Optical: systems and elements – Deflection using a moving element – Using a periodically moving element
Reexamination Certificate
1999-11-12
2003-03-25
Pascal, Leslie (Department: 2633)
Optical: systems and elements
Deflection using a moving element
Using a periodically moving element
C359S199200
Reexamination Certificate
active
06538780
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally to optical switching networks and more particularly to an apparatus and method for reducing the number of control elements for crosstalk reduction devices in an optical switching network.
BACKGROUND OF THE INVENTION
This invention was made with Government support under Agreement No. MDA972-95-3-0027 awarded by ARPA. The Government has certain rights in the invention.
Optical communication systems use optical signals to convey information over an optical transmission medium, typically a waveguiding medium such as optical fiber. The usable transmission capacity of a given waveguiding medium can be substantially increased by the use of wavelength division multiplexing (WDM) techniques. WDM is a method for increasing the capacity of an optical transmission system by simultaneously operating a plurality of optical signals at different wavelengths over one medium and can be used for both long-haul transmission systems and small local area networks. With WDM, different multiplexed optical signals can be transmitted at different wavelengths, referred to as channel wavelengths, through the same transmission medium.
The extensive use of WDM techniques necessitates the use of speedy interconnected elements such as optical switching modules. The switching modules, typically made using lithium niobate (LiNbO
3
), a ferroelectric material, are necessary for the effective routing and control of optical signals from many different paths. However, as the number of interconnected elements and waveguides increase, crosstalk among signals in the waveguides and optical interconnectors is increasingly a problem.
Generally in an optical network, light beams are modulated in a digital or analog fashion and are used as optical carriers of information. There are many reasons why light beams or optical carriers are preferred in these applications. For example, as the data rate required of such channels increases, the high optical frequencies provide a tremendous improvement in available bandwidth over conventional electrical channels such as formed by wires or coaxial cables. In addition, the energy required to drive and carry high bandwidth signals can be reduced at optical frequencies. Further, optical channels can be packed closely and even intersect in space.
Although optical switches have improved dramatically in the last few years, they do not function as perfect digital switches. That is, there is a certain amount of leakage inherently associated with the waveguide structure of the switch. For example, in a typical 1×2 switch with one input port routed to one of two output ports, there may be some signal leakage of the active signal at the undesired output port which is characterized as crosstalk.
In one example of an optical switching network, arbitrary connections between N input channels to any of M output channels can be accomplished by a tree architecture. Although denoted “channels,” the channels may refer to processors in a multiprocessor environment or fiber optic channels or the like. If desired, any electronic information at the input channels can be modulated on optical carriers and reconverted to electronic information at the output channels to emulate any electronic network. Optionally, modulation in the optical domain can be maintained to provide a ready interface to other optical interconnect schemes.
Referring to
FIG. 1
, there is shown, for illustration purposes, a 12×6 tree architecture
10
coupling twelve input channels
65
with six output channels
75
. A 1×6 fan-out tree
15
is formed for each of the twelve input channels and a 12×1 fan-in tree
20
is formed for each of the six output channels. As can be seen, each stage of the fan-out tree
15
is assembled with active 1×2 switches
35
. Similarly, each stage of the fan-in tree
20
is assembled with 2×1 switches
45
. Each 1×2 switch is composed of one input port and two output ports and each 2×1 switch is composed of two input ports and one output port. The fan-out tree
15
is composed of successive stages of 1×2 switches which act as demultiplexers for each input channel. It should be apparent that a unique path for each of the 12×6 possible connections between the input channels and output channels exists in the network of FIG.
1
. Control signals (not shown) are also coupled to each stage of active switches to control the output of the 1×2 switches and 2×1 switches. Typically, to minimize the real estate taken up by control signals, a “ganged” approach is used to control each stage of switches. That is, all switches in the same stage of a given tree are switched by the same control signal. As will be described in detail below, the intersection of the fan-out tree and the fan-in tree is an advantageous location for the placement of crosstalk reduction devices
55
since the intersection serves as a possible opportunity for any crosstalk signals to combine with active signal paths.
Generally, an N×M network arranged in a tree architecture exhibits [Log
X
N]+[Log
X
M] stages wherein X represents the number of output ports on each switch (2 in our example) within the topology and wherein the expression [Y] represents an integer greater than or equal to the argument Y. For instance, a 12×6 tree architecture comprised of 1×2 and 2×1 switches exhibits [Log
2
12]+[Log
2
6] stages, for a total of 7 stages. Additionally, there are generally N·M interconnection sites between the respective N 1×M fan-out trees and the respective M N×1 fan-in trees for insertion of crosstalk reduction devices. In the 12×6 tree architecture of
FIG. 1
, there are 72 crosstalk reduction devices
55
. Conventionally, for the N·M crosstalk reduction devices, N·M control voltage signals are necessary to individually control the crosstalk reduction devices. In the example of a 12×6 tree architecture, 72 individual control voltage signals are necessary to control the crosstalk reduction devices.
Referring to
FIG. 2
, an illustration of crosstalk propagation that can result from a “ganged” approach to controlling a stage of switches is described using an exemplary embodiment of a 1×32 fan-out tree. The nodes on the figure are schematic representations of 1×2 switches in a given 1×32 fan-out tree
15
for a specific input channel. At each switch, the active signal is routed to the desired output port. However, a “knocked down” version of the active signal is transmitted to the other output port because of leakage and is termed level 1 crosstalk. As this crosstalk propagates to the second stage of switches, the crosstalk is “knocked down” another level. That is, any leakage at a succeeding stage weakens the initial crosstalk signal. Referring again to
FIG. 2
, the double arrowed path shows the desired output of the fan-out tree
15
from one input channel denoting the active signal path. The single arrowed path shows the path of any level 1 crosstalk. At node
101
, the active signal is routed to one of the output ports (arbitrarily shown in the Figure as an upward path and denoted by double arrows). At node
102
, the knocked down crosstalk signal appears as level 1 crosstalk (shown in the Figure as a downward path and denoted by single arrows). However, because of the ganged approach to control signals at each stage, level 1 crosstalk is propagated to node
103
. Similarly, nodes
104
,
105
and
106
experience level 1 crosstalk because of the ganged approach to control signals. In contrast to the propagation of level 1 crosstalk, at the unintended output port at node
102
, the level 1 crosstalk is “knocked down” a level to produce level 2 crosstalk at node
107
. It should be apparent that nodes
111
,
121
,
131
,
141
and
106
will have level 1 crosstalk appearing at the output of the fan-out tree
15
for a particular set of control signals for a particular active signal. Node
Murphy Timothy Owen
Richards Gaylord Warner
Leung Christina Y.
Lucent Technologies - Inc.
Pascal Leslie
LandOfFree
Apparatus and method for reducing the number of control... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Apparatus and method for reducing the number of control..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Apparatus and method for reducing the number of control... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3006190