用于安全告警的光纤周界入侵振动信号智能识别

Intelligent identification of optical fiber perimeter intrusion vibration signal for safety alarm

含噪入侵振动信号易引起时域波形中波峰幅值变化,影响信号识别准确率,为此研究用于安全告警的光纤周界入侵振动信号智能识别方法。采用短时傅里叶变换方法转换原始信号,运用互相关函数法对转换得到的信号进行去噪处理。构建控制信号的匹配滤波器,提取帧信号短时能量和每帧信号的短时过零率,降低非入侵信号的误报率。在此基础上,依据光纤周界中时域特征短时能量与短时过零率的大小,判断是否存在入侵信号,从而实现对入侵振动信号的智能识别。实验结果表明:所提方法能够得到入侵振动信号的时域波形,并且信号识别准确率较高,可实现对安全威胁的准确告警。

The noise intrusion vibration signal is easy to cause the peak amplitude change in the time-domain waveform,which affects the accuracy of signal recognition.Therefore,the intelligent identification method of fiber-optic perimeter intrusion vibration signal for safety alarm is studied.The short-time Fourier transform is used to transform the original signal,and the cross-correlation function is used to denoise the converted signal.The matched filter of control signal is constructed to extract short-time energy of frame signal and short-time zero crossing rate of each frame signal,so as to reduce the false alarm rate of non-intrusion signals.On this basis,according to the time-domain characteristic short-time energy and short-time zero crossing rate in the fiber perimeter,judge whether there is intrusion signal,so as to realize the intelligent identification of intrusion vibration signal.The experimental results show that the proposed method can get the time-domain waveform of the intrusion vibration signal,and the signal recognition accuracy is high,which can realize the accurate warning of security threats.