Analysis of Conditional PDF-Based Split VQ

The split vector quantization (SVQ) method results in a ldquocoding lossrdquo due to the independent quantization of split subvectors. To recover the coding loss, conditional pdf-based split vector quantization methods were proposed recently. For a multivariate Gaussian source, we theoretically deri...

Full description

Saved in:
Bibliographic Details
Published in:IEEE signal processing letters 2007-11, Vol.14 (11), p.781-784
Main Authors: Chatterjee, S., Sreenivas, T.V.
Format: Article
Language:English
Subjects:
Allocations
Bit rate
Coding
Conditional pdf
Costs
Decoding
Gain
Gaussian
high rate quantization
Loss measurement
multivariate Gaussian
Optimization
Performance gain
Product codes
Quantization
Rate-distortion
Signal processing
Source coding
Speech coding
split vector quantization
SRA - ICT
SRA - Informations- och kommunikationsteknik
Vector quantization
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The split vector quantization (SVQ) method results in a ldquocoding lossrdquo due to the independent quantization of split subvectors. To recover the coding loss, conditional pdf-based split vector quantization methods were proposed recently. For a multivariate Gaussian source, we theoretically derive the expression of rate-distortion (R/D) performance for conditional pdf-based split vector quantization (CSVQ) method, using high rate quantization theory and optimum bit allocation. We also derive the rate-distortion performance expression of traditional SVQ method and, thus, quantify the coding gain of CSVQ method over SVQ method.
ISSN:1070-9908
1558-2361
1558-2361
DOI:10.1109/LSP.2007.901685