Integer Programming as a Tool for Analysis of Channel Codes

Conference: SCC 2013 - 9th International ITG Conference on Systems, Communication and Coding
01/21/2013 - 01/24/2013 at München, Deutschland

Proceedings: SCC 2013

Pages: 6Language: englishTyp: PDF

Personal VDE Members are entitled to a 10% discount on this title

Authors:
Scholl, Stefan; Kienle, Frank (Microelectronic Systems Design Research Group, University of Kaiserslautern, 67653 Kaiserslautern, Germany)
Helmling, Michael; Ruzika, Stefan (Optimization Research Group, University of Kaiserslautern, 67653 Kaiserslautern, Germany)

Abstract:
Linear and integer programming have recently gained interest as new approach for decoding channel codes. In this paper, we present a framework for the analysis of arbitrary linear block codes based on integer programming.We review how to analyze ML decoding performance and minimum distance. It is shown, that integer programming offers an efficient way for ML decoding. Frame error rates for the ML decoding of various codes with block lengths of several hundred bits have been simulated. Furthermore, we introduce new formulations for weight distributions and reliability based decoding heuristics, like Chase and ordered statistics decoding. New simulation results for WiMAX LDPC and LTE turbo codes under ML decoding are also shown.