Multiple description coding based on Gaussian mixture models

Author

Samuelsson, Jonas ; Plasberg, Jan H.

Author_Institution

Dept. of Signals, R. Inst. of Technol., Stockholm, Sweden

Volume

12

Issue

6

fYear

2005

fDate

6/1/2005 12:00:00 AM

Firstpage

449

Lastpage

452

Abstract

An algorithm for multiple description coding (MDC) based on Gaussian mixture models (GMMs) is presented. Based on the parameters of the GMM, the algorithm combines MDC scalar quantizers, yielding a source-optimized vector MDC system. The performance is evaluated on a speech spectrum source in terms of mean-squared error and log spectral distortion. It is demonstrated experimentally that the proposed system outperforms single description coding and repetition coding over a wide range of channel failure probabilities. The proposed algorithm has a complexity that is linear in rate and dimension while retaining a near optimal vector quantizer point density.

Keywords

combined source-channel coding; decoding; mean square error methods; probability; telecommunication channels; telecommunication network reliability; vector quantisation; GMM; Gaussian mixture model; MDC; MDC scalar quantizers; channel failure probability; decoding; joint source-channel coding; log spectral distortion; mean-squared error; multiple description coding; speech spectrum source; vector MDC system; Communication networks; Communication systems; Computational complexity; Degradation; Lattices; Quantization; Sensor systems; Speech analysis; Tree data structures; Vectors; Gaussian mixture models (GMMs); joint source-channel coding; multiple description coding (MDC); quantization;

fLanguage

English

Journal_Title

Signal Processing Letters, IEEE

Publisher

ieee

ISSN

1070-9908

Type

jour

DOI

10.1109/LSP.2005.847887

Filename

1430744