Performance evaluation of LDPC codes in bliss scheme-based storage systems using density evolution

Density evolution algorithms for scaled min-sum decoding of both regular and irregular low-density parity-check (LDPC) codes in Bliss scheme-based storage systems are presented, to evaluate the performance of LDPC codes in terms of noise threshold. Firstly, with the assumption of fixed error propagation factor and independently and identically distributed (i.i.d.) messages, density evolution algorithms of regular LDPC codes are derived to compare performance among LDPC codes with the same code rate but different bit/check node degrees and determine the optimum scaling factor of scaled min-sum decoding. Secondly, with the looser assumption of fixed error propagation and independently distributed messages, density evolution algorithms of irregular LDPC codes are derived for performance evaluation of LDPC codes with arbitrary degree distributions. Comparisons with simulations show density evolution is a proper performance evaluation method for LDPC codes in Bliss scheme-based storage systems