Wavelet maxima curves of surface latent heat flux
AssigneeGeorge Mason Intellectual Properties, Inc.
InventorsGuido Cervone, Menas Kafatos, Domenico Napoletani, Ramesh Singh
Life of this patent
Estimating the term…
Abstract
The present invention introduces an innovative data mining technique to identify precursory signals associated with earthquakes. It involves a multistrategy approach that employs one-dimensional wavelet transformations to identify singularities in data, and analyzes the continuity of wavelet maxima in time and space to determine the singularities that could be precursory signals. Surface Latent Heat Flux (SLHF) data may be used. A single prominent SLHF anomaly may be found to be associated some days prior to a main earthquake event.
Claims21
A method for identifying earthquake precursory signals comprising the steps of:
- a. selecting at least one grid to generate a grid path,
- b. analyzing a time-series on said at least one grid using wavelet transformation,
- c. identifying at least one local wavelet maximum for each scale of said wavelet transformation,
- d. generating a sequence of said at least one wavelet maximum,
- e. identifying at least one singularity using said sequence,
- f. measuring geometrical space and time continuity of said at least one singularity along said grid path, and
- g. identifying at least one anomaly.
A program storage device readable by a machine, embodying a program of instructions executable by said machine to perform a method for identifying earthquake precursory signals, said method comprising the steps of:
- a. selecting at least one grid to generate a grid path,
- b. analyzing a time-series on said at least one grid using wavelet transformation,
- c. identifying at least one local wavelet maximum for each scale of said wavelet transformation,
- d. generating a sequence of said at least one wavelet maximum,
- e. identifying at least one singularity using said sequence,
- f. measuring geometrical space and time continuity of said at least one singularity along said grid path, and
- g. identifying at least one anomaly.
An earthquake detector comprising:
- a. a grid selector, capable of selecting at least one grid to generate a grid path,
- b. a wavelet transformation analyzer, capable of analyzing wavelet transformation in a time-series,
- c. a local wavelet maximum identifier, capable of identifying at least one local wavelet maximum for each scale of said wavelet transformation,
- d. a sequence generator, capable of generating a sequence of said at least one local wavelet maximum,
- e. a singularity identifier, capable of identifying at least one singularity using said sequence,
- f. a geometrical analyzer, capable of measuring space and time continuity of said at least one singularity along said grid path, and
- g. an anomaly identifier.
Dependencies are inferred from the claim text.