Data examples of logarithm Fourier-domain bidirectional deconvolution |

As we discussed in the previous section, we can decompose the arbitrary data
into three parts: the reflectivity series
, the minimum phase wavelet
and the maximum phase wavelet
:

We wish to solve for the deconvolution filters and , which should be the inverses of wavelets and :

From equation 2, we know that is minimum phase and is maximum phase. If we know the deconvolution filters and , we can get reflectivity series as follows:

Next we transform our problem into the Fourier domain. We use capital letters to denote variables in the Fourier domain:

We use to denote the logarithm of the product of and :

Our problem then becomes

where has become our new unknown in bidirectional deconvolution, and we want to update it in each iteration. After some derivation (Claerbout et al., 2011), we get, in the time domain,

where means cross-correlation and is the hyperbolic penalty function.

By Newton's method (using the only first 2 terms of the Taylor expansion), we can calculate the step length
:

(8) |

Because we use Newton's method, this step length calculated above is not the final value. To obtain the final step length at each iteration, we need another iteration (nested or second-order iteration):

Given the update directions (both for the unknown and for the residual ) and the step length of the update, we have everything we need for each iteration. We can iterate to convergence.

Data examples of logarithm Fourier-domain bidirectional deconvolution |

2011-09-13