Use of machine learning to efficiently predict the confinement loss in anti-resonant hollow-core fiber

Fanchao Meng; Xiaoting Zhao; Jinmin Ding; Yingli Niu; Xinghua Zhang; Mateusz Smietana; Ryszard Buczyński; Ryszard Buczyński; Bo Lin; Guangming Tao; Lvyun Yang; Xin Wang; Shuqin Lou; Xinzhi Sheng; Sheng Liang

doi:10.1364/OL.422511

Optics Letters
Vol. 46,
Issue 6,
pp. 1454-1457
(2021)
•https://doi.org/10.1364/OL.422511

Use of machine learning to efficiently predict the confinement loss in anti-resonant hollow-core fiber

Fanchao Meng, Xiaoting Zhao, Jinmin Ding, Yingli Niu, Xinghua Zhang, Mateusz Smietana, Ryszard Buczyński, Bo Lin, Guangming Tao, Lvyun Yang, Xin Wang, Shuqin Lou, Xinzhi Sheng, and Sheng Liang

Not Accessible

Your library or personal account may give you access

Get PDF
Email
Share
Get Citation
Copy Citation Text
Fanchao Meng, Xiaoting Zhao, Jinmin Ding, Yingli Niu, Xinghua Zhang, Mateusz Smietana, Ryszard Buczyński, Bo Lin, Guangming Tao, Lvyun Yang, Xin Wang, Shuqin Lou, Xinzhi Sheng, and Sheng Liang, "Use of machine learning to efficiently predict the confinement loss in anti-resonant hollow-core fiber," Opt. Lett. 46, 1454-1457 (2021)

Export Citation
- BibTex
- Endnote (RIS)
- HTML
- Plain Text
Citation alert
Save article

Check for updates

Related Topics
Table of Contents Category
- Fiber Optics and Optical Communications
Optics & Photonics Topics
?

The topics in this list come from the Optics and Photonics Topics applied to this article.

About this Article
History
- Original Manuscript: February 10, 2021
- Revised Manuscript: February 18, 2021
- Manuscript Accepted: February 19, 2021
- Published: March 15, 2021

Abstract

The fundamental mode confinement loss (CL) of anti-resonant hollow-core fiber (ARF) is efficiently predicted by a classification task of machine learning. The structure–parameter vector is utilized to define the sample space of ARFs. The CL of labeled samples at 1550 nm is numerically calculated via the finite element method (FEM). The magnitude of CL is obtained by a classification task via a decision tree and $ k $-nearest neighbors algorithms with the training and test sets generated by 290700 and 32300 labeled samples. The test accuracy, confusion matrices, and the receiver operating characteristic curves have shown that our proposed method is effective for predicting the magnitude of CL with a short computation runtime compared to FEM simulation. The feasibility of predicting other performance parameters by the extension of our method, as well as its ability to generalize outside the tested sample space, is also discussed. It is likely that the proposed sample definition and the use of a classification approach can be adopted for design application beyond efficient prediction of ARF CL and inspire artificial intelligence and data-driven-based research of photonic structures.

Full Article | PDF Article

More Like This

Discovering extremely low confinement-loss anti-resonant fibers via swarm intelligence

Fanchao Meng, Xiaoting Zhao, Jinmin Ding, Yingli Niu, Xinghua Zhang, Lvyun Yang, Xin Wang, Shuqin Lou, Xinzhi Sheng, Guangming Tao, and Sheng Liang
Opt. Express 29(22) 35544-35555 (2021)

Confinement loss prediction in diverse anti-resonant fibers through neural networks

Gu Zhenyu, Ning Tigang, Pei Li, Liu Yangmei, Li Jing, Zheng Jingjing, Song Jingyi, Zhang Chengbao, Wang Hua, Jiang Wei, and Wang Wensheng
Opt. Express 32(6) 8903-8918 (2024)

Impact of cladding elements on the loss performance of hollow-core anti-resonant fibers

Md. Selim Habib, Christos Markos, and Rodrigo Amezcua-Correa
Opt. Express 29(3) 3359-3374 (2021)

Previous Article Next Article

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Figures (6)

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Tables (4)

You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Equations (1)

You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

	Classification Boundary	Classes
Binary-class	$C L > 0.01 d B / m$	0
	$0.01 d B / m \geq C L$	1
Five-class	$C L > 0.01 d B / m$	0
	$1 \times 10^{- 2} d B / m \geq C L > 1 \times 10^{- 3} d B / m$	1
	$1 \times 10^{- 3} d B / m \geq C L > 1 \times 10^{- 4} d B / m$	2
	$1 \times 10^{- 4} d B / m \geq C L > 1 \times 10^{- 5} d B / m$	3
	$1 \times 10^{- 5} d B / m \geq C L$	4

