Figure 1
From: Improved Design of Unequal Error Protection LDPC Codes

Average a posteriori variable node MI as a function of decoder iterations for six different PEG-ACE codes with and . For a low number of iterations, a large gives fast convergence for both
shown. For a larger number of iterations (around 60),
dB gives the highest MI at
dB, and
dB gives the highest MI at
dB.