Study On OTDR Dynamic Range Improvement And Event Detection

Dynamic range is an important index that affects the measurement effect of Optical Time Domain Reflectometry(OTDR)and determines the maximum Optical fiber length that OTDR can detect.The range resolution and dynamic range are two interrelated parameters for traditional optical time domain reflectors.In the process of OTDR testing,the noise in the link will affect the event information in the waveform will be covered by noise,and the test accuracy of OTDR system will be reduced.OTDR curve detection also includes event location detection and event type detection.Therefore,it is very important to improve the performance of OTDR system,especially the dynamic range and the accuracy of the event detection algorithm.Based on the above reasons,this paper studied the improvement of the dynamic range of THE OTDR system and the detection algorithm of event location and type,and completed the following work:The sampling rate directly affects the distance accuracy of OTDR test curve.The higher the sampling rate,the smaller the distance accuracy and the higher the accuracy of event location detection.Implemented in 50 MHZ sampling frequency under the simulation analysis of equivalent sampling frequency of 3.2 GHz,to verify the time phase error of the reconstructed signal frequency point location and amplitude of distortion size,adopting a based on equivalent time sampling phase error detection method to accurately detect the various channels of time phase error,the experimental data proved that the equivalent horizontal and vertical equivalent sampling sampling is better than reduce the influence of phase error in the time.Now improve the dynamic range of the OTDR system is the main means to increase test pulse width,but increase the pulse width at the same time reduces the time resolution of the system,in order to solve this problem,this paper designed a kind of gray code coding scheme of OTDR test,using the test of single pulse echo waveform superposition of shift the way of the processing of gray code coding.During the actual measurement of gray code detection,sharp spines will appear near the Fresnel reflection,which will affect the accuracy of subsequent event detection.Constant false alarm detection method can effectively eliminate the influence of sharp spines.OTDR curve analysis includes signal denoising and event detection.In this paper,5-layer wavelet transform is applied to the echo signal of monopulse test,Haar wavelet is selected as the basis function,and improved threshold method is adopted to obtain better denoising effect than traditional soft and hard threshold method.For event detection of de-noised signal,the traditional two-point method is not accurate in event type and location.In this paper,reflecting events and non-reflecting events are detected separately,reflecting events are detected by wavelet decomposition,and high frequency signals are reconstructed by the first layer of detail coefficient.For the detection of non-reflected events,the singularity detection method is used to detect the location of events between adjacent reflected events.The attenuation value of each non-reflected event is calculated by the least square method,and the event is classified by the attenuation value.