(s):

Neural Network ◽

Image Reconstruction ◽

Network Architecture ◽

Neural Network Architecture ◽

Training Time ◽

Modal Dispersion ◽

Multimode Fibers ◽

Speckle Patterns ◽

Hidden Layer ◽

Simple Neural Network

AbstractMultimode fibers (MMFs) have the potential to carry complex images for endoscopy and related applications, but decoding the complex speckle patterns produced by mode-mixing and modal dispersion in MMFs is a serious challenge. Several groups have recently shown that convolutional neural networks (CNNs) can be trained to perform high-fidelity MMF image reconstruction. We find that a considerably simpler neural network architecture, the single hidden layer dense neural network, performs at least as well as previously-used CNNs in terms of image reconstruction fidelity, and is superior in terms of training time and computing resources required. The trained networks can accurately reconstruct MMF images collected over a week after the cessation of the training set, with the dense network performing as well as the CNN over the entire period.

Download Full-text