Research And Realization Of Signal Processing Method In Fiber Temperature-measured System

With the deep research on optical fiber and optical fiber devices, the optical fiber sensor that is used to test devices has gained more and more attention from people. Recently, the optical fiber sensing technology gets rapid development and has been used widely in the scientific research and engineering application field. The distributed optical fiber temperature sensor , as an important branch of optical fiber sensor,distributed optical fiber temperature sensor has been gradually used in all kinds of projects as its technology becomes mutual.The most important part of distribute fiber temperature-measured system are signal collecting, data processing and background software programming. They decide the collecting precision, collecting speed and final response time. How to improve the processing speed and how to coordinate data transmission between different parts are most important in this system.This thesis deeply analyzes the temperature-measuring principle of distributed optical fiber and poses the reform plan for the optical path of system,high speed acquisition circuit and software filter.Through the improvement of each part of the system,it improves the response time and measuring accuracy and so many other performance indexes of the temperature-measuring system of distributed optical fiber.This thesis detailedly introduces the signal processing method of the optical path of system part,high speed acquisition circuit part and software design part of temperature-measuring system of distributed optical fiber. This thesis gives the best splitting ratio of the fiber coupler and makes signal-to-noise ratio of the optical path of system part higher.It designs a high speed cache which can input the data into the DSP to process.In the part of software design, the system usually demands a long measuring period, while guarantee time and space resolution of the system, a method called adaptive threshold filter based on wavelet was proposed in this paper.This parameters of the non-line threshold filtering function in different wavelet scales is based on training signals,then using the adaptive threshold filtering wavelet method to de-noising the temperature signal. The experiments indicate that the measuring period is greatly reduced, and the system becomes more precise.