Data indexing and compression

• US 8,073,824 B2
• Filed: 07/22/2008
• Issued: 12/06/2011
• Est. Priority Date: 06/04/2007
• Status: Expired due to Fees

• Current Assignee: BAE Systems Plc
• Original Assignee: BAE Systems Plc
• Inventors: Appa, Jamil, Standingford, David Fin William
• Primary Examiner(s): Corrielus, Jean M
• Assistant Examiner(s): Ly, Anh
• Application Number:
  US12/177,586
• Publication Number:
  US 20090012931A1
• Time in Patent Office: 1,232 Days
• Field of Search: 707/600, 707/711, 707/741, 707/957, 707/693, 709/247, 715/242
• US Class Current: 707/693
• CPC Class Codes: G06F 16/283 Multi-dimensional databases or data warehouses, e.g. MOLAP or ROLAP
• CPC Class Codes: H04N 19/537 Motion estimation other than block-based
• CPC Class Codes: H04N 19/61 in combination with predictive coding
• CPC Class Codes: H04N 19/63 using sub-band based transform, e.g. wavelets
• CPC Class Codes: Y10S 707/957 Multidimensional

Abstract

A method of indexing multidimensional data obtains (302) a data set (210) describing features defined over at least one dimension other than time. Index data (212) comprising at least one value corresponding to the at least one non-time dimension is also obtained. The data set is compressed (306) and the index data is associated (308) with the compressed data set such that the compressed data set is identifiable by the index data.

Claims

1. A method of indexing multidimensional data including:

obtaining a data set describing features defined over at least one dimension other than time;

obtaining index data comprising at least one value corresponding to the at least one non-time dimension;

compressing the data set;

storing the compressed data set;

associating the index data with the compressed data set such that the compressed data set is identifiable by the index data;

generating mapping data describing a relationship between a first said compressed data set and a second said compressed data set; and

analysing the mapping data to identify differences between the first and second data sets,

wherein the mapping data is in the form of a motion vector describing a change between the first data set and the second data set in terms of at least one of the dimensions over which the data sets are defined.

2. The method according to claim 1, wherein the data set comprises numerical simulation data, such as computational fluid dynamics (CFD) simulation data representing fluid flow about an object.

3. The method according to claim 2, wherein the index data includes a value for an engineering term such as wingspan, number of wheels, bar thickness, velocity of a moving body in the simulation, and material type that relate to the object, with the engineering term corresponding to the at least one non-time dimension of the data set.

4. The method according to claim 3, wherein the index data includes a value for a field flow term such as stagnation point, vortex, boundary layer characteristic, and shock wave that relate to the fluid flow, with the field flow term corresponding to the at least one non-time dimension of the data set.

5. The method according to claim 4, wherein the method is applied to a plurality of said data sets, thereby resulting in a plurality of compressed data sets, with each of the compressed data sets being associated with respective said index data.

6. The method according to claim 5, wherein the compression of the data set includes steps of:

interpolating the data set onto a background mesh;

decomposing the interpolation into wavelet coefficients using a wavelet function based on the at least one non-time dimension, and

storing the wavelet coefficients.

7. The method according to claim 2, wherein the index data includes a value for a field flow term such as stagnation point, vortex, boundary layer characteristic, and shock wave that relate relating to the fluid flow, with the field flow term corresponding to the at least one non-time dimension of the data set.

8. The method according to claim 1, wherein the method is applied to a plurality of said data sets, thereby resulting in a plurality of compressed data sets, with each of the compressed data sets being associated with respective said index data.

9. The method according to claim 1, wherein the compression of the data set includes steps of:

interpolating the data set onto a background mesh;

decomposing the interpolation into wavelet coefficients using a wavelet function based on the at least one non-time dimension, and

storing the wavelet coefficients.

10. A system configured to index multidimensional data, the system including:

a device configured to obtain a data set describing features defined over at least one dimension other than time;

a device configured to obtain index data comprising at least one value corresponding to the at least one non-time dimension;

a device configured to compress the data set;

a device configured to store the compressed data set,

a device configured to associate the index data with the compressed data set such that the compressed data set is identifiable by the index data;

a device configured to generate mapping data describing a relationship between a first said compressed data set and a second said compressed data set; and

a device configured to analyze the mapping data to identify differences between the first and second data sets,

wherein the mapping data is in the form of a motion vector describing a change between the first data set and the second data set in terms of at least one of the dimensions over which the data sets are defined.

11. A method of indexing multidimensional data comprising:

obtaining a plurality of data sets, each data set describing features defined over at least one dimension other than time;

obtaining index data comprising at least one value corresponding to the at least one non-time dimension of each data set;

compressing each data set by interpolating the data set onto a background mesh, decomposing the interpolation into wavelet coefficients using a wavelet function based on the at least one non-time dimension, and storing the wavelet coefficients;

storing each compressed data set;

associating the index data with a respective compressed data set such that the compressed data set is identifiable by the index data; and

generating mapping data describing a relationship between a first compressed data set and a second compressed data set,

wherein each data set comprises numerical simulation data, such as computational fluid dynamics (CFD) simulation data representing fluid flow about an object,

wherein the index data includes a value for an engineering term relating to the object, with the engineering term corresponding to the at least one non-time dimension of the respective data set, and a value for a field flow term relating to the fluid flow, with the field flow term corresponding to the at least one non-time dimension of the respective data set.

12. A method of indexing multidimensional data including:

obtaining a data set describing features defined over at least one dimension other than time;

obtaining index data comprising at least one value corresponding to the at least one non-time dimension;

compressing the data set;

storing the compressed data set,

associating the index data with the compressed data set such that the compressed data set is identifiable by the index data; and

generating mapping data describing a relationship between a first said compressed data set and a second said compressed data set, wherein the mapping data is in the form of a motion vector describing a change between the first data set and the second data set in terms of at least one of the non-time dimensions over which the data sets are defined and further including analysing the mapping data to identify differences between the first and second data sets, wherein the data set comprises numerical simulation data, such as computational fluid dynamics (CFD) simulation data representing fluid flow about an object, wherein the index data includes a value for an engineering term such as wingspan, number of wheels, bar thickness, velocity of a moving body in the simulation, and material type relating to the object, with the engineering term corresponding to the at least one non-time dimension of the data set.

13. A method of indexing multidimensional data including:

obtaining a data set describing features defined over at least one dimension other than time;

obtaining index data comprising at least one value corresponding to the at least one non-time dimension;

compressing the data set;

storing the compressed data set;

associating the index data with the compressed data set such that the compressed data set is identifiable by the index data; and

generating mapping data describing a relationship between a first said compressed data set and a second said compressed data set, wherein the mapping data is in the form of a motion vector describing a change between the first data set and the second data set in terms of at least one of the non-time dimensions over which the data sets are defined and further including analysing the mapping data to identify differences between the first and second data sets, wherein the data set comprises numerical simulation data, such as computational fluid dynamics (CFD) simulation data representing fluid flow about an object, wherein the index data includes a value for a field flow term such as stagnation point, vortex, boundary layer characteristic, and shock wave that relate to the fluid flow, with the field flow term corresponding to at least one non-time dimension of the data set.