LDPC code design for OFDM channel: Graph connectivity and information bits positioning

Alexandre De Baynast, Ashutosh Sabharwal, Behnaam Aazhang

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

9 Scopus citations

Abstract

1 We present an optimized channel coding scheme for OFDM transmission. In the future wireless standards such IEEE 802.11n, frame error rates (FER) as low as 0.0001 are targeting. Irregular LDPC codes have excellent performance at moderate computational complexity: they are strong candidates for channel coding in such systems. In the context of an OFDM transmission, the authors in [Mannoni, 2002] proposed to optimize the graph connectivity of the irregular LDPC code accordingly to the channel spectrum. However, their codes did not have good performance for a FER range of 0.001-0.0001. In this paper, we propose to optimize the LDPC code in two steps: first, we optimize the graph connectivity in order to get a minimum operational average SNR and therefore excellent FER; then we optimize the Information bits placement accordingly to the channel. By simulation, we show that our approach provides substantial performance gain in term of FER over the existing methods (1 dB or more for FER equal to 10 -4).

Original languageEnglish (US)
Title of host publicationISSCS 2005
Subtitle of host publicationInternational Symposium on Signals, Circuits and Systems - Proceedings
Pages649-652
Number of pages4
DOIs
StatePublished - 2005
EventISSCS 2005: International Symposium on Signals, Circuits and Systems - Iasi, Romania
Duration: Jul 14 2005Jul 15 2005

Publication series

NameISSCS 2005: International Symposium on Signals, Circuits and Systems - Proceedings
Volume2

Conference

ConferenceISSCS 2005: International Symposium on Signals, Circuits and Systems
Country/TerritoryRomania
CityIasi
Period7/14/057/15/05

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint

Dive into the research topics of 'LDPC code design for OFDM channel: Graph connectivity and information bits positioning'. Together they form a unique fingerprint.

Cite this