02.12.2014 Views

Annual Report 2000 - WIT

Annual Report 2000 - WIT

Annual Report 2000 - WIT

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Ý<br />

ì<br />

á<br />

ä<br />

é<br />

å¤æDçzè<br />

ä<br />

é<br />

åPæDçzè<br />

è<br />

è<br />

Ý<br />

ì<br />

á<br />

í<br />

í<br />

ß<br />

166<br />

That is what we call the Green's function for random media. It is now possible to<br />

describe seismic pulses as<br />

(20)<br />

ñ8ùúï@é•ó¤ô#û ß<br />

ö ÝÞYßhàâá~ã<br />

where<br />

êNì—÷ø<br />

.<br />

is the Fourier transform of the input signal ö ø<br />

ÝÞná<br />

In order to obtain explicit analytical results for the Green's function in the time domain<br />

(that is to evaluate the integral in equation (19)), we have to introduce some simplifications.<br />

To do so, we study the behavior of equations (17) and (18) in the Fraunhofer<br />

approximation. The latter is characterized by a large wave üþý ä parameter so<br />

that the mean of phase fluctuations can be neglected. The Fraunhofer approximation<br />

becomes valid for large travel-distance . From the behavior of scattering attenuation,<br />

it turns out that for large travel-distances only the low-frequency components of<br />

the transmitted pulse survive. Thus, with increasing à à<br />

÷ø<br />

not only à ÚNÛ but also ÝÎÿ Û<br />

effectively increases. Therefore, equation (19) can be reduced to:<br />

á é¡<br />

Ü£¢¥¤§¦©ÝšÞYßhàâáFã<br />

(21)<br />

éËïN ¥<br />

ñ8ùï@é•óô#û<br />

êNìzí<br />

Relative standard deviation of the phase increment<br />

a=10m, sigma=4%<br />

0.8<br />

a=40m, sigma=4%<br />

a=160m, sigma=4%<br />

a=10m, sigma=8%<br />

sigma<br />

0.6<br />

a=40m, sigma=8%<br />

a=160m, sigma=8%<br />

0.4<br />

0.2<br />

0<br />

200 400 600 800 1000<br />

Travel-distance [m]<br />

Figure 2: The relative standard deviations of the phase increment for a 2-D exponentially<br />

correlated random medium.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!