本文提出用最小熵准则重新建立反射系数谱的方法。作者将新方法(FMED)与传统的最小熵反褶积(MED)以及线性规划(LP)和自回归(AR)等方法进行了比较。MED法是通过使相对于地震道反褶积的线性算子系数的熵的范数达到最大来实现的。相比较而言,我们的方法是使相对于反射系数谱中丢失频率的范数值达到最大来实现的。这种处理方法简化了非 In this paper, a method of rebuilding the reflection coefficient spectrum with the minimum entropy criterion is proposed. The authors compared the new method (FMED) with traditional minimum entropy deconvolution (MED) and linear programming (LP) and autoregressive (AR) methods. The MED method is achieved by maximizing the norm of the entropy of the linear operator coefficients relative to the deconvolution of the seismic traces. In contrast, our approach is to maximize the norm relative to the frequency of missing spectra in the reflection spectrum. This approach simplifies the non