Open AccessApplication of the Prony method for biometric data analysis
Author(s) -
M V Elenets,
М М Немирович-Данченко
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1862/1/012007
Subject(s) - transformation (genetics) , series (stratigraphy) , computer science , exponential function , representation (politics) , process (computing) , biometrics , algorithm , artificial intelligence , linear map , pattern recognition (psychology) , mathematics , data mining , mathematical analysis , paleontology , biochemistry , chemistry , politics , gene , political science , pure mathematics , law , biology , operating system
The paper deals with the processing of numerical series using the spectral estimation methodology. The methodology is based on the Prony transformation. The conversion lets you represent the original number series as a new series. The new series is a linear combination of exponential functions. The article presents the basic concepts and relations of the Prony transformation, and also analyzes the behavior of individual components of the transformation for typical cases. The features of the Prony method are analyzed. Shown are the individual stages of the algorithm, the problems of a specific implementation. The Prony method was used to process the recordings of the electroencephalograms of the operators in their mental representation of movements. This allows us to assume the possibility of using the described methodology in identifying EEG correlates of motor imagination in the space of the roots of polynomials