Page 710 - Introduction to Information Optics
P. 710

694                     12. Networking with Optics

       redirect light. But in this approach, each mirror has multiple possible posi-
       tions— at least N positions. It is called 2N architect because two arrays of N
       mirrors are used to connect N input to N output fibers. The 3D approach can
       scale to thousands of ports with low loss and high uniformity since the distance
       of light propagation does not increase as the port count grows, whereas in 2D
       switch, port count increase results in squared increase in light travel distance
       and increase in the pitch of the micromirrors and the diameter of the light
       beam, placing tight constraints on collimator performance and mirror align-
       ment tolerance. Such a trade-off can rapidly become unmanageable, leading to
       very large silicon devices and low yields. Thirty-two ports are currently
       considered a top-end size for a single-chip solution in 2D MEMS switch. The
       catch for 3D switch is that a sophisticated analog-driving scheme is needed to
       ensure that the mirrors are in the correct position at all times. Although
       MEMS technology can produce 2N 3D mirror arrays with impressive stability
       and repeatability by using a simple open-loop driving scheme, closing the loop
       with active feedback control is fundamental to achieving the long-term stability
       required in carrier-class deployment of all optical OXCs. Using a closed-loop
       control scheme implies that monitoring the beam positions must be implemen-
       ted in conjunction with computation resources for the active feedback loop and
       very linear high-voltage drivers.




       12.2.7. OPTICAL MONITORING

          Optical monitoring provides the base for optical networking. It is crucial
       not only to offer large capacity for various traffics but also to manage all the
       traffic streams. Network management (NM) functions include performance
       monitoring (PM), alarm, provision, and protection/restoration. NM in electri-
       cal networks is usually done by monitoring some predefined overhead (OH)
       bytes or bits in digital signals. For example, SONET/SDH defines path, line,
       and section overhead bytes. SONET equipment reads these OH bytes for
       network management. In principle, an optical network should perform the
       same functions at the optical layer. Unfortunately, optical monitoring technol-
       ogy is still in its infancy stage, way behind optical transmission technology.
          The earliest (and still widely used) optical monitoring system uses an
       asymmetric optical coupler, called a tap coupler, to take a small portion of the
       traffic-bearing optical signal. After O-to-E conversion, this tapped signal
       provides the optical power level of the mainstream optical signal to NM, based
       on which LOS (loss of signal) alarms and protection switching are determined.
       In some DWDM systems, optical channel or wavelength information is
       encoded using a low-frequency dither tone to modulate the transmitter laser.
       This low-frequency content is extracted at the optical monitoring point, so that
   705   706   707   708   709   710   711   712   713   714   715