Abstract: In order to gain effective signal from synthetic seismogram disturbed by random noise and strong surface wave,this paper,to start with,recovers surface wave event by the method of Singular Value Decomposition（SVD）.After eliminating surface wave from synthetic seismogram,the method of wavelet transform is used for suppressing partial random noise.By means of relationship among event trend,first and second singular value of SVD,a new method is introduced to determine the optimal slope of the surface wave.We show that it is effective to process the synthetic seismogram．

摘要： 为了从含强面波干扰和随机噪声的地震记录中提取有效信号,将每条面波同相轴旋转为水平后,用奇异值分解（SVD）方法恢复出来,再从原始地震记录中消除面波,最后用小波变换方法滤除部分随机噪声。根据同相轴走向与SVD第一奇异值、第二奇异值之间的关系,提出了一种新的判别最佳面波同相轴走向的依据,并用于处理合成地震记录,有效地去除了面波。