Soft Q-factor in Optical Communication Systems

High-performance optical communications require a very low bit error rate (BER) down to 10-15 with soft-decision (SD) forward error correction (FEC). There are several performance metrics based on soft information for systems with SD-FEC, but system budgets are usually described with conventional Q-factor converted from pre-FEC BER, based on hard decision for its convenience and history. In this work, for enhancing the feasibility of the soft information, we propose an approximation method for deriving the Q-factor based on soft infor-mation. The proposed method improves the approximation accuracy by more than 20 times compared with the conventional scheme.