Data processing: speech signal processing – linguistics – language – Audio signal bandwidth compression or expansion
Reexamination Certificate
1999-12-10
2003-12-16
To, Doris H. (Department: 2655)
Data processing: speech signal processing, linguistics, language
Audio signal bandwidth compression or expansion
Reexamination Certificate
active
06665646
ABSTRACT:
BACKGROUND
Data compression applications are used for various types of data including audio data, video data and executable content. Typically, data compression occurs at an encoder. The compressed data may be delivered to a decoder via a channel and decompressed into a replica (or sometimes an exact duplicate) of the source data.
Various data compression designs are available in the art. They typically reflect two competing interests. On one hand, it is desirable to eliminate all redundancies from a source data signal so that the compressed data signal occupies as little bandwidth as possible when it is placed in a channel. One the other hand, it becomes necessary to ensure that some amount of redundancy remains in the compressed data signal to make it possible to perform data decompression even in the face of data corruption that may be caused by channel imperfections.
Known “layered” coders (or “hierarchical” or “embedded” coders) attempt to harmonize these competing interests. A layered coder generates a compressed data signal that consists of two or more layers of coded data. These layers are often referred to as “streams.” A first “base” layer of coded data represents basic information about the source data signal. If decompression were performed solely upon the base layer, it would be possible to obtain an acceptable and usable representation of the source data signal. Layered coder typically provides additional information in one or more enhancement layers that, when decoded together with the base layer, refine the estimate that is obtained from the base layer. The enhancement layers typically do not completely represent the source data.
In a layered coder, the base layer is critical because a usable replica of the source signal cannot be obtained without it. Such coders are disadvantageous because it may not always be possible to ensure that the base layer is available at the decoder. Accordingly, there is a need in the art for a coding system that generates multiple streams with the property that a usable replica can be decoded from any single stream. Further, there is a need in the art for a coding system that generates such streams with sufficient data compression.
SUMMARY OF THE INVENTION
The present invention provides a balanced multiple descriptive coder, one that codes data in streams so that an acceptable replica of source data can be generated if either stream is lost.
REFERENCES:
patent: 5576765 (1996-11-01), Cheney et al.
patent: 5651090 (1997-07-01), Moriya et al.
patent: 5740283 (1998-04-01), Meeker
patent: 5777679 (1998-07-01), Cheney et al.
patent: 5867602 (1999-02-01), Zandi et al.
patent: 6195465 (2001-02-01), Zandi et al.
patent: 6292589 (2001-09-01), Chow et al.
patent: 6377914 (2002-04-01), Yeldener
Ingle et al, “DPCM System Design for Diversity Systems with Applications to Packetized Speech”,IEEE Transactions on Speech and Audio Processing, vol. 3, No. 1, pp 48-58, publication of the IEEE Signal Processing Society.
Vaishampayan, V.A., “Design of Multiple Description Scalar Quantizers”,IEEE Transactions on Speech and Audio Processing,vol. 39, No. 3, pp 821-835, The Institute of Electrical and Electronics Engineers, Inc.New York, New York.
Vaishampayan et al, “Speech Predictor Design for Diversity Communication Systems”, Proceedings 1995 IEEE Workshop on Speech Coding for Telecommunications,Speech Coding for Interoperable Global Communications,Annapolis, Maryland, Sep. 20-22, 1995.
John Sam
Vaishampayan Vinay A.
AT&T Corp.
Opsasnick Michael N.
To Doris H.
LandOfFree
Predictive balanced multiple description coder for data... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Predictive balanced multiple description coder for data..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Predictive balanced multiple description coder for data... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3153473