Ratio configurable multiple description correlating transforms coding

Daichi Saitoh, Takahiro Yakoh

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Multiple description coding (MDC) is one of source coding techniques which generate several descriptions from a single data. A decoder of MDC reconstructs the data from these descriptions. Even if some descriptions are lost, the decoder can reconstruct a similar data from the other descriptions. Therefore, this coding technique is expected to apply applications which prefer accepting error of data to retransmitting the lost or erroneous data over multipath noisy network. Multiple description correlating transforms coding (MDCTC) is a method to realize MDC with a matrix. Based on this method, this paper proposes the method, ratio configurable multiple description correlating transforms coding (RMDCTC), which can adjust the data size ratio among descriptions to fit the ratio of channel bandwidths. The proposed method is suitable to apply error allowable communicating applications over asymmetric multipath noisy network. Mathematical analysis and experimental results verify the effectiveness of the proposed method.

Original languageEnglish
Title of host publicationProceedings - ICIT 2011
Subtitle of host publication2011 IEEE International Conference on Industrial Technology
Pages352-357
Number of pages6
DOIs
Publication statusPublished - 2011 May 18
Event2011 IEEE International Conference on Industrial Technology, ICIT 2011 - Auburn, AL, United States
Duration: 2011 Mar 142011 Mar 16

Publication series

NameProceedings of the IEEE International Conference on Industrial Technology

Other

Other2011 IEEE International Conference on Industrial Technology, ICIT 2011
Country/TerritoryUnited States
CityAuburn, AL
Period11/3/1411/3/16

ASJC Scopus subject areas

  • Computer Science Applications
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Ratio configurable multiple description correlating transforms coding'. Together they form a unique fingerprint.

Cite this