为了提高光纤周界安防系统中扰动信号端点检测的性能,提出一种新型端点检测方法。首先通过计算信号的短时能量和过零率分布分别找出信号中短时能量比和过零率满足一定门限的信号点,通过对这两信号点进行一定的加权确定扰动信号的端点位置。与离散小波提取起始点信号相比,本文方法由于综合了短时能量和过零率法的优势,能够适应更多的扰动事件类型,通过加权极大提高了端点检测的精度,并且运算时间大幅降低,提高了系统的实时性。实验结果表明,本文算法在10 MHz的采样速度和采集0.3s数据情况下,93.47%的数据定位误差低于10m,计算时间少于0.3s。 In order to improve the performance of the endpoint detection of disturbing signals in the perimeter security system of optical fiber, a new endpoint detection method is proposed. Firstly, the short-time energy and zero-crossing rate distribution of the signal are respectively used to find out the signal points which meet the certain short-time energy ratio and the zero-crossing rate. The end points of the disturbance signal are determined by weighting these two signal points. Compared with discrete wavelet extraction of the initial point signal, this method can adapt to more types of disturbance events by combining the advantages of short-time energy and zero-crossing rate method, greatly improve the accuracy of endpoint detection by weighting, Reduce, improve the system’s real-time. The experimental results show that the proposed algorithm achieves 93.47% of the data localization error less than 10m with a sampling rate of 10 MHz and 0.3s data acquisition, and the calculation time is less than 0.3s.