Extending ANN for Optical Elements – EDFA Characteristics
V. S. Lavanya1, V. K. Vaidyan2

1V. S. Lavanya, Department of Physics, Mar Ivanios College, Thiruvananthapuram (Kerala). India.
2V. K. Vaidyan, Department of Physics, Mar Ivanios College, Thiruvananthapuram (Kerala). India.

Manuscript received on 13 June 2016 | Revised Manuscript received on 20 June 2016 | Manuscript Published on 30 June 2016 | PP: 13-17 | Volume-5 Issue-5, June 2016 | Retrieval Number: E4580065516/16©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Artificial Neural Network has proved to be one of the best and widely used soft computation techniques in diversified fields such as Biology, Medicine, Energy, Bioinformatics etc. Modelling in Communication has come far way forward when the industry realized its benefits over conventional method of research and development. It mainly helps in two ways. The first advantage is such that the fabrication cost or wastage is highly reduced, second being the time to final solution implementation. There are various computational methods available in market, which were effectively used in the modelling of different application in diversified fields. In this work, we will discuss how effectively we can use ANN for optical elements and extend it to address the rapid explosion of information traffic and emerging applications in communication. We consider here a basic set up of forward pumped EDFA in a WDM long haul communication system and analyze the characteristics of it through proper signaling. The characterization of the gain, and amplifier noise is again modelled with the help of ANN by appropriately using the experimental data for both modelling and testing. The simulated output from the model agrees well with the experimental data and this approach can be extended to serve as a prediction tool for designing the complex systems in optical communication. The computational time(~ms) taken to model the system and mean-square error(10-5 ) limited is very promising to adapt the model for future activities as desired in further modelling or fabrication of the amplifier with preferred throughput. The results of modeling envisage how favorable ANN is on building the prediction formula in optical communication networks.
Keywords: ANN, EDFA, Modelling, Optical Amplifier

Scope of the Article: Optical Communication