The possibility of the classification of infrasonic signals using methods for checking statistical hypotheses

This work is devoted to the possibility of the classification of signals using methods for checking statistical hypotheses. On the basis of analysis of the typical peculiarities of signals that belong to the same class we performed an empirical derivation of the shape of the class. The mechanism of the definition of the separability of signals of each class, as well as the level of criterion for defining critical domain were suggested. A classification algorithm was obtained on the basis of the research. The efficiency of the suggested methodology was tested on the problem of separability of infrasonic signals recorded in the atmosphere.