PseudoCodeword Analysis of Tanner Graphs from Projective and Euclidean Planes.pdf

Submitted to IEEE Transactions on Information Theory February 25, 2006 Pseudo-Codeword Analysis of Tanner Graphs from Projective and Euclidean Planes∗ Roxana Smarandache† and Pascal O. Vontobel‡ 6 0 0 2 Abstract b In order to understand the performance of a code under maximum-likelihood (ML) e F decoding, one studies the codewords, in particular the minimal codewords, and their Hamming weights. In the context of linear programming (LP) decoding, one’s attention 6 2 needs to be shifted to the pseudo-codewords, in particular to the minimal pseudo- codewords, and their pseudo-weights. In this paper we investigate some families of codes ] that have good properties under LP decoding, namely certain families of low-density T parity-check (LDPC) codes that are derived from projective and Euclidean planes: we I . study the structure of their minimal pseudo-codewords and give lower bounds on their s c pseudo-weight. [ Index terms — Linear programming decoding, message-passing iterative decoding, min- 1 v imal codewords, minimal pseudo-codewords, pseudo-weight, codes from projective planes, 9 codes from Euclidean planes, pseudo-weight spectra. 8 0 2 1 Introduction 0 6 0 The motivation of this paper to look at minimal codewords and minimal pseudo-codewords / s is twofold. On the one hand we would like to give performance gua