Correcting generalized matrix erasures with applications to random network coding

Rank-metric codes are widely used in communications and recently in random network coding. In this paper generalized matrix erasures are defined and an algorithm to correct them is described. However, the main result is that we show that for the common channel model in random network coding only erasures, and no random errors have to be corrected.

[1]  Simon Plass,et al.  Fast decoding of rank-codes with rank errors and column erasures , 2004, International Symposium onInformation Theory, 2004. ISIT 2004. Proceedings..

[2]  Frank R. Kschischang,et al.  Fast encoding and decoding of Gabidulin codes , 2009, 2009 IEEE International Symposium on Information Theory.

[3]  Ernst M. Gabidulin,et al.  Error and erasure correcting algorithms for rank codes , 2008, Des. Codes Cryptogr..

[4]  Ron M. Roth,et al.  Author's Reply to Comments on 'Maximum-rank array codes and their application to crisscross error correction' , 1991, IEEE Trans. Inf. Theory.

[5]  Frank R. Kschischang,et al.  Coding for Errors and Erasures in Random Network Coding , 2007, IEEE Transactions on Information Theory.

[6]  Ernst M. Gabidulin.  A Fast Matrix Decoding Algorithm for Rank-Error-Correcting Codes , 1991, Algebraic Coding.

[7]  Pierre Loidreau,et al.  A Welch-Berlekamp Like Algorithm for Decoding Gabidulin Codes , 2005, WCC.

[8]  Martin Bossert,et al.  Space-time codes based on rank codes , 2000, 2000 IEEE International Symposium on Information Theory (Cat. No.00CH37060).