Achievable information rates calculation for optical OFDM few-mode fiber long-haul transmission systems

We propose a method to estimate the lower bound of achievable information rates (AIRs) of high speed orthogonal frequency-division multiplexing (OFDM) in spatial division multiplexing (SDM) optical long-haul transmission systems. The estimation of AIR is based on the forward recursion of multidimensional super-symbol efficient sliding-window Bahl-Cocke-Jelinek-Raviv (BCJR) algorithm. We consider most of the degradations of fiber links including nonlinear effects in few-mode fiber (FMF). This method does not consider the SDM as a simple multiplexer of independent data streams, but provides a super-symbol version for AIR calculation over spatial channels. This super-symbol version of AIR calculation algorithm, in principle, can be used for arbitrary multiple-input-multiple-output (MIMO)-SDM system with channel memory consideration. We illustrate this method by performing Monte Carlo simulations in a complete FMF model. Both channel model and algorithm for calculation of the AIRs are described in details. We also compare the AIRs results for QPSK/16QAM in both single mode fiber (SMF)- and FMF-based optical OFDM transmission.