Modulation format identification in heterogeneous fiber-optic networks using artificial neural networks and genetic algorithms

摘要

The use of genetic algorithm (GA) to simplify the structures of artificial neural network-based modulation format identification is proposed in next-generation dynamic and heterogeneous fiber-optic networks. Simulation results show that with 80 asynchronous amplitude histogram bins, by virtue of GA, the identification error rate decreases from 4.24 to 1.04 %.