局域波分解及其在地震信号时频分析中的应用

Local waves decomposition and its applications to seismic signal time-frequency analysis

讨论非平稳、非线性局域波信号分解方法的物理意义和自适应性,数值实现三种局域波分解算法,研究局域波分解在地震信号时频分析中的应用方法和实践.通过Hilbert变换得到地震信号时频分布特征,与Fourier变换谱物理意义不同.采用一定的计算策略,改善瞬时频率计算精度,突出瞬时频率属性的物理意义;将局域波分解同Wigner分布结合,计算地震信号基于局域波分解的Wigner分布,抑制交叉项.模型和实际地震数据试算结果表明实现算法的正确性和有效性.研究表明:在地震信号时频分析中采用局域波分解求Hilbert谱具有自适应性,时域和频域的分辨率也较高;而基于局域波分解的地震信号Wigner分布,保留Wigner分布优良特性,抑制交叉项方法简单易实现;局域波分解方法适合处理地震信号,值得进一步研究和推广.

The local waves decomposition method is suitbale for processing nonstaionanry and nonlinear signals, because the Intrinsic Mode Functions are of determined physical meanings and adaptive. In this papers , the local waves decomposition algorithms and their applications to seismic signals time-frequency analysis are studied. By the Hilbert ttransform , the distributions in the amplititude-time-frequency dimensions can be computed, and the computed instantaneous frequency not only is of determined physical meanings, but also more accurate with some tricks. Combining the local waves decomposition with Wigner distribution , the crossing iterms effects can be reduced very well. The numerical experiments of model and real seismic data show the method is adaptive , the time--frequency resolution is better and the local decomposition Wigner distribution is feasible . On base of the results, it can be concluded that the local wave decomposition is advantagous for seismic signal processing and worth doing more researchs.