基于Fisher检测算法的次声信号检测

In order to reduce the number of false noise signals and reduce the false alarm rate of the algorithm when Fisher detection algorithm is used for infrasound detection, this paper applies a similarity clustering algorithm based on Euclidean-distance, that is, the Fisher detection algorithm will set a lower Fisher threshold to obtain as many significant infrasonic events as possible. At the same time, the beamforming algorithm is used to calculate the azimuth Angle, velocity, data similarity, frequency and other parameters corresponding to the significant infrasound event, and then the parameters are clustered according to the similarity rules. In order to verify the feasibility of the clustering algorithm in reducing the false alarm rate, this paper uses the infrasound data of Tonga volcanic eruption for event detection. The results show that the similarity clustering algorithm can effectively reduce the number of false detected noise signals and improve the accuracy and reliability of Fisher detection algorithm.