U.S. patent application number 09/811136 was filed with the patent office on 2003-02-20 for spectrum division multiplexing for high channel count optical networks.
Invention is credited to Qian, Charles X.W., Qin, Yi, Wang, Hongchuan.
Application Number | 20030035168 09/811136 |
Document ID | / |
Family ID | 24291539 |
Filed Date | 2003-02-20 |
United States Patent
Application |
20030035168 |
Kind Code |
A1 |
Qian, Charles X.W. ; et
al. |
February 20, 2003 |
Spectrum division multiplexing for high channel count optical
networks
Abstract
A versatile, wavelength-slicing methodology, referred to herein
as spectrum division multiplexing (SDM), provides new avenues and
technologies for optical communication applications. Specifically,
SDM separates a composed optical signal into a group of output
spectra. Each output spectrum carries a multiple of optical
communication signal channels. The bandwidth of each channel and
spacing between adjacent channels may differ from one output
spectrum to another. A critical building block of SDM technology is
a spectrum filter with periodic passbands, referred to herein as an
optical spectrum synthesizer (OSS). The cascade of OSS devices, the
combinations of OSS with prior art components and modules, and
other new devices to be used in conjunction with OSS, lead to new
spectrum devices that add new dimensions to existing and new
optical network architectures. The invention of OSS leads to new
Spectrum Division Multiplexing and management devices based on
cascading OSS devices. Examples of these devices include Spectrum
Exchanger, Spectrum Multiplexer, Spectrum Demultiplexer and
Spectrum Add Drop Module. The combinations of OSS and other prior
art devices also lead to several new Spectrum devices and modules.
Examples of these include, Spectrum Switch, Spectrum Cross-Connect
and Spectrum Long Haul Transport Modules. Other devices designed to
be used in conjunction with OSS, e.g., 1/n Multiplexer and 1/n
Demultiplexer, can also be used to form new devices and modules. In
SDM methodology, spectra (group of channels), instead of single
channels, are managed collectively thereby offering both
flexibility and efficiency for next generation high channel count
optical networks.
Inventors: |
Qian, Charles X.W.;
(Cupertino, CA) ; Qin, Yi; (Pleasanton, CA)
; Wang, Hongchuan; (Fremont, CA) |
Correspondence
Address: |
LEONARD TACHNER
A PROFESSIONAL LAW CORPORATION
SUITE 38-E
17961 SKY PARK CIRCLE
IRVINE
CA
92614
US
|
Family ID: |
24291539 |
Appl. No.: |
09/811136 |
Filed: |
March 17, 2001 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
09811136 |
Mar 17, 2001 |
|
|
|
09573330 |
May 18, 2000 |
|
|
|
Current U.S.
Class: |
398/79 |
Current CPC
Class: |
G02B 6/2938 20130101;
G02B 6/29358 20130101; H04J 14/02 20130101; G02B 6/2706 20130101;
G02B 6/12007 20130101 |
Class at
Publication: |
359/124 |
International
Class: |
H04J 014/02 |
Claims
1. A method for combining a plurality of optical signals of
different center frequency into a reduced number of composite
signals; the method comprising the steps of: receiving a plurality
of N parallel input optical signals each defined by a channel
having a selected bandwidth and center frequency; and combining
said input optical signals in at least one wavelength dependent
optical device to form a plurality of M parallel optical output
signals each defined by a spectrum of multiple channels where
N>M.
2. The method recited in claim 1 wherein N.gtoreq.4 and
M.gtoreq.2.
3. A method for segregating a plurality of composite signals each
composed of multiple channels, each channel having a selected
bandwidth and center frequency, into separate individual signals
each having more than one channel; the method comprising the steps
of: receiving said composite signals each have said multiple
channels of selected wavelength and bandwidth; passing said
composite signals into at least one wavelength dependent optical
device to form a plurality of parallel output individual signals
each defined by a group of channels each having selected bandwidth
and a center frequency.
4. An optical device for receiving an input optical composite
signal having a plurality of M channels, each such channel having a
selected bandwidth and center frequency and splitting the input
composite signal into a plurality of output optical signals each
output signal having fewer than M channels; the device comprising:
a broadband beam splitter receiving said input composite signal at
a selected angle of incidence; a pair of optical cavities
positioned on opposite sides of said beam splitter, each such
cavity having a selected optical thickness and a reflective cavity
surface to obtain an output spectrum having selected channel
spacing and wavelength separation.
