Convolutional Neural Network-assisted Optical Microfiber Interferometer for Refractive Index Measurement

A team of researchers from Nanchang Hangkong University has made significant advancements in the field of optical fiber sensing with the development of a convolutional neural network (CNN)-assisted optical microfiber interferometer. The proposed system has demonstrated high accuracy in measuring surrounding refractive index (RI) with a coefficient of determination (R-2) of 0.992. The findings suggest that the system has good potential for application in medical and environmental monitoring fields.

Key Takeaways:

• The researchers proposed a 1-D convolutional neural network (1-D CNN) to extract information from a U-shape single-mode-tapered seven core-single (STSS)-mode structure interferometer.
• The 1-D CNN method demonstrated higher accuracy in measuring surrounding refractive index (RI) compared to the traditional dip/peak wavelength tracking method.
• The coefficient of determination (R-2) of the RI predicted by 1-D CNN is as high as 0.992, indicating a high level of accuracy.
• The effects of bandwidth and sampling points on the RI measurement accuracy were studied, and it was found that reducing the spectral sampling resolution and bandwidth can improve the measurement accuracy to above 0.990 and 0.989, respectively.
• The proposed RI optical fiber sensing system has good application potential in medical and environmental monitoring fields.
• The research was conducted by Bin Liu and his colleagues from Nanchang Hangkong University, with support from the National Natural Science Foundation of China (NSFC) and the Natural Science Foundation of Jiangxi Province.
• The research has been peer-reviewed and published in the IEEE Transactions on Instrumentation and Measurement.

Statistics:

• The coefficient of determination (R-2) of the RI predicted by 1-D CNN is 0.992, indicating a high level of accuracy.
• The effects of bandwidth on the RI measurement accuracy showed that a reduction in bandwidth can improve the measurement accuracy to above 0.990.
• The effects of sampling points on the RI measurement accuracy showed that a reduction in sampling points can improve the measurement accuracy to above 0.989.

Sources:

• Liu, B., et al. "Convolutional Neural Network-assisted Optical Microfiber Interferometer for Refractive Index Accurately Measurement." IEEE Transactions on Instrumentation and Measurement (2025): 74.
• Bin Liu, Nanchang Hangkong University, Key Lab Optoelect Information Percept and Instrumentation, Nanchang 330063, People's Republic of China. Email: lbin@ncapu.edu.cn.
• National Natural Science Foundation of China (NSFC).
• Natural Science Foundation of Jiangxi Province.
• IEEE Transactions on Instrumentation and Measurement.