Font Size: a A A

Research On The Intrusion Recognition Method For The Fiber Perimeter Security System Based On Dual Mach-Zehnder

Posted on:2019-12-05Degree:MasterType:Thesis
Country:ChinaCandidate:Z C LiFull Text:PDF
GTID:2428330593951460Subject:Optical Engineering
Abstract/Summary:PDF Full Text Request
With the development of optical fiber sensing technology,Dual Mach-Zehnder interference vibration sensor is becoming mature.It has been widely used in optical fiber perimeter security monitoring field.However,the sensing system is still lacking in the Diversity of intrusion events and the accuracy of recognition.What?s more,the way to deal with the intrusion signal needs to be improved.This paper briefly introduces the development status of optical fiber perimeter security system and analyzes its advantages.Then,the paper focus on the following aspects:1.We introduce the principle of the Dual Mach-Zehnder fiber perimeter security system.Then,we analyze significance of pattern recognition for intrusion events.2.The time frequency information of the interference signal is obtained by the method based on Short-time Fourier transform,and we analyze characteristics of the time frequency information caused by different behaviors.We can reduce the interference of the noise signal by the difference between the the noise signal and the intrusion signal.3.An endpoint detection method based on Short-time Fourier transform(STFT)is proposed.The core of the method is getting the relationship of time and the average amplitude of high-frequency.The result of the proposed method endpoint detection is accurate,anti-interference and stable.And the method is applied to practical engineering.4.An integrated scheme to distinguish invasive events based on Short-time Fourier transform(STFT)and singular value decomposition is proposed for optical perimeter security system.The result of 360 sets of experiments demonstrates that the proposed scheme can discriminate three common invasive events(climbing the fence,knocking the cable,and waggling the fence)with a high recognition rate.The recognition rates of the three events were all above 90%.
Keywords/Search Tags:Fiber vibration sensing, DMZI system, Time frequency analysis, Endpoint detection, Pattern recognition
PDF Full Text Request
Related items