5. The optical device recited in claim 4 wherein each said cavity
is adjustable in optical thickness.
6. The optical device recited in claim 5 further comprising a
piezoelectric device for adjusting said optical thickness.
7. The optical device recited in claim 4 wherein each said cavity
encloses a controlled gas content.
8. The optical device recited in claim 4 further comprising means
for precisely controlling the temperature of said cavity.
9. An optical device for modifying the spectra of input optical
signals and generating output optical signals having the modified
spectra; the device comprising: a broadband beam splitter receiving
each of said input optical signals at respective selected angles of
incidence; a pair of optical cavities, one such cavity being
positioned on each of opposing sides of said beam splitter; each
said optical cavity having a selected optical thickness and a
mirror of selected reflectivity to produce a selected phase
modification.
10. The optical device recited in claim 9 wherein each said cavity
is adjustable in optical thickness.
11. The optical device recited in claim 9 further comprising a
piezoelectric device for adjusting said optical thickness.
12. The optical device recited in claim 9 wherein each said cavity
encloses a controlled gas content.
13. The optical device recited in claim 9 further comprising means
for precisely controlling the temperature of said cavity.
14. A spectrum exchange apparatus for interchanging selected
spectral components of input composite optical signals, each such
composite signal having a plurality of distinct spectral channels
of selected bandwidth and center frequency; the apparatus
comprising: a spectrum multiplexer receiving said input composite
optical signals and generating a unified composite signal output;
and a spectrum demultiplexer receiving said multiplexer output and
segregating said multiplexer output into a plurality of demux
output signals each having spectral channels from at least two of
said input composite optical components.
15. The spectrum exchange apparatus recited in claim 14 wherein
each of said multiplexer and said demultiplexer comprises at least
one wavelength dependent optical device.
16. A spectral processor for segregating an input composite optical
signal having a plurality of spectral channels of selected
bandwidth and distinct center frequency, into a plurality of
separate output composite optical signals each having selected
spectral components of the input composite optical signal; the
processor comprising: a plurality of cascaded optical spectrum
synthesizers, each said synthesizer having at least one wavelength
dependent optical device for separating a multiple channel optical
signal spectrum into broad and narrow spectral portions, the narrow
spectral portion of each synthesizer forming one of the output
composite optical signals, the broad spectral portion of all but
the last synthesizer forming an input to each subsequent
synthesizer.
17. An optical communications system comprising: a spectral
multiplexer for receiving a plurality of parallel input optical
signals each having at least one frequency channel of selected
bandwidth and center frequency and generating a composite output
signal having all of the frequency channels of the input optical
signals; an add-drop module for removing certain selected frequency
channels of said multiplexer composite output signal and adding
certain selected frequency channels to said multiplexer composite
output signal; and a spectral demultiplexer for generating a
plurality of segrated parallel spectral components from said
add-drop module.
18. A spectrum switch apparatus for use in an optical communication
system; the switch apparatus comprising: a spectral demultiplexer
for receiving a composite optical signal input having a multiple
channel spectrum each channel having a selected bandwidth and
center frequency and generating a plurality of N parallel output
signals each having at least one of said channels; and an N.times.N
switch having N inputs and N outputs and switching means for
placing any of said N inputs on any of said N outputs, said N
inputs corresponding to said N parallel output signals of said
demultiplexer.
19. The spectrum switch of claim 18 wherein N.gtoreq.2.
20. An apparatus for receiving a composite optical signal defined
by a plurality of distinct channels having spaced center
wavelengths in a continuous frequency spectrum; the apparatus
generating two separate output optical signals from the received
signal; the apparatus comprising: a wavelength-dependent optical
device for segregating said received signal into said two separate
output optical signals having non-continuous spectra; one of said
output signals having a greater number of said distinct channels
than the other of said output signals.
21. The apparatus recited in claim 20 wherein the non-continuous
spectrum of one of said output signals is the complement of the
non-continuous spectrum of the other of said output signals.
22. The apparatus recited in claim 20 wherein the combined
non-continuous spectra of said two output signals contain all of
said distinct channels of said continuous frequency spectrum of
said received optical signal.
23. The apparatus recited in claim 20 wherein each of said
non-continuous spectra of said output optical signals comprises a
plurality of passbands that are spaced from one another in
frequency; the number of said distinct channels in each of said
passbands of one of said output signals being greater than the
number of said distinct channels in each of said passbands of the
other of said output signals.
