Fixed, variable and adaptive bit rate data source encoding (compression) method

• US 7,391,918 B2
• Filed: 05/16/2007
• Issued: 06/24/2008
• Est. Priority Date: 08/18/2000
• Status: Expired due to Term

• Current Assignee: Regents of the University of California (University of California)
• Original Assignee: Regents of the University of California (University of California)
• Inventors: Subramaniam, Anand D., Rao, Bhaskar D.
• Primary Examiner(s): Bella; Matthew
• Assistant Examiner(s): Hung; Yubin
• Application Number:
  US11/804,028
• Publication Number:
  US 20070225974A1
• Time in Patent Office: 405 Days
• Field of Search: 382/232, 382/239, 382/251, 382/253, 348/420.1, 348/422.1, 704/222
• US Class Current: 382/253
• CPC Class Codes: G10L 19/008 Multichannel audio signal coding or decoding using interchannel correlation to reduce redundancy, e.g. joint-stereo, intensity-coding or matrixing
• CPC Class Codes: G10L 19/038 Vector quantisation, e.g. TwinVQ audio
• CPC Class Codes: G10L 19/06 Determination or coding of the spectral characteristics, e.g. of the short-term prediction coefficients
• CPC Class Codes: H03M 7/3082 Vector coding for television signals, see H04N19/94

Abstract

According to the invention, quantization encoding is conducted using the probability density function of the source, enabling fixed, variable and adaptive rate encoding. To achieve adaptive encoding, an update is conducted with a new observation of the data source, preferably with each new observation of the data source. The current probability density function of the source is then estimated to produce codepoints to vector quantize the observation of the data source.

Claims

1. A computer readable medium including computer readable instructions for vector encoding a data source comprising:

instructions for taking an observation of a data source;

instructions for estimating a probability density function from the data observation; and

instructions for determining codepoints for vector encoding the data source based upon the estimated probability density function from the data source.

2. The computer readable medium of claim 1, further comprising computer readable instructions for transmitting an index of a determined codepoint to a receiver.

3. The computer readable medium of claim 1, further comprising computer readable instructions for determining with each observation of current data from the data source.

4. The computer readable medium of claim 2, wherein the computer readable instructions for estimating further comprise:

instructions for quantizing an observation of current data from the data source using all data clusters of the current data to obtain m prospective candidates; and

instructions for choosing a codepoint from among the m prospective candidates that minimizes relevant distortion; and

transmitting an index of the codepoint to the receiver.

5. The computer readable medium of claim 1, wherein the computer readable instructions for estimating comprise:

instructions for quantizing an observation of current data from the data source using all data clusters of the current data to obtain m prospective candidates;

instructions for choosing a codepoint from among the m prospective candidates that minimizes relevant distortion;

instructions for identifying the cluster which is used to quantize the observation and using the number of bits allocated to the cluster to obtain the index in the variable-rate case.

6. The computer readable medium of claim 1, further comprising computer readable instructions for decorrelating output of the data source.

7. The computer readable medium of claim 1, further comprising computer readable instructions for transmitting a set of model parameters of the estimated probability density function to a receiver.

8. The computer readable medium of claim 7 further comprising:

instructions for observing new data;

instructions for determining a current estimate of the set of model parameters by applying a re-estimation algorithm to a previous estimate of the set of model parameters; and

instructions for updating the codebook for each of the one or more clusters.

9. The computer readable medium of claim 8, further comprising:

instructions for receiving an update to an index of the codepoint that was determined from a re-estimation of the set of model parameters based on a new data source.

10. A system for vector encoding a data source comprising:

means for taking an observation of a data source;

means for estimating a probability density function from the data observation; and

means for determining codepoints for vector encoding the data source based upon the estimated probability density function from the data source.

11. The system of claim 10, further comprising means for transmitting an index of a determined codepoint to a receiver.

12. The system of claim 11, wherein the means for estimating further comprise:

means for quantizing an observation of current data from the data source using all data clusters of the current data to obtain m prospective candidates; and

means for choosing a codepoint from among the m prospective candidates that minimizes relevant distortion; and

transmitting an index of the codepoint to the receiver.

13. The system of claim 10, wherein the means for estimating comprise:

means for quantizing an observation of current data from the data source using all data clusters of the current data to obtain m prospective candidates;

means for choosing a codepoint from among the m prospective candidates that minimizes relevant distortion;

means for identifying the cluster which is used to quantize the observation and using the number of bits allocated to the cluster to obtain the index in the variable-rate case.

14. A method of vector encoding a data source comprising:

observing a data source and generating a parametric model of the data source, wherein the parametric model is defined by a parameter set and the parameter set is associated with one or more clusters;

generating a codebook for each of the one or more clusters;

quantizing the observation of the data source to obtain one or more codepoints;

selecting a codepoint that minimizes distortion; and

transmitting an index of the codepoint to a receiver.

15. The method of claim 14 further comprising:

observing new data;

determining a current estimate of the parametric model by applying a re-estimation algorithm to a previous estimate of the parametric model; and

updating the codebook for each of the one or more clusters.

16. The method of claim 14, further comprising:

receiving the index for a codepoint by a receiver.

17. The method of claim 16, further comprising:

receiving an update to the index of the codepoint that was determined from a re-estimation of the probability density function based on a new data source.

18. A system for vector encoding a data source comprising:

means for observing a data source and generating a parametric model of the data source, wherein the parametric model is defined by a parameter set and the parameter set is associated with one or more clusters;

means for generating a codebook for each of the one or more clusters;

means for quantizing the observation of the data source to obtain one or more codepoints;

means for selecting a codepoint that minimizes distortion; and

means for transmitting an index of the codepoint to a receiver.

19. The system of claim 18 further comprising:

means for observing new data;

means for determining a current estimate of the parametric model by applying a re-estimation algorithm to a previous estimate of the parametric model; and

means for updating the codebook for each of the one or more clusters.

20. The system of claim 18, further comprising:

means for receiving an update to the index of the codepoint that was determined from a re-estimation of the probability density function based on a new data source.

21. A computer readable medium including computer readable instructions for vector encoding a data source comprising:

instructions for observing a data source and generating a parametric model of the data source, wherein the parametric model is defined by a parameter set and the parameter set is associated with one or more clusters;

instructions for generating a codebook for each of the one or more clusters;

instructions for quantizing the observation of the data source to obtain one or more codepoints;

instructions for selecting a codepoint that minimizes distortion; and

instructions for transmitting an index of the codepoint to a receiver.

22. A system for encoding a non-stationary data source comprising:

means for determining a current estimate of probability density function model parameters of the data source by applying a re-estimation algorithm to a previous estimate of the model parameters and new data; and

means for updating a set of codepoints using the model parameters.

23. The system of claim 22, further comprising means for transmitting the set of model parameters to a receiver.

24. The system of claim 22, further comprising means for decorrelating output of the data source.

25. A computer readable medium including computer readable instructions for encoding a non-stationary data source comprising:

instructions for determining a current estimate of probability density function model parameters of the data source by applying a re-estimation algorithm to a previous estimate of the model parameters and new data; and

instructions for updating a set of codepoints using the model parameters.