Study On Optical Communications, Volume. 51, Issue 1, 230123-01(2025)
Optimized Neural Network Method for Temperature Estimation along Optical Fiber
Figures & Tables(8)
Fig. 2. Brillouin frequency shift for the training and testing samples, and corresponding typical Brillouin spectra
Fig. 3. Temperature error along the optical fiber for different algorithms
Fig. 4. Influence of number of hidden layer on ANN training time, computation time, temperature error and its standard deviation
Fig. 5. Influence of number of hidden layer neurons on ANN training time, computation time, temperature error and its standard deviation
Fig. 6. Influence of training goal on ANN training time, computation time and temperature error
Table 1. Average computation time of different algorithms
View tableView in ArticleTable 1. Average computation time of different algorithms
SNR/dB A1/μs A2/ms A3/μs 22 1.62 34.95 35.97 37 1.87 31.57 34.46
Table 2. Temperature error of different algorithms
View tableView in ArticleTable 2. Temperature error of different algorithms
SNR/dB EM/℃ Em/℃ 单斜坡 谱拟合 ANN 单斜坡 谱拟合 ANN 22 10.96 0.75 1.18 2.28 0.23 0.34 37 1.88 0.11 0.63 0.44 0.04 0.22
Tools
Get Citation
Copy Citation Text
Suya LI, Yanwei DONG, Lin LI, Chi ZHANG, Nan LI, Qi NING, Yonghui CHEN. Optimized Neural Network Method for Temperature Estimation along Optical Fiber[J]. Study On Optical Communications, 2025, 51(1): 230123-01
Paper Information
Category:
Received: Oct. 2, 2023
Accepted: --
Published Online: Feb. 24, 2025
The Author Email: DONG Yanwei (joeldream@163.com)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20