Automatic and optimum sensing of Fresnel reflections in optical fiber systems with connectors using OTDR

This paper discusses an automatic method for setting the optimum OTDR measurement conditions for optical fiber lines with Fresnel reflections. An algorithm is proposed that can find the optimum attenuation of the OTDR detector sensitivity by sensing the peak values of Fresnel reflections on OTDR traces. A neural networks software technology is used in order to select the Fresnel reflection at the cable end with great accuracy. The algorithm is applied to an automatic fiber testing system prototype that is composed of couplers, fiber selectors, an OTDR, a personal computer and fiber lines with connectors and filters. The experimental result shows that the best OTDR measurement conditions can be obtained in a faster process time than the conventional method by the automatic and optimum sensing of Fresnel reflections with the algorithm proposed here.