Decoding with Early Termination for Raptor Codes

Rateless codes, and especially Raptor codes, have received considerable attention in the recent past due to their inherent ability to adapt to channel conditions and their capacity- approaching performance. Since decoding of rateless codes typically involves multiple decoding attempts, early termination of such attempts is mandatory for overall efficient decoding. In this letter, we propose a new decoding scheme with early termination that is particularly suited for rateless codes. Simulation results for the example of the binary symmetric channel show complexity reductions (in terms of the total required number of decoding iterations) by 87% compared to conventional message-passing decoding and 54% compared to a recently proposed incremental decoding scheme for Raptor codes.

[1]  Payam Pakzad,et al.  Abstract—two Decoding Schedules and the Corresponding Serialized Architectures for Low-density Parity-check (ldpc) , 2001 .

[2]  Michael Luby,et al.  LT codes , 2002, The 43rd Annual IEEE Symposium on Foundations of Computer Science, 2002. Proceedings..

[3]  H. Kfir,et al.  Parallel versus sequential updating for belief propagation decoding , 2002, cond-mat/0207185.

[4]  Yongyi Mao,et al.  Performance-Complexity Tradeoffs of Raptor Codes over Gaussian Channels , 2007, IEEE Communications Letters.

[5]  Richard D. Wesel,et al.  Informed Dynamic Scheduling for Belief-Propagation Decoding of LDPC Codes , 2007, 2007 IEEE International Conference on Communications.

[6]  Yongyi Mao,et al.  Reduced-Complexity Decoding of Raptor Codes over Fading Channels , 2006 .

[7]  Omid Etesami,et al.  Raptor codes on binary memoryless symmetric channels , 2006, IEEE Transactions on Information Theory.