Prediction of Cross Port Output in Electro-optic Mach-Zehnder interferometer Using Artificial Neural Network

Abstract

Artificial neural networks are essential for the design and analysis of complex algorithms. We have designed a circuit with an electro-optic Mach-Zehnder interferometer (EOMZI) driven by a photodiode. The output from the EOMZI’s cross bar port is used. Then, we have developed a photonic artificial neuron based on this setup. The outputs are validated using MATLAB, and the Artificial neural networks successfully generate the sigmoid function corresponding to the EOMZI switch’s cross port.