Method and system for wavelet packet transmission using a best base algorithm

• US 7,436,884 B2
• Filed: 03/25/2003
• Issued: 10/14/2008
• Est. Priority Date: 03/26/2002
• Status: Expired due to Fees

• Current Assignee: Lockheed Martin Corporation (Martin Marietta Corporation)
• Original Assignee: Lockheed Martin Corporation (Martin Marietta Corporation)
• Inventors: Olson, Teresa L., Chen, Hal-Wen
• Primary Examiner(s): WANG, TED M
• Application Number:
  US10/395,255
• Publication Number:
  US 20030185247A1
• Time in Patent Office: 2,030 Days
• Field of Search: 375/240, 375/240.25, 382/252, 382/253
• US Class Current: 375/240
• CPC Class Codes: G06F 18/25 Fusion techniques
• CPC Class Codes: G06F 18/251 of input or preprocessed data
• CPC Class Codes: G06F 18/256 of results relating to different input data, e.g. multimodal recognition
• CPC Class Codes: G06T 2207/10048 Infrared image
• CPC Class Codes: G06T 5/40 using histogram techniques
• CPC Class Codes: G06T 5/92 based on global image properties

Abstract

A data compression system and method provides a best base searching method to determine the best orthogonal basis function for wavelet-based, data compression. A processing device may receive data signals from a source and wavelet-based data compression may be performed on the received data before transmission (to a receiving component) and/or further digital signal processing (DSP) of the data signals. An encoder portion of the processing device performs the data compression using a predetermined algorithm that determines the best orthogonal basis function (base) of the signal transform (used for representing the data signals) by searching a set of bases including (approximately) all orthogonal bases available to provide the minimum number of coefficients for efficient data compression while maintaining a low error rate in reconstructing the original data signals.

Claims

1. A method for performing data compression, comprising:

a) receiving a digital data signal;

b) determining a true best basis (TBB) function to assign a best basis function for encoding the digital data signal;

c) transforming digital data signal based on a wavelet-based transform coefficient at various decomposition levels to produce wavelet coefficients;

d) determining the best basis function based on the wavelet coefficients; and

e) encoding at least a portion of said received digital data signal using a signal transform said encoding being based on the best basis function satisfying a predetermined threshold.

2. The method of claim 1, wherein said determining the best basis function based on the wavelet coefficients further comprises performing the following steps:

d1) sorting all wavelet coefficients from each decomposition level into a one dimensional vector, V_s(n) with a length of N*log₂(N) where N is a sample length of the digital data signal;

d2) selecting first sample coefficient V_s(n) as a best vector of the best basis function where n is equal to 1;

d3) generating a look-up-table (LUT) containing a non-orthogonal tree for a first best vector V_s(l);

d4) selecting V_s(n+1) as a succeeding sample coefficient; and

d5) determining the best basis function according to an orthogonality of the succeeding sample coefficient.

3. The method of claim 2, wherein said determining the best basis function according to the orthogonality of the succeeding sample coefficient further comprises performing the following steps:

if the succeeding sample coefficient V_s(n+1) is orthogonal to V_s(n), said determining step iteratively performing the following steps;

d5-1) selecting the sample coefficient V_s(n+1) as a succeeding best vector of the best basis function and storing to the LUT, the non-orthogonal tree for updated best vector;

d5-2) if n<

N*log₂(N), iterate steps (g)-(h) and n increment by 1 for each iteration; and

d5-3) if n is not less than N*log₂(N), the updated sample succeeding sample coefficient V_s(n+1) is determined as the best basis function.

4. The method of claim 3, wherein said determining the best basis function according to the orthogonality of the succeeding sample coefficient further comprises performing the following steps:

if the succeeding sample coefficient V_s(n+1) is not orthogonal to V_s(n), said determining step iteratively performing the following steps;

d5-4) if n<

N*log₂(N), repeat steps (g)-(h) with n increment by 1 for each iteration; and

d5-5) if n is not less than N*log₂(N), the current sample succeeding sample coefficient V_s(n+1) is determined as the best basis function.

5. The method of claim 1, wherein said predetermined threshold relates to one of number of errors in, number of coefficients for, amplitude of coefficients for, and computational complexity for reconstructing said at least a portion of the received digital data signal.

6. The method of claim 1, wherein said predetermined threshold relating to a number of coefficients is 50 or less.

7. The method of claim 1, wherein said encoding includes encoding at least a portion of said received data signal using a digital signal transform having a basis function being selected from one of a child and parent node on decomposition levels of said transform.

8. A system for performing data compression, comprising:

a plurality of sensors for acquiring data; and

a controller for performing the steps of;

a) receiving a digital data signal from said plurality of sensors;

b) determining a true best basis (TBB) function to assign a best basis function for encoding the digital data signal;

c) transforming digital data signal based on a wavelet-based transform coefficient at various decomposition levels to produce wavelet coefficients;

d) determining the best basis function based on the wavelet coefficients; and

e) encoding at least a portion of said received digital data signal using a signal transform said encoding being based on the best basis function satisfying a predetermined threshold.

