Post-FEC BER Prediction for Bit-Interleaved Coded Modulation with Probabilistic Shaping
Accurate performance prediction after forward error correction (FEC) decoding is essential for system design in optical fiber communications. While generalized mutual information (GMI) has been shown to be successful for soft-decision decoded systems, for systems using probabilistic shaping (PS), GMI is less well correlated with the bit-error rate after softdecision decoding. The proposed metrics for such systems are instead normalized GMI (NGMI) and asymmetric information (ASI). They are good to predict BER after FEC decoding or to give an FEC limit in bit-interleaved coded modulation (BICM) with PS, but their relation has not been clearly explained so far. In this paper, we show that NGMI and ASI are equal under matched decoding but not under mismatched decoding. We also examine pre-FEC BER and ASI/NGMI over Gaussian and nonlinear fiber-optic channels with approximately matched decoding. ASI/NGMI always shows better correlation with post-FEC BER than pre-FEC BER for BICM with PS. On the other hand, post-FEC BER can differ at a given ASI/NGMI when we change the bit mapping, which describes how each bit in a codeword is assigned to a bit tributary.
READ FULL TEXT