Improvement of RMSE in optical fiber communication systems using RBF ANN post processing approach

In this paper, improvement of Bit Error-Rate (BER) in optical fiber data transmission using an Artificial Neural Networks (ANN) post processing approach is developed. The approach has been tested on optical fiber systems operating at bit rates of up to 10Gbps for a non-return-to-zero (NRZ) modulation formats. The transmission system considered is built using standard optical transmitters and receivers operating in the region of 1550 nm using a standard single mode fiber (SSMF). Performance of signals detected and post processed has been evaluated through Numerical simulations and have shown a very good improvement in the system RMSE after applying the RBF ANN post processing scheme on the detected electrical signal.