9. The system of claim 8, wherein said determining the best basis function based on the wavelet coefficients further comprises performing the following steps:

d1) sorting all wavelet coefficients from each decomposition level into a one dimensional vector, V_s(n) with a length of N*log₂(N) where N is a sample length of the digital data signal;

d2) selecting first sample coefficient V_s(n) as a best vector of the best basis function where ii is equal to 1;

d3) generating a look-up-table (LUT) containing a non-orthogonal tree for a first best vector V_s(1);

d4) selecting V_s(n+1) as a succeeding sample coefficient; and

d5) determining the best basis function according to an orthogonality of the succeeding sample coefficient.

10. The system of claim 9, wherein said determining the best basis function according to the orthogonality of the succeeding sample coefficient further comprises performing the following steps:

if the succeeding sample coefficient V_s(n+1) is orthogonal to V_s(n), said determining step iteratively performing the following steps;

d5-1) selecting the sample coefficient V_s(n+1) as a succeeding best vector of the best basis function and storing to the LUT, the non-orthogonal tree for updated best vector;

d5-2) if n<

N*log₂(N), iterate steps (g)-(h) and n increment by 1 for each iteration; and

d5-3) if n is not less than N*log₂(N), the updated sample succeeding sample coefficient V_s(n+1) is determined as the best basis function.

11. The system of claim 10, wherein said determining the best basis function according to the orthogonality of the succeeding sample coefficient further comprises performing the following steps:

if the succeeding sample coefficient V_s(n+1) is not orthogonal to V_s(n), said determining step iteratively performing the following steps;

d5-4) if n<

N*log₂(N), repeat steps (g)-(h) with n increment by 1 for each iteration; and

d5-5) if n is not less than N*log₂(N), the current sample succeeding sample coefficient V_s(n+1) is determined as the best basis function.

12. The system of claim 8, wherein said predetermined threshold relates to one of number of errors in, number of coefficients for, amplitude of coefficients for, and computational complexity for reconstructing said at least a portion of the received digital data signal.

13. The system of claim 8, wherein said predetermined threshold relating to a number of coefficients is 50 or less.

14. The system of claim 8, wherein said encoding includes encoding at least a portion of said received data signal using a digital signal transform having a basis function being selected from one of a child and parent node on decomposition levels of said transform.

15. A computer-readable storage medium executable by a computer encoded with a computer program stored therein, the computer program when executed by a computer performing data compression, comprising:

a) receiving a digital data signal;

b) determining a true best basis (TBB) function to assign a best basis function for encoding the digital data signal;

c) transforming digital data signal based on a wavelet-based transform coefficient at various decomposition levels to produce wavelet coefficients;

d) determining the best basis function based on the wavelet coefficients; and

e) encoding at least a portion of said received digital data signal using a signal transform said encoding being based on the best basis function satisfying a predetermined threshold.

16. The computer-readable storage medium of claim 15, wherein said determining the best basis function based on the wavelet coefficients further comprises performing the following steps:

d1) sorting all wavelet coefficients from each decomposition level into a one dimensional vector, V_s(n) with a length of N*log₂(N) where N is a sample length of the digital data signal;

d2) selecting first sample coefficient V_s(n) as a best vector of the best basis function where n is equal to 1;

d3) generating a look-up-table (LUT) containing a non-orthogonal tree for a first best vector V_s(1);

d4) selecting V_s(n+1) as a succeeding sample coefficient; and

d5) determining the best basis function according to an orthogonality of the succeeding sample coefficient.

17. The computer-readable storage medium of claim 16, wherein said determining the best basis function according to the orthogonality of the succeeding sample coefficient further comprises performing the following steps:

if the succeeding sample coefficient V_s(n+1) is orthogonal to V_s(n), said determining step iteratively performing the following steps;

d5-1) selecting the sample coefficient V_s(n+1) as a succeeding best vector of the best basis function and storing to the LUT, the non-orthogonal tree for updated best vector;

d5-2) if n<

N*log₂(N), iterate steps (g)-(h) and n increment by 1 for each iteration; and

d5-3) if n is not less than N*log₂(N), the updated sample succeeding sample coefficient V_s(n+1) is determined as the best basis function.

18. The computer-readable storage medium of claim 17, wherein said determining the best basis function according to the orthogonality of the succeeding sample coefficient further comprises performing the following steps:

if the succeeding sample coefficient V_s(n+1) is not orthogonal to V_s(n), said determining step iteratively performing the following steps;

d5-4) if n<

N*log₂(N), repeat steps (g)-(h) with n increment by 1 for each iteration; and

d5-5) if n is not less than N*log₂(N), the current sample succeeding sample coefficient V_s(n+1) is determined as the best basis function.

19. The computer-readable storage medium of claim 15, wherein said predetermined threshold relates to one of number of errors in, number of coefficients for, amplitude of coefficients for, and computational complexity for reconstructing said at least a portion of the received digital data signal.

20. The computer-readable storage medium of claim 15, wherein said predetermined threshold relating to a number of coefficients is 50 or less.

21. The computer-readable storage medium of claim 15, wherein said encoding includes encoding at least a portion of said received data signal using a digital signal transform having a basis function being selected from one of a child and parent node on decomposition levels of said transform.