Dispersion Compensating Fibers

In this paper we briefly review the work on dispersion compensating fibers (DCFs) in the last few years. Starting with the basic principle behind DCF, its need for upgrading the 1310 nm optical fiber links is discussed. The experimental and theoretical results of some researchers have been cited. The latest trend in optical communication has been to use dense wavelength division multiplexed (DWDM) systems in the gain window of an optical fiber amplifier. However, nonlinear effects like four wave mixing (FWM) impose serious limitations. To overcome this difficulty use of fibers with non-zero dispersion has been suggested. For long distance repeater less transmission this dispersion will go on accumulating and will limit the number of bits one can send at each wavelength. Properly designed dispersion compensating fibers can overcome this difficulty. Already there has been some work in this direction by various researchers. More than 70 million km of conventional single mode fiber (CSF) with zero dispersion wavelength (Xz) ~ 1300 nm had already been laid underground operating around this wavelength. It is known that fiber has lowest loss at 1550 nm and also that efficient amplifiers operate around this wavelength. So it is desirable to operate around 1550 nm. Dispersion shifted fibers (DSF) having zero dispersion wavelength at 1550 nm were developed for this purpose. If we use the existing CSFs at 1550 nm, they will have dispersion coefficient D ~16 ps/knvnm which will limit the bit rate one can transmit. The solution of this problem is through dispersion compensation [1]. In order to understand dispersion compensation let us consider a Gaussian input pulse It can be shown [2] that due to group velocity dispersion in an optical fiber; the pulse at a distance x along the fiber will be given by