A denoising and positioning method of long-distance fiber optic perimeter security system based on φ-OTDR

Abstract

In this paper, we propose an improved denoising and positioning method for fiber optic perimeter security system based on the φ-OTDR. We combine the lifting scheme of the discrete wavelet transform (LSDWT) and the total variation (TV) based image denoising algorithm to suppress the background noises and magnify the man-made intrusion signals, and the positions of man-made intrusion can be acquired with the denoised signals through the variance-based positioning method, which mainly determines the man-made intrusion signals by calculating the variance of the temporal signal. A series of experiments have been carried out to validate this method. In the scheme, three types of man-made intrusion behaviors such as knocking, waggling, and tapping were implemented at the end of the 40 km optical cable, while the background noises such as natural noise and sound of car driving were continuing. Compared with the original signals, the peak signal-to-noise ratio (PSNR) of the signals after the proposed scheme has been greatly enhanced, and a ±15m positioning error can be realized at the end of the 40 km fiber cable. Therefore, the proposed method can provide new thinking to address the high positioning error at the end of the long-distance optical fiber cable.