A low-complexity MAP detection for pattern-dependent impairments mitigation in Nyquist-WDM coherent optical systems

In the long-haul Nyquist wavelength division multiplexing (Nyquist-WDM) optical fiber communication systems, pattern-dependent impairments, such as the Nyquist filtering and fiber nonlinearity, have adverse effects on system performance. In this paper, we propose a modified maximum-a-posteriori-probability (MAP) detection based on low-complexity Euclidean distance approximate calculation to reduce the implementation complexity of pattern-dependent impairments mitigation. Comprehensive numerical simulation results show its advantages of comparable performance for pattern-dependent impairments mitigation to conventional MAP detection with lower complexity. Furthermore, the three-carrier 20GBaud/subcarrier coherent polarization-multiplexed 16 quadrature-amplitude-modulation Nyquist-WDM (Co PM 16QAM Nyquist-WDM) optical transmission off-line experiment with 7×100km fiber loops is carried out to validate the performance of the modified MAP detection. Compared with the conventional MAP, the modified MAP attain comparable performance with much lower computational complexity.