24. The apparatus recited in claim 20 wherein said
wavelength-dependent optical device comprises: a plurality of
optical cavities each having at least one partially reflective
surface.
25. The apparatus recited in claim 20 wherein said
wavelength-dependent optical device comprises: at least two optical
cavities having a total of at least two partially reflective
surfaces; said optical cavities having a selected thickness for
achieving said separate output optical signals.
26. The apparatus recited in claim 25 wherein at least one of said
optical cavities comprises an air spaced optical cavity.
27. The apparatus recited in claim 25 wherein at least one of said
cavities comprises a controlled gas content.
28. The apparatus recited in claim 27 wherein at least one of said
optical cavities comprises a piezoelectric device for selecting
thickness.
29. The apparatus recited in claim 25 wherein each of said
partially reflective surfaces has a reflection coefficient in the
range of 5% to 99.5%.
30. The apparatus recited in claim 25 wherein each of said
partially reflective surfaces has a reflection coefficient in the
range of 18% to 99.5%.
31. The apparatus recited in claim 20 wherein said received
composite signal is incident on said wavelength-dependent optical
device at an angle of less than 10 degrees from normal.
32. The apparatus recited in claim 20 wherein said
wavelength-dependent optical device comprises materials having
selected thermal expansion coefficients to reduce the temperature
sensitivity of said device.
33. The apparatus recited in claim 20 wherein said
wavelength-dependent optical device is positioned in proximity to
temperature control apparatus for selecting temperature adjacent
said device.
34. An apparatus for receiving a composite optical signal defined
by a plurality of distinct channels having center wavelengths in a
continuous frequency spectrum; the apparatus comprising: a
wavelength-dependent optical device for segregating said received
signal into N separate output optical signals having non-continuous
spectra, where N.gtoreq.3; each of said output optical signals
having a substantially equal number of said distinct channels.
35. A spectrum add and drop apparatus for receiving a first
composite optical signal defined by a plurality of distinct
channels having spaced center wavelengths in a continuous frequency
spectrum and generating a second composite optical signal wherein
at least some of said distinct channels from said first composite
signal are replaced by substitute distinct channels in said second
composite signal; the apparatus comprising: a first
wavelength-dependent optical device for segregating said first
composite signal into two separate output optical signals having
non-continuous spectra; one of said output signals having a greater
number of said distinct channels than the other of said output
signals; a second wavelength-dependent optical device connected to
said first wavelength-dependent optical device for receiving said
output signal having a greater number of said distinct channels,
but receiving a substitute for the other output signal of said
first wavelength-dependent optical device; said second
wavelength-dependent optical device generating said second
composite optical signal.
36. The apparatus recited in claim 20 further comprising at least
one wavelength periodic filter connected for filtering of at least
one of said output signals.
37. A method for demultiplexing a composite optical signal with
different center-wavelengths represented by .lambda..sub.1,
.lambda..sub.2, .lambda..sub.3, .lambda..sub.4, . . .
.lambda..sub.n where n is a positive integer and said wavelengths
are equally spaced, comprising steps of: a) receiving said
composite optical signal into an asymmetric wavelength slicing
device through a device input port; and b) slicing said composite
signal and extracting a first composite optical signal comprising a
first set of channels .lambda..sub.1, .lambda..sub.a,
.lambda..sub.b, .lambda..sub.c, . . . .lambda..sub.n-a+2 through a
first output port, and a second composite optical signal comprising
a second set of channels .lambda..sub.2, .lambda..sub.d,
.lambda..sub.e, .lambda..sub.f, . . . .lambda..sub.n through a
second output port wherein said second set of data channels is
complimentary to said first set of data channels and a spacing
(.lambda..sub.1-.lambda..sub.a) between .lambda..sub.1 and
.lambda..sub.a is different from a spacing
(.lambda..sub.2-.lambda..sub.d) between .lambda..sub.2 and
.lambda..sub.d.
38. A method for demultiplexing a composite optical signal with
different center-wavelengths represented by .lambda..sub.1,
.lambda..sub.2, .lambda..sub.3, .lambda..sub.4, . . .
