Journal de génie informatique et des technologies de l'information

On Studying Characteristics of Periodic Signals Based on the Cluster Analysis

Znak Vladimir

Periodic (harmonic or frequency-modulated) signals, which reflect the behavior of some real processes in the course of a discrete time, are widely used. At the same time, such signals are quite often corrupted by different kinds of noise. The examples of such signals are seismic waves which are deformed by a noise in the course of their propagation. Therefore, much attention is given by many researchers to problems of processing and studying such data. The objective of this paper is studying parameters and characteristics of periodic signals by using the cluster analysis methods. The efficiency of the approach proposed is demonstrated by attracting a periodic signal model.