数学物理学报 ›› 2000, Vol. 20 ›› Issue (zk): 650-657.

• 论文 • 上一篇    下一篇

褶积模型参数的估计及模拟试验

胡必锦   

  1. 华中科技大学 武汉430074
  • 收稿日期:1998-11-16 修回日期:1999-03-09 出版日期:2000-11-10 发布日期:2000-11-10

Parametric estimation of convolutional-Model and simulation

HU Bi-Jin   

  1. department of mathematics|huazhong university of science and technology,wuhan 430074
  • Received:1998-11-16 Revised:1999-03-09 Online:2000-11-10 Published:2000-11-10

摘要:

在褶积模型中关于混合相位子波反演问题一直受到人们的广泛的关注,这类模型在地震记录反演过程中起着特别重要的作用.该文在引入介入子波的概念的基础上拟订了一套算法将统计递归算法与以高阶标准累积矩作为判据的迭代算法结合起来,最终获得的不是反子波的估计,而是混合相位子波的估计.模拟试验表明:该文所拟订的算法对混合相位子波(其反子波具有有限长度)的估计是有效的.

关键词: 统计递归算法, 混合相位子波, 介入子波, 相对反子波标准累积矩

Abstract:

Theinverseproblemonmixedphasewaveletofconvolutionmodelwhichtakes impotentpoleininverseprocessofseismicrecordshaswidelybeenpayedattentionin.In thepaper,thealgorithmdesignedinthebaseonconcptofmiddlewaveletcombinesstatisticalrecursivealgorithm withiterationformatinwhichcriteriafunctionisahigherorder standardizingcumulant.Usingthealgorithminthepaper,estimationofmixedphasewaveletisobtained,andnotoneofantiwavelet.Simulationshawsthatthealgorithmisavailable totheestimationofmixedphasewaveletwhoseantiwavelethasafiniteenght.

Key words: Statisticalrecursivealgorithm, Mixedphasewavelet, Middle-wavelet, Relative anti-wavelet, Standardizingcumulant

中图分类号: 

  • 62F