f-x域内提高地震资料的信噪比

Improvement of signal/noise ratio of seismic data in f-x domain

在 f-x 域,地震记录可以表示成有限个调频调幅正弦波的叠合,叠合的结果是一个连续且光滑的函数。我们可以用一个简单的 n 次多项式去逼近这个函数。本文提出了一种在 f-x 域用最小平方拟合原理消除随机噪声的新方法,并证明了在噪声为自噪以及空间采样足够密的条件下,拟合结果与噪声无关,即 f-x 域最小平方拟合处理可以有效地消除记录上的随机白噪,而对有效信号的破坏甚微。由于理论上允许振幅值及同相轴形态在空间上发生变化,因而该方法不具有混波作用,是高分辨率处理的理想工具。

Seismograms in f-x domain can be expressed as the superposition of finite sine waves which have experienced frequency modulation and amplitude modula- tion.The superposed result is a continuous and smooth function,which can be approximated by the Nth power of a polynomial. A new method is proposed which removes random noises in f-x domain by using least square fit principle.It has been proved that the fit result is indepen- dent of noises in the case of white noise and enough dense spacial sampling,in other words,the least square fit processing in f-x domain can well eliminate random white noise,with very slinght damage to useful signals.This method, without wave mixing,is ideal for high resolution data processing because the spatial variations of amplitude and event shape are permitted theoretically.