论文部分内容阅读
本文研究地震道上信号与噪音的频率成分的区别。为了发展适应于相当大空间范围的滤波,特别研究了这些地震道上单个信号子波与噪音子波(同相轴)在波形及频率成分上的基本差别。为此,介绍了一种能压制“噪音”子波提高“信号”子波的所谓“子波”维纳滤波(“Wavelet”Wiener filter);该滤波与普通的维纳滤波大不相同,它是根据子波沿着地震道重复出现的统计特征来辨别信号与噪音的。用数字计算机自动推导用于地震资料的(子波)维纳滤波技术得到了发展。这个滤波是时间相关到无关的范围,滤波是根据算子标定的数据窗的序列而导出的;提供了选择窗孔位置的准则。计算机推导维纳滤波的全部技术用合成地震资料及实际地震资料来阐明。本文附录里讨论了子波维纳滤波与普通维纳滤波的联系,并讨论了子波维纳滤波引出的有限数据窗的影响。
This paper studies the difference between the frequency components of signal and noise in a seismic trace. In order to develop a filter that accommodates a large spatial range, the fundamental differences in waveform and frequency components of individual signal and noise wavelets (in-phase axes) on these traces have been studied in particular. To this end, we introduce a so-called “wavelet” Wiener filter that can suppress the “noise” wavelet to improve the “signal” wavelet. This filter is very different from the ordinary Wiener filter. Discrimination of signals and noise is based on the statistical characteristics of the repeated wavelets along the seismic traces. Automatic generation of (wavelet) Wiener filtering techniques for seismic data has been developed using digital computers. This filter is time-to-irrelevant and the filter is derived from the sequence of operator-calibrated data windows; it provides a guideline for selecting the position of the aperture. The computer derives all the techniques of Wiener filtering from the synthetic seismic data and the actual seismic data. The appendix of this article discusses the relationship between wavelet and ordinary Wiener filtering, and discusses the influence of the finite data window derived from wavelet Wiener filtering.