.lambda..sub.n where n is a positive integer and the wavelengths
are equally spaced, comprising steps of: a) receiving said
composite optical signal into an asymmetric wavelength slicing
device through a device input port; and b) slicing said composite
signal and extracting a first composite optical signal comprising a
first set of channels .lambda..sub.1, .lambda..sub.3,
.lambda..sub.5, .lambda..sub.7, . . . .lambda..sub.n-1 through a
first output port, and a second composite optical signal comprising
a second set of channels .lambda..sub.2, .lambda..sub.4,
.lambda..sub.6, .lambda..sub.8, . . . .lambda..sub.n through a
second output port wherein said second set of data channels is
complimentary to said first set of data channels but having a
different bandwidth.
39. A asymmetric wavelength slicing device for demultiplexing a
composite optical signal with different center-wavelengths
represented by .lambda..sub.1, .lambda..sub.2, .lambda..sub.3,
.lambda..sub.4, . . . .lambda..sub.n where n is a positive integer
and the wavelengths are equally spaced, comprising at least an
input port and two output ports, said composite signal being sliced
into a first composite optical signal comprising a first set of
channels .lambda..sub.1, .lambda..sub.a, .lambda..sub.b,
.lambda..sub.c, . . . .lambda..sub.n-a+2 through a first output
port, and a second composite optical signal comprising a second set
of channels .lambda..sub.2, .lambda..sub.d, .lambda..sub.e,
.lambda..sub.f, . . . .lambda..sub.n through a second output port
wherein said second set of data channels is complimentary to said
first set of data channels, but the spacing between .lambda..sub.1
and .lambda..sub.a is different from the spacing between
.lambda..sub.2 and .lambda..sub.d.
40. A asymmetric wavelength slicing device for demultiplexing a
composite optical signal with different center-wavelengths
represented by .lambda..sub.1, .lambda..sub.2, .lambda..sub.3,
.lambda..sub.4, . . . .lambda..sub.n where n is a positive integer
and the wavelengths are equally spaced, comprising: at least an
input port and two output ports, said composite signal being sliced
into a first composite optical signal comprising a first set of
channels .lambda..sub.1, .lambda..sub.3, .lambda..sub.5,
.lambda..sub.7, . . . .lambda..sub.n-1 through a first output port,
and a second composite optical signal comprising a second set of
channels .lambda..sub.2, .lambda..sub.4, .lambda..sub.6,
.lambda..sub.8, . . . .lambda..sub.n through a second output port
wherein said second set of data channels is complimentary to said
first set of data channels, but the bandwidth is different from the
bandwidth of said first set of data channels.
41. The method recited in claim 37 wherein step b) is carried out
by placing a wavelength slicing device in the path of said received
composite optical signal, said device having at least two optical
cavities having a total of at least two partially reflective
surfaces, said optical cavities having a selected thickness for
achieving said first and second composite optical signals.
42. The method recited in claim 38 wherein step b) is carried out
by placing a wavelength slicing device in the path of said received
composite optical signal, said device having at least two optical
cavities having a total of at least two partially reflective
surfaces, said optical cavities having a selected thickness for
achieving said first and second composite optical signals.
43. The device recited in claim 39 further comprising a wavelength
slicing device in the path of said received composite optical
signal, said device having at least two optical cavities having a
total of at least two partially reflective surfaces, said optical
cavities having a selected thickness for achieving said first and
second composite optical signals.
44. The device recited in claim 40 further comprising a wavelength
slicing device in the path of said received composite optical
signal, said device having at least two optical cavities having a
total of at least two partially reflective surfaces, said optical
cavities having a selected thickness for achieving said first and
second composite optical signals.
45. A spectral demultiplexer for use in optical communications
systems: the demultiplexer receiving a composite optical signal
having spectral components in any of a plurality of wavelength
channels in a continuous spectrum and generating a plurality of N
output optical signals each having spectral components in 1/N of
said wavelength channels in respective non-continuous spectra.
46. The spectral demultiplexer recited in claim 45 where
N.gtoreq.3.
47. A spectral multiplexer for use in optical communications
systems; the multiplexer receiving a plurality of N input optical
signals each having different discontinuous spectral components in
1/N wavelength channels of a plurality of wavelength channels in a
continuous spectrum, and generating an output composite optical
signal having the spectral components of all of said N input
optical signals.
48. The spectral multiplexer recited in claim 47 wherein
N.gtoreq.3.
49. A group of optical signal demultiplexers comprising a plurality
of demultiplexers each receiving a different composite optical
signal having a plurality of spaced center channel wavelengths in a
non-continuous spectrum and each such demultiplexer producing a
plurality of individual output optical signals each having a unique
one of said spaced center channel wavelengths.
