ON LOW-FREQUENCY ERRORS OF UNIFORMLY MODULATED FILTERED WHITE-NOISE MODELS FOR GROUND MOTIONS.
被引:0
|
作者:
Safak, Erdal
论文数: 0引用数: 0
h-index: 0
机构:
US Geological Survey, Menlo Park,, CA, USA, US Geological Survey, Menlo Park, CA, USAUS Geological Survey, Menlo Park,, CA, USA, US Geological Survey, Menlo Park, CA, USA
Safak, Erdal
[1
]
Boore, David M.
论文数: 0引用数: 0
h-index: 0
机构:
US Geological Survey, Menlo Park,, CA, USA, US Geological Survey, Menlo Park, CA, USAUS Geological Survey, Menlo Park,, CA, USA, US Geological Survey, Menlo Park, CA, USA
Boore, David M.
[1
]
机构:
[1] US Geological Survey, Menlo Park,, CA, USA, US Geological Survey, Menlo Park, CA, USA
EARTHQUAKES - MATHEMATICAL MODELS - NOISE;
SPURIOUS SIGNAL - White Noise - SIGNAL FILTERING AND PREDICTION;
D O I:
暂无
中图分类号:
学科分类号:
摘要:
Low-frequency errors of a commonly used non-stationary stochastic model (uniformly modulated filtered white-noise model) for earthquake ground motions are investigated. It is shown both analytically and by numerical simulation that uniformly modulated filter white-noise-type models systematically overestimate the spectral response for periods longer than the effective duration of the earthquake, because of the built-in low-frequency errors in the model. The errors, which are significant for low-magnitude short-duration earthquakes, can be eliminated by using the filtered shot-noise-type models (i. e. white noise, modulated by the envelope first, and then filtered).