Research On Signal Analysis Technology In Optical Fiber Sensing Applications

Compared with the traditional sensor,the optical fiber sensor has the advantages of small size,anti-electromagnetic interference,corrosion resistance,light weight,etc.For different applications and different sensor working principle,this paper takes the environment-based optical fiber security system as an example.With the mature hardware system of perimeter security detection system based on the principle of Mach-Zehnder fiber interferometer,this paper mainly studies basic theory and experiment of the techniques used in the signal interpretation of optical fiber sensing applications.The mainly work is as follows:(1)This paper makes a theoretical analysis of different types of optical fiber sensors,proposes an optical fiber sensing signal analysis system model which gives the signal analysis process from signal acquisition to signal recognition.(2)This paper analyzes the data characteristics of different event signals,and propose the optimization design of feature extraction methods in time domain and frequency domain respectively.The project achieved online production template by KMeans algorithm,improve the recognition accuracy of the system.(3)In view of the problem of parameter adjustment in system,the paper proposes a parameter adaptive adjustment strategy theory model based on Q learning algorithm and KMeans clustering algorithm.This method is implemented in engineering,which makes the system operation more concise and reduces the system false alarm rate.(4)To reduce the false alarm rate of the system further,the paper propose multi-template recognition.Based on only two types of intrusion and non-intrusive templates,the system can save multiple typical environment templates and use multiple template matching to identify event numbers to further improve system performance.