50. A group of optical signal multiplexers comprising a plurality
of multiplexers each receiving a plurality of individual input
signals each such signal having a center channel wavelength which
is spaced from the center channel wavelength of the other such
signals; each such multiplexer producing a different composite
output signal, each such different output signal comprising all of
the center channel wavelengths of the individual input signals of
the multiplexer from which the output signal is produced.
51. The apparatus recited in claim 34 further comprising an
N.times.N switch for placing said output optical signals on N
output lines in any selected order.
Description
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] This application is a continuation-in-part of pending
applications Ser. No. 09/573,330 filed May 18, 2000.
BACKGROUND OF THE INVENTION
[0002] 1. Field of the Invention
[0003] The present invention relates generally to the field of
optical communications and more particularly to a method and
apparatus for symmetric and asymmetric wavelength/spectrum slicing
for use in dense wavelength division multiplexing (DWDM)
applications.
[0004] 2. Background Art
[0005] Optical communications is an active area of new technology
and is crucial to the development and progress of several important
technologies, e.g., Internet and related new technologies. A key
technology that enabled higher data transmission rate is the dense
wavelength division multiplexing (DWDM) technology. In the DWDM
technology, optical signals generated from different sources
operating at predetermined, densely spaced center wavelengths, are
first combined to form a single optical output. This single optical
output is then transmitted, frequently amplified during
transmission, through an optical fiber. The single optical output
is then de-multiplexed, a process to separate individual data
channels and each channel is then directed to its own destinations.
In the DWDM technology, each data channel is assigned to a center
frequency and the spacing between any two adjacent channels is a
constant (e.g., 200 GHz or 100 GHz, per ITU standard). It is also
understood that all channels are given frequency windows with
identical widths. The width of these windows is kept great enough
to pass information associated with these data channels and at the
same time as narrow as possible to prevent cross-talk between
different data channels. It is generally understood that the
narrower the frequency spacing between different data channels, the
greater the transmission capacity a DWDM system will have at a
given bit rate.
[0006] Several multiplexing and de-multiplexing devices are
essential to the operation of a DWDM system. FIG. 1A is a diagram
illustrating the operation of a group of devices known as optical
filters. An optical filter (100) has the function of separating
signals within a predetermined frequency window (104) from the
input spectrum (102). The remaining signals are output as OUT2
(106). In a DWDM system, to de-multiplex composite data, an optical
filter is employed to separate signals associated with a particular
data channel as depicted in FIG. 1A. Because each channel requires
a specific filter, a DWDM de-multiplexer will require n optical
filters in cascade in order to separate all of n channels into
separate outputs. Using these filter cascades in the reverse
direction will enable the construction of a multiplexer with which
individual signal channels with different center wavelengths, can
be combined together to form a single composite optical output
signal. There are several types of optical filters and brief
descriptions are provided for two types of commonly available
filters. In FIG. 1B, a filter made with optical fiber, known as
fiber Bragg grating (FBG) (110), is illustrated. In a FBG, the
index of refraction of the optical fiber is periodically modified.
The period of the modification, d, is related to the center
wavelength .lambda..sub.m of the given filter as .lambda..sub.m=2 n
d/m. Where m is the order of the Bragg grating and n is average of
the index of refraction of the fiber. Another type of filter
frequently used in DWDM systems is a multi-layer interference
filter (120). These filters are constructed with several, sometimes
many layers of different optical materials with varying thickness
such that a desired transmission (or reflection) curve centered
near a predetermined channel center-frequency is obtained as
depicted in FIG. 1C.
[0007] In the filter approach to DWDM, each data channel is
associated with a specific optical filter. The DWDM system
therefore consists of many filters, each of which has to be
connected or placed in a particular location and/or orientation. A
more systematic way to construct a DWDM system is to use wavelength
dispersion devices such that many channels can be multiplexed or
de-multiplexed with a single device. In FIG. 2A, a device commonly
known as an arrayed waveguide grating (AWG) (200) is displayed. As
depicted, these AWG can be used to separate all data channels
simultaneously. The output channels (204-i) can be connected
directly to individual optical fibers. When using an AWG in the
reverse direction, many different signal channels can be combined
into a single optical fiber. A prism (210) can also be used to
multiplex or de-multiplex optical signals. As illustrated in FIG.
