Hybrid random network coding

Tsi-Ui Ik*

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

To avoid possible dependence among coded packets, most research works of random linear network coding suggest encoding packets over a Galois field with enough many elements, e.g., GF (28) or GF (216). However, Galois operations are with higher computational cost than ordinary integer operations. On the other hand, random Boolean network coding, based on the Boolean algebra in which XOR is the only operation needed for encoding and encoding, is a computation-friendly random network coding scheme but at the same time suffered for the possible dependence among coded packets. In this work, a hybrid coding scheme is proposed. Most coded packets are encoded over the Boolean field and only a few augmented coded packets, more precisely 4 packets, are needed to be encoded over the Galois field. Both the computational efficiency and bandwidth effectiveness are taken into consideration. The proposed scheme is theoretically verified by probabilistic analysis.

Original languageEnglish
Title of host publicationWireless Algorithms, Systems, and Applications - 6th International Conference, WASA 2011, Proceedings
Pages202-210
Number of pages9
DOIs
StatePublished - 5 Sep 2011
Event6th International Conference on Wireless Algorithms, Systems, and Applications, WASA 2011 - Chengdu, China
Duration: 11 Aug 201113 Aug 2011

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume6843 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference6th International Conference on Wireless Algorithms, Systems, and Applications, WASA 2011
Country/TerritoryChina
CityChengdu
Period11/08/1113/08/11

Keywords

  • Boolean field
  • Galois fields
  • Hybrid random network coding
  • random Boolean network coding
  • random linear network coding
  • random rank

Fingerprint

Dive into the research topics of 'Hybrid random network coding'. Together they form a unique fingerprint.

Cite this