Year: 2009
Type: conference-paper
Abstract: Independent Component Analysis (ICA) is a statistical and computational technique for\nrevealing hidden factors that underlie set of random variables, measurements, or signals. ICA is a\ngeneral purpose technique which is used to linearly transform the observed random data into\ncomponents. The ICA can be estimated by using the concept of maximum nonGaussianity,\nmaximum\nlikelihood estimation, or minimisation of mutual information. This paper applies ICA to seismic\nacceleration time histories in order to locate any hidden components of ground rotational motion or\ntilts. Normally the three components of seismically induced rotations are not recorded in most of the\navailable seismic instruments, primarily because previous devices did not provide the required\nsensitivity to observe rotations in a wide frequency band and distance range (the two horizontal\ncomponents, equal to tilt at the free surface, are generally recorded at low frequencies) Igel et al 2003.\nFrom the x, y and z components usually recorded the Extended Generalised Lambda Distributions\n(EGLD) – ICA model was used to examine whether rotational or tilt trends were embedded within the\n3 components. The algorithm tries to fit a matrix from the data which will separate any other trends\nwithin the available components. The results show that the EGLDICA\nseparates trends within the 3\ncomponents; however these are not yet identified as tilts or rotations.
Authors: Ramaswany Thiyagu, A.A. Chanerley
Language: English
Subfield: Signal Processing
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W24797666