2B, due to dispersion, i.e., the index of refraction is different
for different frequencies so that the exit angle is different for
channels having different center frequencies. Different output
channels (214i) are separated in space and connected into
individual fibers. Another commonly used device is a diffraction
grating (220), an optical surface which is modified periodically
(with a period d) such that when light is directed to this surface,
the angle of incidence (.alpha.) and diffraction (.beta.) are
related to the wavelength of the incoming light, .lambda. according
to: d (sin .alpha.+sin .beta.)=m.lambda., where m is an integer
commonly referred as the order of diffraction. Such a diffraction
grating is illustrated in FIG. 2C.
[0008] A third type of wavelength separating and combing devices
are known as interleavers. FIG. 3A provides a function diagram of
an interleaver (300). These interleavers separate a composite
optical signal (302) into two complementary signals in which the
odd data channels are branched into one output (304) and the even
channels are directed into the other output (306). In an
interleaver application, the frequency space is divided into two
parts, 50% for output 1 and 50% for output 2, as illustrated in
FIG. 3B. Two typical interleaver devices are depicted in FIG. 3C
and FIG. 3D. In FIG. 3C, an interleaver design based upon a
Gires-Toumois (GT) mirror and a Michelson interferometer is
displayed (320). This prior art interleaver was first described by
Dingel and Izutsu in a publication (Optics Letters, Jul. 15, 1998,
vol 23, pages 1099-1101) and is incorporated herein by reference as
relevant background material. In this device, the input signal
(322) is coupled to a 50% beam splitter (321) through a collimating
lens (329). A GT mirror (325) and a regular mirror (327) are used
to form the interferometer. The odd channels return to one output
fiber (324) through a lens (329) whereas the even channels return
to the other fiber (326) through another lens (329). This type of
interleaver and related devices have been disclosed in a recent
U.S. Pat. (No. 6,169,626 issued Jan. 2, 2001). This patent is also
incorporated herein by reference as relevant background material.
In FIG. 3D, another prior art interleaver (330) based on a 50% beam
splitter and a GT mirror is displayed. This prior art device has
been disclosed recently in U.S. Pat. No. 6,169,604 issued on Jan.
2, 2001 to Cao. This patent is therefore incorporated herein by
reference as relevant background material. In this prior art
device, the input signal (332) is coupled to a 50% beam splitter
(331) through a collimating lens (339). Two sections of a phase
modified GT mirror (335) are used as two mirrors of the
interferometer. The odd channels return to one output fiber (334)
through lens (339) whereas the even channels return to the other
fiber (336) through another lens (339).
[0009] These prior art interleavers can provide some flexibility to
DWDM system designers and engineers. In FIG. 4, two stages of
interleavers (400, 410, 420) are cascaded to provide four outputs
(414, 416, 424, 426) each carrying one fourth of the original data
channels. The frequency spacing of the adjacent data channels for a
particular output is therefore four times the spacing between
adjacent data channels in the input signal (402). Another practical
configuration, as demonstrated in FIG. 4B, utilizes both the
interleaver (430) and wavelength dispersion devices (440, 450). In
this configuration, the optical alignment and/or temperature
stability requirements for the dispersion devices are significantly
less stringent when the channel spacing is increased to twice that
of the original spacing. In a different configuration shown in FIG.
5, an interleaver (500), or a two-stage cascade of interleavers, is
followed by individual filters. In this configuration, filters with
a larger channel spacing and hence lower tolerance (e.g., 200 GHz
filters) can be used to construct DWDM systems with a smaller
channel-spacing (e.g., 100 GHz or 50 GHz).
[0010] In many optical network applications, one needs to separate
a group of signal channels and redirect these channels. This is
accomplished via prior art add-drop modules. In FIG. 6, a DWDM long
haul system (600) with multiple add-drop channels is illustrated.
The optical signals of different center wavelengths (602) are
combined through a DWDM multiplexer (603) and amplified via 605. At
a branching point (606), a group of channels is dropped through
add-drop modules, and replaced with signals from alternate sources.
This modified composite signal is transferred to a demultiplexer,
separated into individual channels and sent to their corresponding
receivers (608).
[0011] There are several prior art add/drop module designs. In FIG.
6B, a particular prior art design (610) utilizing interference
filters is illustrated. The incoming signal (612) is directed to
the first interference filter (614) where signals associated with
the channel to be dropped pass through as the drop output (613).
The remaining signal channels reflect from the first filter (614)
to the second filter (615), and are combined with the add input
(616) to form the output (618).
[0012] One of the disadvantages of this prior art add/drop
methodology is that when a group of signal channels is added and
dropped, many filtering components and modules must be used.
