Continuous Wavelet Transform as a Tool for Fractal Brownian Motion Analysis and Synthesis
18th International Conference on Mathematical Modeling and Analysis (MMA2013): Abstracts 2013
Andrejs Pučkovs, Andrejs Matvejevs

According to research results, Wavelet coecients of Fractal Brownian process upper interval bound demonstrate more stability at longest time horizons for higher H (Hurst exponent) values. In this case, Wavelet coecients form a pyramidal shape with a high centre at large scales and large time horizons. For lower H (Hurst exponent values) Wavelet coecients of Fractal Brownian process upper interval bound are quite small. Wavelet coecients form convex - concave shape with unexpressed peak in the high-scales and mid-time. Wavelet coecients for greater Hurst exponent are much divergent from average, while the fractional Brownian process with lower Hurst exponent consistently demonstrates a return to average, so the deviation from average is smaller.


Keywords
Wavelet coecients of Fractal Brownian process, wavelet coecients probability density function
Hyperlink
http://www.ut.ee/mma-amoe2013/nmd3/abstraktid09876/Matvejevs.pdf

Pučkovs, A., Matvejevs, A. Continuous Wavelet Transform as a Tool for Fractal Brownian Motion Analysis and Synthesis. In: 18th International Conference on Mathematical Modeling and Analysis (MMA2013): Abstracts, Estonia, Tartu, 27-30 May, 2013. Tartu: 2013, pp.100-100. ISBN 9789949918058.

Publication language
English (en)
The Scientific Library of the Riga Technical University.
E-mail: uzzinas@rtu.lv; Phone: +371 28399196