Calculation of Analogs for the Largest Lyapunov Exponents for Acoustic Data by Means of Artificial Neural Networks. / Chernykh, G.A.; Kuperin, Y.A.; Dmitrieva, L.A.; Navleva, A.A.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics): 13th International Symposium on Neural Networks, ISNN 2016; St. Petersburg; Russian Federation; 6 July 2016 through 8 July 2016; Code 177689. Vol. 9719 Springer Nature, 2016. p. 108-114.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review
}
TY - GEN
T1 - Calculation of Analogs for the Largest Lyapunov Exponents for Acoustic Data by Means of Artificial Neural Networks
AU - Chernykh, G.A.
AU - Kuperin, Y.A.
AU - Dmitrieva, L.A.
AU - Navleva, A.A.
PY - 2016
Y1 - 2016
N2 - A method for calculating the largest Lyapunov exponents analogs for the numerical series obtained from acoustic experimental data is proposed. It is based on the use of artificial neural networks for constructing special additional series which are necessary in the process of calculating the Lyapunov exponents. The musical compositions have been used as acoustic data. It turned out that the error of the largest Lyapunov exponent computations within a single musical composition is sufficiently small. On the other hand for the compositions with different acoustic content there were obtained various numerical values Lyapunov exponents. This enables to make conclusion that the proposed procedure for calculating the Lyapunov exponents is adequate. It also allows to use the obtained results as an additional macroscopic characteristics of acoustic data for comparative analysis.
AB - A method for calculating the largest Lyapunov exponents analogs for the numerical series obtained from acoustic experimental data is proposed. It is based on the use of artificial neural networks for constructing special additional series which are necessary in the process of calculating the Lyapunov exponents. The musical compositions have been used as acoustic data. It turned out that the error of the largest Lyapunov exponent computations within a single musical composition is sufficiently small. On the other hand for the compositions with different acoustic content there were obtained various numerical values Lyapunov exponents. This enables to make conclusion that the proposed procedure for calculating the Lyapunov exponents is adequate. It also allows to use the obtained results as an additional macroscopic characteristics of acoustic data for comparative analysis.
U2 - 10.1007/978-3-319-40663-3_13
DO - 10.1007/978-3-319-40663-3_13
M3 - Conference contribution
SN - 978-331940662-6
VL - 9719
SP - 108
EP - 114
BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PB - Springer Nature
T2 - Advances in Neural Networks
Y2 - 6 June 2016 through 8 June 2016
ER -
ID: 8625443