Frequently, DWDM Multiplexers and Demultiplexers are also required.
There is therefore a need for a single device that can be used to
accomplish the multichannel add-drop function in a single step.
SUMMARY OF THE INVENTION
[0013] In accordance with the present invention, a new methodology
for optical network data transport and routing is disclosed. In
this Spectrum Division Multiplexing (SDM) method, the dense signal
channels are arranged into groups of channels (spectra) and are
transported accordingly. A critical enabling component of SDM is a
spectrum filter, Optical Spectrum Synthesizer (OSS). OSS separates
a composite, multi-channel optical communication signal into two
groups of channels. Each output signal has a different spectrum
that allows the selection of a different group of channels or the
passage of different frequency regions of the original optical
spectrum. Specifically, each spectrum can be characterized as
comprising periodic pass bands. The width and period of the pass
bands can be designed to accommodate specific network requirements.
The two output spectra are complements of each other, but may have
different pass bandwidths. An OSS can be used to separate two
groups of channels having different OC protocols requiring
different bandwidths, e.g., one output is used to pass OC-192
channels whereas the other is used to pass OC-768 channels. With a
modification to the OSS, a Spectrum Exchanger (SE) is formed. The
SE has the function of exchanging two groups of channels of two
input signals and can be used as an Optical Spectrum Add/Drop
(OSAD) device. An OSAD module provides the network system designer
with a means to add and drop a group of signal channels
collectively. A Spectrum De-Multiplexer (SDEMUX), constructed by
cascading n OSS devices, separates a composite multi-channel
optical signal into n spectra each containing a different subgroup
of the incoming channels. The SDEMUX has a similar functional
structure in comparison with DEMUX devices used in prior DWDM
technology. Instead of having outputs each carrying an individual
signal channel, each output of SDEMUX carries a subgroup of
channels. The individual channels contained in a particular output
of SDEMUX can be further separated using a 1/n DEMUX where the
separation between adjacent channels is n times the spacing of a
prior art DEMUX. Similarly, a Spectrum Multiplexer (SMUX) is
obtained by using the SDEMUX in the reverse direction. A 1/n MUX
can be constructed by using a 1/n DEMUX in the reverse direction.
In an additional embodiment, a long haul transmission system is
disclosed which utilizes SMUX, SDEMUX and EDFA devices. An
alternate long haul system is also disclosed consisting of 1/n MUX,
1/n DE-MUX and EDFA devices. An OSAD Module can also be implemented
with a cascade of two OSS devices. The combination of a SDEMUX with
an optical switch allows the formation of a Spectrum Switch (SS)
where different groups of signal channels can be switched
simultaneously. The SS can be connected to form a Spectrum
Cross-Connect in a way similar to the construction of a
conventional optical cross-connect using conventional optical
switches. Another device comprises two (or more) OSS devices
connected with a branch coupler. Such a device maximizes the usage
of frequency space and hence can be used to achieve a higher
overall data throughput rate in a network system. In still another
embodiment of the invention, a Spectrum Processor is disclosed in
which flexible usage of the frequency space is enabled by dividing
that frequency space to accommodate different OC protocols and
provide a group of channels all within a specific frequency window
and with a different channel spacing and width. The term Nano-T.TM.
as used herein is a trademark used on a product of the assignee of
the present invention. The Nano-T.TM. product is described in
co-pending application Ser. No. ______ filed on Feb. ______, 2001
and that application is hereby incorporated herein in its entirety
by reference.
BRIEF DESCRIPTION OF THE DRAWINGS
[0014] The aforementioned objects and advantages of the present
invention, as well as additional objects and advantages thereof,
will be more fully understood hereinafter as a result of a detailed
description of a preferred embodiment when taken in conjunction
with the following drawings in which:
[0015] FIGS. 1A through 1C (prior art) are simplified diagrams
illustrating conventional filters and their use in DWDM technology.
FIG. 1A is a block diagram illustrating the operation of a generic
filter device. FIG. 1B depicts a fiber Bragg grating filter. FIG.
