Localization of non-reflective events in OTDR data combining DWT with template matching

A new method of detecting and localizing non-reflective events in optical time-domain reflectometry (OTDR) data based on biorthogonal wavelet transform and template matching is proposed in this paper. According to the non-stationary characteristics of OTDR logarithmic measurement data, discrete wavelet transform(DWT) is used to detect and coarsely locate the non-reflective events, and then the locations are determined at high accuracy by using template matching. The result shows that without the requisition of pre-treatment like de-noising, the proposed method is able to locate the events accurately both in the low and high SNR levels, which proves that the algorithm is robust and adaptive for noisy OTDR data.