光通信研究Issue(2)：99-104,6.DOI:10.13756/j.gtxyj.2025.240159
基于1D-CNN的射频强度温度传感研究
Research on RF Intensity Temperature Sensing based on 1D-CNN
丁美琪 1桂林 2王子怡 1尚荻森 1钱敏 1李乾坤1
作者信息
- 1. 上海第二工业大学智能制造与控制工程学院,上海 201209
- 2. 上海第二工业大学计算机与信息工程学院,上海 201209
- 折叠
摘要
Abstract
[Objective]In order to improve the accuracy and efficiency of temperature sensing,the application of Microwave Pho-tonic Filter(MPF)based on One-Dimensional Convolutional Neural Network(1D-CNN)in Radio Frequency(RF)intensity temperature sensing is studied.[Methods]The MPF system based on Mach-Zehnder Interferometer(MZI)structure is built ex-perimentally,and the RF spectral data of 20～70℃under the condition of notch depth of 8.1 dB are collected by changing the am-bient temperature.30 sets of data are collected under each temperature condition.Then the 1D-CNN structure is designed and op-timized by greedy strategy to determine the number of network layers,the size of the convolutional kernel,the size of the pooled kernel and the type of activation function.The model is trained with the training set data and validated with the test set data to opti-mize the model parameters for optimal performance.Its nonlinear mapping capability is used to extract features from RF spectral data to achieve high-precision demodulation of RF intensity and temperature changes.Finally,the Root Mean Square Error(RMSE)is used as the evaluation index,and the performance of 1D-CNN is compared with the traditional algorithms(maxi-mum-value method,centroid method and Gaussian fitting method)to analyze its performance under different temperature condi-tions.[Results]The experimental results show that the RMSE of the prediction model based on 1D-CNN reaches the order of 10-3,while the RMSE of the traditional algorithms is usually in the order of 10-1.Compared with the traditional Gaussian fitting algorithm,the demodulation speed of the 1D-CNN-based algorithm is improved by 2.72 times.1D-CNN shows high stability and low error under different temperature conditions.[Conclusion]1D-CNN has significant advantages in dealing with complex nonlinear relationships and feature extraction,not only superior in computational efficiency and robustness,but also effective in dealing with noise and environmental interference.The research in this paper provides new ideas and methods for the application of MPF in the field of RF intensity temperature sensing.关键词
一维卷积神经网络/微波光子滤波器/光纤传感/温度传感/射频强度Key words
1D-CNN/MPF/fiber optic sensing/temperature sensing/RF intensity分类
信息技术与安全科学引用本文复制引用
丁美琪,桂林,王子怡,尚荻森,钱敏,李乾坤..基于1D-CNN的射频强度温度传感研究[J].光通信研究,2025,(2):99-104,6.
