Adaptive Neuro-Fuzzy Inference System-based Nonlinear Equalizer for CO-OFDM Systems

Abstract

Abstract The principle of orthogonal frequency-division multiplexing (OFDM) is to transmit the data through a large number of multiple orthogonal subcarriers. The coherent optical OFDM (CO-OFDM) is OFDM data that are being modulated to light frequency and being detected in coherent manner. CO-OFDM brings to optical communications the combination of two powerful techniques, coherent optical detection and OFDM. One of the primary challenges in the CO-OFDM system is to remove optical fiber nonlinear effects. This makes nonlinearity compensation a critical task of the CO-OFDM system. So a nonlinear equalizer (NLE) based on adaptive neuro-fuzzy inference system (ANFIS) is presented for CO-OFDM systems to mitigate nonlinearities on long-haul optical communications with high bit rate and bit error rate (BER)of the system. Various performance metrics were analyzed for the proposed ANFIS–NLE, and it is compared with existing techniques such as support vector machine and artificial neural network. From the experimental results, our proposed approach gives better performance in terms of BER and Q-factor on comparing with existing methods.