No.	Symbols	Range	Description
1	$s_{1}$	1, 2, 3	Structural style of the first tubes
2	$s_{2}$	0, 1, 2, 3	Structural style of the second tubes
3	$N$	5–10	Number of first/second tubes
4	$D_{c o r e}$	30 µm	Core diameter
5	$M a_1 c$	20–40 µm	Major axis of the first cladding tube
6	$M i_1 c$	20–40 µm	Minor axis of the first cladding tube
7	$M a_2 c$	10–20 µm	Major axis of the second cladding tube
8	$M i_2 c$	10–20 µm	Minor axis of the second cladding tube
9	$M a_2 n$	$0.3 - 0.8 \cdot M a_2 c$	Major axis of the second nested tube
10	$M i_2 n$	$0.3 - 0.8 \cdot M a_2 n$	Minor axis of the second nested tube
11	$M a_1 n$	$0.3 - 0.8 \cdot M a_1 c$	Major axis of the first nested tube
12	$M i_1 n$	$0.3 - 0.8 \cdot M a_1 n$	Minor axis of the first nested tube
13	$t_1 c$	0.3–0.7 µm	Thickness of the first cladding tube
14	$t_1 n$	0.3–0.7 µm	Thickness of the first nested tube
15	$t_2 c$	0.3–0.7 µm	Thickness of the second cladding tube
16	$t_2 n$	0.3–0.7 µm	Thickness of the second nested tube

$s_{1}$			$s_{1}$			$s_{1}$			$s_{1}$
$s_{2}$	$N$	Labels	$s_{2}$	$N$	Labels	$s_{2}$	$N$	Labels	$s_{2}$	$N$	Labels
	5	140		5	100		5	1860		5	2270
	6	270		6	770		6	6150		6	5180
1	7	580	1	7	1790	1	7	11880	1	7	9350
0	8	1870	1	8	9000	2	8	790	3	8	620
	9	2640		9	5040		9	20660		9	11790
	10	230		10	10680		10	820		10	1830
	5	190		5	1800		5	1290		5	1410
	6	180		6	5290		6	1970		6	4810
2	7	4980	2	7	9870	2	7	4450	2	7	8700
0	8	0	1	8	380	2	8	10410	3	8	700
	9	0		9	10650		9	7960		9	0
	10	0		10	450		10	0		10	0
	5	0		5	2040		5	4510		5	0
	6	55440		6	9300		6	11620		6	0
3	7	7160	3	7	20790	3	7	5260	3	7	15770
0	8	5300	1	8	2060	2	8	650	3	8	1190
	9	0		9	0		9	0		9	110
	10	0		10	0		10	0		10	0

	Classification Boundary	Classes
Binary-class	$C L > 0.01 d B / m$	0
	$0.01 d B / m \geq C L$	1
Five-class	$C L > 0.01 d B / m$	0
	$1 \times 10^{- 2} d B / m \geq C L > 1 \times 10^{- 3} d B / m$	1
	$1 \times 10^{- 3} d B / m \geq C L > 1 \times 10^{- 4} d B / m$	2
	$1 \times 10^{- 4} d B / m \geq C L > 1 \times 10^{- 5} d B / m$	3
	$1 \times 10^{- 5} d B / m \geq C L$	4

No.	Symbols	Range	Description
1	$s_{1}$	1, 2, 3	Structural style of the first tubes
2	$s_{2}$	0, 1, 2, 3	Structural style of the second tubes
3	$N$	5–10	Number of first/second tubes
4	$D_{c o r e}$	30 µm	Core diameter
5	$M a_1 c$	20–40 µm	Major axis of the first cladding tube
6	$M i_1 c$	20–40 µm	Minor axis of the first cladding tube
7	$M a_2 c$	10–20 µm	Major axis of the second cladding tube
8	$M i_2 c$	10–20 µm	Minor axis of the second cladding tube
9	$M a_2 n$	$0.3 - 0.8 \cdot M a_2 c$	Major axis of the second nested tube
10	$M i_2 n$	$0.3 - 0.8 \cdot M a_2 n$	Minor axis of the second nested tube
11	$M a_1 n$	$0.3 - 0.8 \cdot M a_1 c$	Major axis of the first nested tube
12	$M i_1 n$	$0.3 - 0.8 \cdot M a_1 n$	Minor axis of the first nested tube
13	$t_1 c$	0.3–0.7 µm	Thickness of the first cladding tube
14	$t_1 n$	0.3–0.7 µm	Thickness of the first nested tube
15	$t_2 c$	0.3–0.7 µm	Thickness of the second cladding tube
16	$t_2 n$	0.3–0.7 µm	Thickness of the second nested tube

Abstract

Cited By

Figures (6)

Tables (4)

Equations (1)

Optics Letters