最大方差模反褶积

MAXIMUM VARIANCE NORM DECONVOLUTION

1977年,Wiggins提出最小熵反褶积(MED)。这个方法依赖于定义一个能衡量信号简单性的模,这个模就是在“数据分析”中已应用多年的规范方差模。最小熵反褶积就是寻找反褶积因子使规范方差模达到最大值。因此,我们直接称这种反褶积为最大方差模反褶积,简记为MVND (Maximum Variance Norm Deconvolution的缩写)。为什么对地震记录能使用最大方差模反褶积呢?Wiggins的文章并没有分析这个问题。这个问题是很复杂的,这篇文章只是试图作一初步讨论。首先,我们给出了规范方差模的两个基本性质;其次,指出了在什么条件下能使用最大方差模反褶积。我们希望这篇文章能起到抛砖引玉的作用,能引起物探工作者的兴趣,以便进行更多的试验和更深入的研究。

In 1977 , Ralph A . Wiggins put forward a new method called "Minimum Entropy Deconvolution". This method depends on defining a norm with which we can estimate the simplicity of the signal. The norm is called "Normalized Variance Norm". As a matter of fact, it has been applied in" data analysis "for several years. In MED method, it is necessary to seek a deconvolution factor to maximize the normalized variance norm. For this reason, this method of deconvolution is called" Maximum Variance Norm Deconvoluton", abbreviated as "MVND". But for what reason can the method of MVND be applied to the processing of seismogram? R. A. Wiggins made no discussion in his paper. This problem is rather complicated and we only try to make a preliminary discussion here. We first presented two basic characteristics of the normalized variance norm, then pointed out in what cases the MVND could be used. We hope that the publication of this article might be an attraction for geophysists and bring them to make more tests and studies and draw forth more valuable contributions.