Optical: systems and elements – Optical frequency converter
Reexamination Certificate
2001-09-14
2003-12-02
Lee, John D. (Department: 2874)
Optical: systems and elements
Optical frequency converter
C385S016000, C398S082000
Reexamination Certificate
active
06657773
ABSTRACT:
CROSS-REFERENCE TO RELATED APPLICATIONS
This application is based on French Patent Application No. 00 11 888 filed Sep. 18, 2000, the disclosure of which is hereby incorporated by reference thereto in its entirety, and the priority of which is hereby claimed under 35 U.S.C. §119.
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates to optical transmission networks and more precisely to switches for wavelength division multiplexed optical signals organized into packets.
2. Description of the Prior Art
Generally speaking, optical packet switching networks include nodes provided with fast packet switches for routing fixed or variable size groups of data, usually referred to as “packets” in the case of the Internet or “cells” in the case of an ATM network.
Photonic switching matrices are “all optical” switches in which data, generally in the form of amplitude modulation of an optical carrier wave, is routed from one optical link to another without changing its optical nature, i.e. without conversion into the electrical domain. One function of photonic switching matrices is to synchronize the packets with a view to managing conflicts in order to minimize the loss of packets. If wavelength division multiplexing (WDM) is used, the matrices must also be designed to take account of the wavelength of the signals to be switched.
The invention relates to a wavelength selector and converter that can be used in matrices for controlling wavelength division multiplexing.
The invention also relates to a photonic switching matrix incorporating the wavelength selector and converter.
FIG. 1
shows an example of an optical switch to which the invention may be applied.
The switch essentially consists of a photonic switching matrix
1
and an associated electronic control unit
2
.
The matrix
1
, which receives a plurality of input WDM optical signals We, We′, is made up of a plurality of modules
3
,
3
′. For simplicity, only two modules and two WDM signals are shown.
The signals We, We′, each consisting of a plurality of channels &lgr;
1
-&lgr;n, are respectively connected on the one hand to the modules
3
,
3
′ via variable delay lines DL, DL′ and on the other hand to optical-electrical converter interfaces OE, OE′ of the control unit
2
via demultiplexers De, De′.
The switching matrix
1
includes, connected in cascade, sets of delay lines
5
each belonging to one of the modules
3
,
3
′, a common “crossbar” space switch
6
, wavelength selector stages
7
, wavelength reallocator stages
8
, and output coupler stages
4
each belonging to one of the modules.
The electronic control unit
2
includes a processor
9
connected on the one hand to the outputs of the optical-electrical converter interfaces OE, OE′ and on the other hand to a control circuit
10
.
A first function of the processor
9
is to decode the headers of the packets received to extract their respective destinations. Depending on the destination information, conditioned by choices imposed by a routing table, the unit
9
then detects any possible conflicts. Accordingly, for each packet received conveyed by each wavelength, the unit
9
determines to which output port of the matrix the packet must be directed and at what time. This routing information is transmitted to the control circuit
10
which then sends appropriate control signals to the space switch
6
and to the wavelength selector stages
7
.
FIG. 2
shows one of the modules
3
of the matrix
1
in more detail. The set
5
of delay lines essentially consists of k delay lines L
2
, Lu, Lk of different lengths and each adapted to create a time-delay that is an integer multiple of the time to transmit a packet. Each delay line receives the input multiplex We associated with the module via a broadcast coupler
11
.
The outputs of the k delay lines are respectively connected to k inputs of the common space switch
6
.
The wavelength selector stage
7
consists of n wavelength selectors SEL
1
, SEL
2
, SELj, SELn respectively controlled by control signals CC
1
, CC
2
, CCj, CCn from the control circuit
10
.
The inputs of the n selectors are respectively connected to n outputs of the space switch
6
supplying the signals S
1
, S
2
, Sj, Sn. The outputs of the selectors SEL
1
, SEL
2
, SELj, SELn are respectively connected to n inputs of a multiplexer
4
via wavelength converters C&lgr;
1
, C&lgr;
2
, C&lgr;j, C&lgr;n constituting the wavelength reallocator stage
8
.
Accordingly, depending on the state of the space switch
6
and the wavelengths selected by the various selectors, each packet belonging to any input multiplex and conveyed by any wavelength can be subjected to a chosen time-delay, routed to any output of the matrix, and conveyed at that output by a new wavelength.
The wavelength of the signals can be taken into account by virtue of the wavelength selector stage
7
and the wavelength reallocator stage
8
. Accordingly, each signal Sj from an output of the switch
6
is processed by a wavelength selector and converter SELj, C&lgr;j of one of the modules
3
before it is injected into the multiplexer
4
of that module.
The selector function implemented by each selector SELj consists of extracting from the WDM signal Sj received from the switch
6
a signal that is conveyed by only one of the wavelengths assigned to the channels of the WDM signal. The function of the wavelength converters C&lgr;
1
-&lgr;n of the same module
3
is to have the signals extracted in this way conveyed by new wavelengths &lgr;
1
-&lgr;n that are fixed and different from each other so that they can be recombined by the multiplexer
4
to constitute an output WDM signal Ws.
FIG. 3
shows a prior art wavelength selector and converter SELj, C&lgr;j associated with one output of the switch
6
.
The selector and converter SELj includes an input broadcast coupler Ce with one input and n outputs respectively connected to n inputs of a multiplexer MX via n optical gates OG
1
, OG
2
, OGx, OGn electrically controlled by the signals CCj. The output of the multiplex MX is connected to the input of the wavelength converter C&lgr;j.
In operation, the input coupler Ce receives the WDM signal Sj and the optical gates receive the control signals CCj such that only one of the gates is open, for example the gate OGx. Accordingly, only the input of the multiplexer MX that is connected to the gate OGx receives the signal Sj and, given the filter function of the multiplexers, only the wavelength that matches that input is transmitted to the output of the multiplexer MX, for example the wavelength &lgr;x. The multiplexer then supplies to the converter C&lgr;j a signal S&lgr;x that is part of the channel carried by the wavelength &lgr;x. Depending on the signal S&lgr;x selectively extracted in this way, the converter C&lgr;j delivers the converted signal S&lgr;j conveyed by the wavelength &lgr;j imposed by that converter.
Accordingly, the converter C&lgr;j must be able to convert any wavelength of the received WDM signal.
Conventional wavelength converters are based on semiconductor optical amplifiers operating under crossed gain modulation conditions or interferometer structures, for example Mach-Zehnder structures, using crossed phase modulation. They have a particular bandwidth, i.e. can convert correctly optical signals whose carrier has a wavelength included in that bandwidth.
In practice the bandwidth of the converters is related to the frequency of the signal modulating the carriers and therefore to the bit rate of the data to be transmitted. To be more precise, for a given type of converter, to increase the bit rate it is necessary to increase the speed of its optical components, and therefore that of the converter, although this reduces the bandwidth. For example, for a bit rate of 10 Gbit/s, a converter can be used that is based on a semiconductor optical amplifier with a low confinement factor and providing a bandwidth in excess of 40 nm. To achieve a bit rate of 40 Gbit/s, it is necessary to increase the confine
Chiaroni Dominique
Le Sauze Nicolas
Mestric Roland
Pons Alain
LandOfFree
Wavelength selector and converter and a photonic switching... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Wavelength selector and converter and a photonic switching..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Wavelength selector and converter and a photonic switching... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3175747