Disturbance Location And Recognition Of Airport Perimeter Based On Fiber Mach-Zehnder And Φ-OTDR Combined System

Distributed optical fiber perimeter technology is a new detection technology,which has the advantages of fast response speed and high real-time performance.It was widely used in the security monitoring fields such as national defense borders and civil aviation airports.It is mainly divided into interferometric optical fiber sensing systems and backscattering sensing systems.The interferometric system has the capability of broadband detection,but the location ability of airport perimeter invasion is weak.The backscattering sensing system can not only accurately locate the location of the intrusion disturbance point,but also can monitor the location of multiple intrusion sources in real time.The backscattering system has the disadvantage of weak recognition ability.The current airport perimeter security technology is mainly based on traditional technologies,which will make the airport perimeter security technology have certain disadvantages in real-time positioning and pattern recognition.This paper not only combines the advantages of the two systems on the basis of the original single system to build a system based on the combination of phase-sensitive optical time domain reflectometer and Mach-Zinde interferometer,but also optimizes the system’s early warning method on the basis of the combined system.In addition,this paper also studies the signal processing methods of the Φ-OTDR system and the Mach-Zehnder system.In summary,this solution enables the joint system to have the best positioning capability and pattern recognition capability.The main work of this paper is as follows:1.This article first studies the sensing principles of the Φ-OTDR system and the Mach-Zehnder system and the corresponding signal processing methods.Then this article theoretically analyzes the feasibility,advantages and disadvantages of the combination of the two systems.Finally,based on the research of the research group,this paper not only optimizes the system early warning mode and signal acquisition software,but also adds the function of signal software filtering.Through the above-mentioned research,this solution can make the signal be preliminarily filtered before saving to improve the signal-to-noise ratio.This scheme enables the system to reduce the false alarm rate of the system when performing endpoint detection,thereby achieving the effect of improving the system’s early warning performance.2.Aiming at the problem of low signal-to-noise ratio and large positioning error caused by severe noise interference of Φ-OTDR signal,this paper proposes a Φ-OTDR denoising location algorithm based on LCD combined wavelet threshold.Through the use of field experimental data,the experimental results show that this method can reduce the interference of noise to the signal(signal-to-noise ratio increased by 4.4d B),thereby reducing the positioning error.3.Aiming at the problems of single feature extraction of Mach-Zehnder signal and low recognition efficiency,this paper studies an improved recognition algorithm combining local feature scale decomposition and deep belief network.In the field experiment verification,this method not only improves the signal-to-noise ratio by 8 d B,but also greatly improves the recognition efficiency(the recognition rate reaches more than 95%).