基于离散Gabor分布的地震事件识别

Seismic events recognition based on discrete Gabor distribution

描述了一种联合时频分析方法— Gabor分布 ,并且讨论了信号离散 Gabor分布的计算机实现方法。根据公认有效的地震事件识别判据 -体波震级 mb与面波震级 MS之比 mb/ MS,结合信号的 Gabor分布谱图 ,在联合时频域对这一识别判据进行了新的描述 。

A method for signal recognition, applied to seismic event classification, is presented. According to the mb/MS—one effective criterion on seismic event verification, we re explain it by their power in the time frequency plane based on discrete Gabor distribution. The experiment proves that this method, Pb/Ps, is able to classify the nuclear explosion and earthquake evidently. We consider the power ratios of some typical seismic phases useful on monitoring of CTBT(Comprehensive Test Ban Treaty).