For long haul coherent optical fiber communication systems, it is significant to precisely monitor the quality of transmission links and optical signals. The channel capacity beyond Shannon limit of Single-mode optical fiber (SMOF) is achieved with the help of Multi-mode optical fiber (MMOF), where the signal is multiplexed in different spatial modes. To increase single-mode transmission capacity and to avoid a foreseen “capacity crunch”, researchers have been motivated to employ MMOF as an alternative. Furthermore, different multiplexing techniques could be applied in MMOF to improve the communication system. One of these techniques is the Optical Code Division Multiple Access (Optical-CDMA), which simplifies and decentralizes network controls to improve spectral efficiency and information security increasing flexibility in bandwidth granularity. This technique also allows synchronous and simultaneous transmission medium to be shared by many users. However, during the propagation of the data over the MMOF based on Optical-CDMA, an inevitable encountered issue is pulse dispersion, nonlinearity and MAI due to mode coupling. Moreover, pulse dispersion, nonlinearity and MAI are significant aspects for the evaluation of the performance of high-speed MMOF communication systems based on Optical-CDMA. This work suggests a hybrid algorithm based on nonlinear algorithm (Dynamic evolving neural fuzzy inference (Dy-NFIS)) and linear algorithm (Recursive least squares (RLS)) equalization for ZCC code in Optical-CDMA over MMOF. Root mean squared error (RMSE), mean squared error (MSE) and Structural Similarity index (SSIM) are used to measure performance results.
