Method and system for information signal coding using combinatorial and huffman codes

• US 6,662,154 B2
• Filed: 12/12/2001
• Issued: 12/09/2003
• Est. Priority Date: 12/12/2001
• Status: Active Grant

• Current Assignee: Google Technology Holdings LLC (Alphabet Inc.)
• Original Assignee: Motorola, Inc. (Motorola Solutions, Inc.)
• Inventors: Mittal, Udar, Ashley, James Patrick, Cruz-Zeno, Edgardo Manuel
• Primary Examiner(s): Chawan, Vijay
• Assistant Examiner(s): Nolan, Daniel A.
• Application Number:
  US10/015,871
• Publication Number:
  US 20030110027A1
• Time in Patent Office: 727 Days
• Field of Search: 704/219, 704/211, 704/222, 704/220, 704/503, 386/39, 375/341, 327/18
• US Class Current: 704/219
• CPC Class Codes: G06T 9/005 Statistical coding, e.g. Huffman, run length coding
• CPC Class Codes: G10L 19/10 the excitation function being a multipulse excitation

Abstract

The invention provides a method of coding an information signal. An information signal is represented by a sequence of pulses. A plurality of pulse parameters are determined based on the sequence of pulses including a non-zero pulse parameter corresponding to a number of non-zero pulse positions in the sequence of pulses. The non-zero pulse parameter is coded using a variable-length codeword.

Claims

1. A method of coding an information signal comprising:

representing the information signal by a sequence of pulses;

determining a plurality of pulse parameters based on the sequence of pulses including a non-zero pulse parameter corresponding to a number of non-zero pulse positions in the sequence of pulses;

coding the non-zero pulse parameter using a variable-length instantaneously decodable codeword; and

coding at least one of a remaining plurality of pulse parameters to produce a codeword having a corrsponding length based on the number of non-zero pulse positions.

2. The method of claim 1 wherein the plurality of pulse parameters include a pulse position parameter, a pulse magnitude parameter and a pulse phase parameter.

3. The method of claim 2 wherein the pulse position parameter is coded in accordance with a function:

$F<br><br>\left(n,m\right)=\frac{n<br><br>\text{!}}{\left(n-m\right)<br><br>\text{!}<br><br>m<br><br>\text{!}}$

where n is a is a total number of available positions in the sequence of pulses and m is the number of non-zero pulse positions in the sequence of pulses.

4. The method of claim 3 wherein the function F(n,m) is accessed using a medium memory storage routine (MMSR).

5. The method of claim 3 wherein the function F(n,m) is accessed using a medium memory storage routine (SMSR).

6. The method of claim 2 wherein the pulse magnitude parameter is based on a predetermined sum of pulse magnitudes in the sequence of pulses.

7. The method of claim 2 wherein the pulse phase parameter presents signs of the non-zero pulse positions in the sequence of pulses.

8. The method of claim 2 wherein the pulse magnitude parameter is coded in accordance with a function:

$D<br><br>\left(i,m\right)=\frac{\left(i-1\right)<br><br>\text{!}}{\left(i-m\right)<br><br>\text{!}<br><br>\left(m-1\right)<br><br>\text{!}}$

where i is a sum of pulse magnitudes in the sequence of pulses and m is the number of non-zero pulse positions in the sequence of pulses.

9. The method of claim 8 wherein the function D(i,m) is accessed using a medium memory storage routine.

10. The method of claim 8 wherein the function D(i,m) is accessed using a smallest memory storage routine.

11. The method of claim 1 further comprising:

assembling the coded pulse parameters into a sequence of codewords.

12. The method of claim 11 wherein a first codeword represent the number of non-zero pulse positions in the sequence of pulses.

13. The method of claim 1 wherein the non-zero pulse parameter is coded using a variable-length Huffman code.

14. The method of claim 1 wherein the information signal comprises a time-domain signal.

15. The method of claim 1 wherein the information signal comprises a frequency-domain signal.

16. The method or claim 1 wherein the information signal is selected from a group consisting of a speech signal, a transformed speech signal, an audio signal, a transformed audio signal, an image signal, a transformed image signal, a video signal, and a transformed video signal.

17. A system for coding an information signal comprising:

means for representing the information signal by a sequence of pulses;

means for determining a plurality of pulse parameters based on a sequence of pulses including a non-zero pulse parameter corresponding to a number of non-zero pulse positions in the sequence of pulses;

means for coding the non-zero pulse parameter using a variable-length instantaneously decodable codeword; and

means for coding at least one of a remaining plurality of pulse parameters to produce a codeword having a corresponding length based on the number of non-zero pulse positions.

18. The system of claim 17 further comprising:

means for assembling the coded pulse parameters into a sequence of codewords.

19. A computer usable medium including a program for coding an information signal comprising:

computer program code for representing the information signal by a sequence a of pulses;

computer program code for determining a plurality of pulse parameters based on the sequence of pulses including a non-zero pulse parameter corresponding to a number of non-zero pulse positions in the sequence of pulses;

computer program code for coding the non-zero pulse parameter using a variable-length instantaneously decodable codeword; and

computer program code for coding at least one of a remaining plurality of pulse parameters to produce a codeword having a corresponding length based on the number of non-zero pulse positions.

20. The computer usable medium of claim 19 further comprising:

computer program code for assembling the coded pulse parameters into a sequence of codewords.

21. A method of decoding an information signal comprising:

receiving a plurality of codewords from a source;

determining a first characteristic, representing a number of non-zero pulse positions of a pulse sequence, based on a first variable length codeword;

determining a length of a second variable length codeword, representing a second characteristic of the pulse sequence, based on the first variable length codewords representing the first and second characteristics of the pulse sequence.

22. The method of claim 21 wherein the second characteristic of the pulse sequence is one of the set of characteristics of the pulse sequence to include location of non-zero pulses, pulse magnitudes, and pulse phases.

23. The method of claim 21 wherein the information signal is selected from a group consisting of a speech signal, a transformed speech signal, an audio signal, a transformed audio signal, an image signal, a transformed image signal, a video signal, and a transformed video signal.

24. The method of claim 21 wherein the information signal comprises a time-domain signal.

25. The method of claim 21 wherein the information signal comprises a frequency-domain signal.

26. The method of claim 21 wherein the plurality of codewords is received from a communications channel.

27. The method of claim 21 wherein the plurality of codewords is received from a storage device.

28. A system for decoding an information signal comprising:

means for receiving a plurality of codewords from a source;

means for determining a first characteristic, representing a number of non-zero pulse positions of a pulse sequence, based on a first variable length codeword;

means for determining a length of a second variable length codeword, representing a second characteristic of the pulse sequence, based on the first variable length codeword; and

means for decoding the pulse sequence based on the first and second codewords representing the first and second characteristics of the pulse sequence.

29. A computer usable medium including a program for decoding an information signal comprising:

computer program code for receiving a plurality of codewords from a source;

computer program code for determining a first number of non-zero pulse positions of a pulse sequence, based on a first variable length codeword;

computer program code for determining a length of a second variable length codeword, representing a second characteristic of the pulse sequence, based on the first variable length codeword; and

computer program code for decoding the pulse sequence based on the first and second codewords representing the first and second characteristics of the pulse sequence.