Energy minimization for classification, pattern recognition, sensor fusion, data compression, network reconstruction and signal processing

• US 7,272,262 B2
• Filed: 04/01/2005
• Issued: 09/18/2007
• Est. Priority Date: 12/29/1997
• Status: Expired due to Fees

• Current Assignee: Abel Wolman, Jeff B. Glickman
• Original Assignee: Abel Wolman, Jeff B. Glickman
• Inventors: Glickman, Jeff B., Wolman, Abel
• Primary Examiner(s): Desire, Gregory M
• Application Number:
  US11/097,733
• Publication Number:
  US 20050185848A1
• Time in Patent Office: 900 Days
• Field of Search: 382/133, 382/159, 382/170, 382/224, 707/102
• US Class Current: 382/224
• CPC Class Codes: G06F 18/2137 based on criteria of topology preservation, e.g. multidimensional scaling or self-organising maps
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation

Abstract

A data analyzer/classifier comprises using a preprocessing step, and energy minimization step, and a postprocessing step to analyze/classify data.

Claims

1. A signal analysis method comprising the steps of:

receiving input data representative of time varying signals;

wherein the time varying signals are selected from the group consisting of digital signals, analog signals and electrical signals;

mapping the input data into two or more matrices;

applying individual differences multidimensional scaling to the two or more matrices to produce a source space output; and

interpreting the source space output as a spatial representation of the time varying signals.

2. The signal analysis method of claim 1 wherein interpreting the source space output comprises separating elements of the source space output using hyperplanes.

3. The signal analysis method of claim 1 wherein interpreting the source space comprises:

producing a graphical representation of the source space output for analysis of the time varying signals.

4. A signal analysis method comprising the steps of:

receiving input data representative of time varying signals, wherein the time varying signals are selected from the group consisting of digital signals, analog signals and electrical signals;

mapping the input data into one or more matrices;

applying individual differences multidimensional scaling to the one or more matrices to produce a common space output; and

interpreting the common space output as a spatial representation of the time varying signals.

5. The signal analysis method of claim 4 wherein interpreting the common space output comprises separating elements of the common space output using hyperplanes.

6. The signal analysis method of claim 4 wherein interpreting the common space comprises:

producing a graphical representation of the common space output for analysis of the time varying signals.

7. A signal analysis method comprising the steps of:

receiving input data representative of time varying signals, wherein the time varying signals are selected from the group consisting of digital signals, analog signals and electrical signals;

mapping the input data into one or more matrices;

applying individual differences multidimensional scaling to the one or more matrices to produce a rate of change of stress/energy; and

using the rate of change of stress/energy to analyze the time varying signals.

8. The signal analysis method of claim 7 wherein interpreting rate of change of stress/energy comprises:

producing a graphical representation of the rate of change of stress/energy for analysis of the time varying signals.

9. A signal analysis method comprising the steps of:

receiving input data representative of time varying signals wherein the time varying signals are selected from the group consisting of digital signals, analog signals and electrical signals;

mapping the input data into two or more symmetric matrices;

applying individual differences multidimensional scaling to the two or more symmetric matrices to produce a source space output; and

interpreting the source space output as a spatial representation of the time varying signals.

10. The signal analysis method of claim 9 wherein the step of mapping the input data comprises forming two or more hollow symmetric matrices.

11. The signal analysis method of claim 9 wherein the step of mapping the input data into two or more symmetric matrices comprises straight substitution of the received data to populate the two or more symmetric matrices.

12. The signal analysis method of claim 9 wherein the step of mapping the input data into two or more symmetric matrices comprises forming two or more proximity matrices.

13. The signal analysis method of claim 9 wherein the step of mapping the input data into two or more symmetric matrices from the received input data comprises forming two or more distance matrices.

14. The signal analysis method of claim 13 wherein the two or more distance matrices comprise Euclidean distance matrices.

15. The signal analysis method of claim 9 wherein the step of mapping the input data into two or more symmetric matrices comprises populating the two or more symmetric matrices with absolute values of differences of the received data.

16. The signal analysis method of claim 9 wherein the two or more symmetric matrices contain missing values.

17. The signal analysis method of claim 9 wherein the step of applying individual differences multidimensional scaling comprises applying nonmetric individual differences multidimensional scaling.

18. The signal analysis method of claim 9 wherein the step of interpreting the source space output as a spatial representation of the received input data representative of time varying signals further comprises the steps of:

interpreting the spatial representation of the received data representative of time varying signals as containing contextual signatures of the received data; and

using the contextual signatures to analyze the received data.

19. The signal analysis method of claim 9 wherein interpreting the source space comprises:

producing a graphical representation of the source space output for analysis of the time varying signals.

20. A method for signal analysis comprising:

receiving input data representative of time varying signals;

wherein the time varying signals are selected from the group consisting of digital signals and analog signals and electrical signals;

using entry-wise substitution to form two or more hollow symmetric matrices from the received data;

applying individual differences multidimensional scaling to the two or more hollow symmetric matrices to produce at least a source space output; and

interpreting the source space output as a spatial representation of the time varying signals.

21. The signal analysis method of claim 20 wherein interpreting the source space comprises:

producing a graphical representation of the source space output for analysis of the time varying signals.