Multicriteria Classification for Photonic-Crystal Fiber Design

摘要

The photonic crystal fibers (PCF) are considered as the future information support for the telecommunication system. In this paper, a multicriteria method is used for the design of the PCFs with the user-defined optical proprieties. This method combines the deductive and the inductive learning and it is introduced for the first time in the field of optical fibers. These simulation tools will be optimized for PCF structures in order to optimize the parameters necessary for the improvement of the communication system performances. The multicriteria decision analysis makes it possible to evaluate the optical proprieties of PCFs by determining the effects of attenuation and distortion caused by Physics Phenomena. This decision is done by the means of a relational model preferably. As a result, this method avoids the recourse to distances and makes it possible to use quantitative and/or qualitative criteria. Moreover, it defeat some difficulties encountered when data are expressed in different units. These advantages allow the new multicriteria classification method to be employed easily to the diagnosis and to the design of photonic-crystals fibers.