应用小波分析提高MT资料信噪比

IMPROVING THE S/N RATIO OF MT DATA BY WAVELET ANALYSIS

大地电磁测深法（ＭＴ）野外观测得到的原始资料是包含多种频谱的时间序列，这种时间序列中常常包含固定源噪声。由于这种噪声的频率相对稳定，应用小波分析容易得到一个或多个以这种噪声为主的时间序列块。对实验数据应用小波变换，将时间序列分解成不同频率的多个时间序列块，压制噪声比较集中的时间序列块后重建时间序列。通过与未作这种处理的时间序列的谱分析结果对比。

Magnetotelluric Sounding (MT) is one of the basic prospecting methods for oil and gas. Noise is a vital affect on its precision and a stumbling block to its development. It is necessary to hunt a good way to lower the noises. Wavelet analysis could decompose the composite signal which consists of several components of different frequencies into a series of signal blocks, so it is an effective method for dissociating the noise from signal. The authors detects that MT original data are time series. This time series often composes several spectra fltied by band pass filter, and sometimes composes regular frequency noises caused by a certain source. In order to filter this regular noise, the authors used wavelet analysis to process the added noise MT data. Judging by the comparison with the synthetic MT data, the result is perfect. Thus the authors consider that wavelet analysis is useful for improving the S/N ratio of MT data.