Linear-Complexity Overhead-Optimized Random Linear Network Codes (1311.2123v1)

Abstract: Sparse random linear network coding (SRLNC) is an attractive technique proposed in the literature to reduce the decoding complexity of random linear network coding. Recognizing the fact that the existing SRLNC schemes are not efficient in terms of the required reception overhead, we consider the problem of designing overhead-optimized SRLNC schemes. To this end, we introduce a new design of SRLNC scheme that enjoys very small reception overhead while maintaining the main benefit of SRLNC, i.e., its linear encoding/decoding complexity. We also provide a mathematical framework for the asymptotic analysis and design of this class of codes based on density evolution (DE) equations. To the best of our knowledge, this work introduces the first DE analysis in the context of network coding. Our analysis method then enables us to design network codes with reception overheads in the order of a few percent. We also investigate the finite-length performance of the proposed codes and through numerical examples we show that our proposed codes have significantly lower reception overheads compared to all existing linear-complexity random linear network coding schemes.