Fiber perimeter security intrusion event identification method and apparatus based on integrated characteristics

Abstract

The invention discloses a fiber perimeter security intrusion event identification method and apparatus based on integrated characteristics. The fiber perimeter security intrusion event identification method based on integrated characteristics includes the steps: feeding the signal disturbing the start point in each channel of an all-phase filter bank each channel to perform frequency domain separation processing, calculating the normalization power value of the output signal of each channel, wherein the plurality of normalization power values are output in parallel; integrating the normalization power values and the time domain zero-crossing rate of the whole section of disturbance signals to generate a characteristic vector, wherein the characteristic vector contains two aspects of information: the time domain and the frequency domain; and feeding the characteristic vector in a radial primary function neural network to realize intrusion motion quick high precision identification. The fiber perimeter security intrusion event identification apparatus based on integrated characteristics includes an analog-to-digital converter and a DSP (Digital Signal Processor) device. The fiber perimeter security intrusion event identification method and apparatus based on integrated characteristics can accurately distinguish four types of intrusion events. Besides, compared with a current high precision intrusion event identification classifier, the DMZI (dual Mach-Zehnder interferometer) intrusion motion identifier has great advantages of work efficiency.

Description

technical field [0001] The invention relates to the technical field of digital signal processing, in particular to a method and device for identifying fiber optic perimeter security intrusion events based on comprehensive features. Background technique [0002] With the continuous maturity of optical fiber and optical fiber communication technology, optical fiber sensor has been developed rapidly. Perimeter Security System Based on Optical Fiber Sensing Technology [1][2][3] It is also gradually being valued in the field of security. with conventional infrared [4] ,electric fence [5] Compared with other security systems, the optical fiber sensing system can monitor various disturbances that are directly touched or indirectly transmitted to the optical fiber in real time, with higher sensitivity and thus more practical value. As a phase-modulated fiber optic sensor, the dual Mach-Zehnder interferometer (DMZI) distributed fiber optic sensing system [6][7] It has the advant...

AI Technical Summary

Problems solved by technology

[0004] Among them, although the wavelet identification method proposed in literature [14] can be characterized by the energy of different frequency bands obtained by multi-level decomposition, the eigenvector [17] , but the amount of calculation will increase with the increase of the number of wavelet decomposition layers. In addition, the feature vector only considers the frequency domain features and lacks sufficient and rich time domain information, so the recognition accuracy is low, and only three types of intrusions can be recognized event;

[0005] However, the Empirical Mode Decomposition (EMD) method proposed in literature [16] needs to obtain the kurtosis value of the intrinsic mode function (IMF) in the decomposition process one by one, and the kurtosis value obtained by combining these kurtosis values Four types of common actions can be recognized with high precision by classifying the feature vectors. However, these IMFs need to go through multiple complex iterations to obtain, so the recognition efficiency is not high and affects the practicability.

Images