			Home About us Contact

Third-order Statistics (third-order + statistics)

Distribution by Scientific Domains

Earth and Environmental Science	75%

Selected Abstracts

Short note: Applications of third-order statistics for the automatic time picking of seismic events

GEOPHYSICAL PROSPECTING, Issue 1 2001
Karthik Srinivasan
We present a new automatic time-picking method based on third-order statistics, namely bicoherence correlation. Contrary to conventional methods, which are based on second-order statistics (i.e. cross-correlation or neural-network trainings), our method is less sensitive to coloured noise as well as the bandwidth of the signal. Bicoherence correlation can also be used for autotracking events in seismic data for an interpretation. [source]

Blind identification of sparse Volterra systems

INTERNATIONAL JOURNAL OF ADAPTIVE CONTROL AND SIGNAL PROCESSING, Issue 7 2008
Hong-Zhou Tan
Abstract This paper is concerned with blind identification for single-input single-output Volterra systems with finite order and memory with the second-order and the third-order statistics. For the full-sized Volterra system (i.e. all its kernels are nonzero) excited by unknown independently and identically distributed stationary random sequences, it is shown that blind identifiability does not hold in the second-order moment (SOM) and the third-order moment (TOM) domain. However, under some sufficient conditions, a class of truncated sparse Volterra systems, where some kernels are restricted to being zero, can be identified blindly and more Volterra parameters can be estimated in TOM than in SOM. Numerical examples illustrate the effectiveness of the proposed methods. Copyright © 2007 John Wiley & Sons, Ltd. [source]

The local isotropy hypothesis and the turbulent kinetic energy dissipation rate in the atmospheric surface layer

THE QUARTERLY JOURNAL OF THE ROYAL METEOROLOGICAL SOCIETY, Issue 603 2004
M. Chamecki
Abstract We test the applicability of the local isotropy hypothesis to surface-layer turbulent flow; turbulent velocities measured with a three-dimensional sonic anemometer are used for this purpose, and the predictions of local isotropy for the spectra, second- and third-order structure functions are assessed against measured data. Also investigated are scale interactions via the correlation between velocities and velocity increments, and the ability of isotropic spectral models to reproduce measured spectra. In general, second-order structure functions display a narrower inertial range than the corresponding spectra; both the known effects of path-averaging and the predictions of the spectral models show that the sonic anemometer is unable to resolve the whole inertial range, even at a measurement frequency of 60 Hz. We confirm previous results that unstable runs tend to be more isotropic, but find that, for third-order statistics, isotropy does not hold well for the data analysed. Turbulence intensity, and not atmospheric stability, plays a determining role on the correlation coefficient between velocities and velocity increments. The observed anisotropic behaviour has important implications for the calculation of the turbulent kinetic energy dissipation rate from Kolmogorov's four-fifths law, whose estimates are consistently smaller than those from the inertial range of the spectrum or the structure functions. Copyright © 2004 Royal Meteorological Society [source]