CFAR Detection Algorithm And Implementation Of Fiber Optical Vibration Signal

Safety in the military, industry, agriculture and other fields has a very important role, but for the border, a long-distance security detection monitoring system such as oil and gas pipelines, complex security work inefficient. Fiber optic vibration signal detection system based on ΦOTDR, can effectively solve the problems of the long-distance monitoring, but as the threshold detection system cannot be adaptive, unknown false alarm rate, false alarm rate is not controllable, algorithm difficult problems, such as the optical fiber vibration signal detection system application in real life. This article mainly from the constant false alarm rate (constant false alarm rate, CFAR) testing perspective, the original vibration signal preprocessing, the pretreatment of data to establish the corresponding mathematical model, through the CFAR detection algorithm, the detection threshold can be adaptive, and ensure constant false alarm rate and improve the accuracy of detection, and the algorithm is successfully implemented in the DSP hardware platform. This paper mainly completed the following work:First of all, the original optical fiber vibration signal preprocessing, including reducing sampling rate, multi pulse cumulative difference, signal fusion pretreatment method to preprocess the signal. Through the pretreatment method can effectively facilitate the feature extraction of vibration, the next step in the design of the detection algorithm.Secondly, the basic principle of signal detection algorithm is studied, including Bayesian detection theory, the rule of Norman Pearson, CFAR detection and so on the classical test theory. Probability density after preprocessing the data statistical analysis, validate the feasibility of CFAR detection and the theory is correct.Furthermore, according to the geographical location and the actual need, cell averaging constant false alarm rate (CA-CFAR) and order statistic constant false alarm rate (OS-CFAR) has carried on the theoretical analysis and the actual data simulation.Finally, the design of CFAR detection algorithm is implemented on DSP hardware platform, in order to ensure the consistency of the background reference unit segmenting the background reference unit for processing. In addition, in order to save the time in the DSP, CFAR detection algorithm for the algorithm optimization.