LTS/MAX/UMBC PMD Project

Stochastic Optical Impairments and Network Layer Mitigation Techniques in High Speed Optical Networks

This project is a collaboration between MAX and the University of Maryland Baltimore County (UMBC), with support from the Laboratory for Telecommunication Sciences (LTS). The goal is to provide a proof-of-concept experimental demonstration that we can monitor different impairments in the physical layer of an optical network, and use this information as a controlling element in a DRAGON optical network, causing dynamic optical link management based on time-varying changes in the physical layer condition. We will focus on polarization mode dispersion (PMD), a well-known impairment which is already problematic for installed 10 Gb/s systems and will become even more serious as data rates increase to 40 Gb/s and beyond [1]. The PMD on a link can vary significantly over time, and is sensitive to the physical environment of the fiber - temperature, stress, etc. While significant effort has been put into developing PMD compensating devices, such solutions have several drawbacks. For example, it will typically be quite costly, requiring one device for each wavelength channel on a given link. Further, such devices are inefficient in the sense that for a reasonably good link, the compensator will not be needed most of the time.

In this project, we will demonstrate an alternative to PMD compensation, an approach which has been dubbed "soft failure" [2]. Using a simple sensor scheme, we will monitor the PMD level on a given fiber link in real time, and migrate that information about the physical layer up to the control plane software. When the fiber is determined to be unsuitable for 10 Gb/s transmission, the control plane will route new traffic to alternative paths, such as a different fiber, or a lower data rate channel. If the threshold for switching is done carefully, this can be done in such a way as to prevent the loss of critical data travelling over the link. Such an approach should be much more cost effective than compensators, and more easily applied to multi-channel systems, or networks which have multiple fiber paths that must be monitored.

References

[1] For a review of PMD and related issues, see for example, H. Kogelnik, R. M. Jopsen, and L. E. Nelson, "Polarization-Mode Dispersion," in Optical Fiber Communications, vol. IVb, I. Kaminow and T. Li, Ed., p. 725 (Academic Press, San Diego, CA, 2002).

[2] J. Zweck and C. R. Menyuk, "Detection and Mitigation of Soft Failure due to Polarization-Mode Dispersion in Optical Networks," In Proc. Opt. Fiber Commun. Conf. (OFC2006), Anaheim, CA, Paper OFG5.