Decoupled Digital Filtering-enabled Simplification and Fast Convergence of IM/DD System Linearization Algorithms for Strong Band-limited Power Fading Channels

Chromatic dispersion limits practically achievable transmission distances of high data-rate intensity modulation and direct detection (IM/DD) optical transmission systems at C-band. Previously reported IM/DD system linearization algorithms can improve their transmission performances, which are, however, sensitive to the band limitation effect. In this paper, for linearizing IM/DD fiber channels with severe power fading and band-limitation, we propose and evaluate numerically and experimentally a decoupled system linearization scheme called equalization-cascaded multi-constraint iterative algorithm (EC-MCIA) using decoupled, channel partial-aware adaptive digital filters and a multiple constraint iteration algorithm. Compared with the previously reported hybrid MCIA with an embedded filter pair, the EC-MCIA can reduce the complexity and improve the linearization performance. In simulations, we observe two orders of magnitude bit-error-rate (BER) reductions for 100-Gb/s@100-km pulse amplitude modulation (PAM4) transmissions. We also experimentally demonstrate C-band 224 Gbit/s PAM4 transmissions over 25-km standard single mode fibers (SSMF). Experimental results show that the EC-MCIA achieves BERs lower than the 20% forward error correction (FEC) limit and outperforms the hybrid MCIA by nearly an order of magnitude. The required number of taps for the decoupled linear filter is only 15.