1C represents a multi-layer interference filter;
[0016] FIGS. 2A through 2C (prior art) are simplified diagrams
illustrating conventional dispersion multi-channel devices and
their use in DWDM technology. FIG. 2A is a diagram illustrating the
operation of an arrayed waveguide grating (AWG) device. FIG. 2B
represents a prism wavelength dispersion device. FIG. 2C shows the
operation of a conventional grating device;
[0017] FIGS. 3A through 3D (prior art) are simplified diagrams
illustrating conventional interleaver devices and their use in DWDM
technology. FIG. 3A is a block diagram illustrating the operation
of an interleaver. FIG. 3B displays the output frequency spectra
associated with two output signals. FIG. 3C shows an interleaver
based on a GT mirror and a regular mirror and FIG. 3D depicts the
operation of an interleaver based upon a GT mirror and a Michelson
interferometer;
[0018] FIGS. 4A through 4B (prior art) are schematic diagrams
illustrating DWDM applications utilizing interleavers. FIG. 4A is a
block diagram of three interleavers in a cascade. The four outputs
each carries 1/4 of the signal channels from the original composed
input signal. FIG. 4B is a schematic diagram illustrating the
combination of an interleaver and two multi-channel dispersion
devices (prisms);
[0019] FIG. 5 (prior art) depicts a device composed of interleaver
and filters. Each output of the device carries only one signal
channel;
[0020] FIGS. 6A through 6B (prior art) are diagrams illustrating a
multichannel add/drop function in an optical network. FIG. 6A
depicts a multichannel add/drop arrangement in a long haul system.
FIG. 6B shows a filter based add-drop module;
[0021] FIGS. 7A through 7C are diagrams illustrating the
methodology of SDM and the operation of a versatile interleaver,
OSS, according to embodiments of the present invention. FIG. 7A is
diagram illustrating a systematic way of grouping signal channels
into spectra. FIG. 7B is a block diagram of an OSS and FIG. 7C
displays the spectra associated with output signals;
[0022] FIGS. 8A through 8C are diagrams illustrating the
construction of OSS based upon a design with two Nano Tuner
(Nano-T) GT mirrors and a 50% beam splitter. FIG. 8A displays a
generic OSS. FIG. 8B depicts a modified OSS device having two
inputs and two outputs. FIG. 8C shows an OSS constructed with a
polygon 50% beam splitter and two Nano-T reflectors;
[0023] FIGS. 9A through 9C are diagrams illustrating the function
of Spectrum Exchanger (SE) in accordance with embodiments of the
present invention wherein FIG. 9A displays the operation of a
Spectrum Exchanger. FIG. 9B illustrate a symbol for this device.
FIG. 9C shows the operation of an Optical Spectrum Add/Drop module
based on the SE;
[0024] FIGS. 10A through 10B are diagrams depicting the separation
of a composite optical signal into two outputs of signals carrying
different protocol channels;
[0025] FIG. 11 is a diagram illustrating a Spectrum DeMultiplexer
(SDEMUX) constructed with three OSS devices;
[0026] FIGS. 12A and 12B are 1/3 DEMUX and 1/3 MUX devices
according to the present invention;
[0027] FIGS. 13A and 13B are diagrams illustrating long haul
systems according to the present invention. In FIG. 13A a system
using 1/n MUX, EDFA and 1/n DEMUX is depicted whereas in FIG. 13B a
system based on SDEMUX, EDFAs and SMUX is shown;
[0028] FIGS. 14A through 14C are diagrams illustrating Spectrum
Add-Drop Module and one application based on the present invention.
In FIG. 14A an SADM constructed using two OSS is displayed whereas
in FIG. 14C a system using SADM is displayed, a symbol for this
device is illustrated in FIG. 14B;
[0029] FIGS. 15A and 15B are diagrams illustrating the construction
of a 1.times.4 Spectrum Switch. In this case, a 4.times.4 switch
follows an SDEMUX to allow flexible redirection of subgroups of
signal channels, FIG. 15B illustrates a symbol for this device;
[0030] FIG. 16 is a diagram which shows the construction of a
4.times.4.times.4 Spectrum Cross-Connect. Eight SS are connected to
form this SCC;
[0031] FIGS. 17A and 17B are diagrams illustrating a module for
which overlapping spectra were generated as the outputs. Device of
this type can be used to maximize the net data throughput rate by
allowing certain amount of crosstalk between adjacent channels;
and
[0032] FIGS. 18A and 18B are diagrams which illustrate a Spectrum
Processor in accordance with an embodiment of the present invention
wherein FIG. 18A illustrates the frequency space usage and FIG. 18B
illustrates the structure of a Spectrum Processor module.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0033] In the following the details of various preferred
embodiments of the present invention are disclosed. The preferred
embodiments are described with the aid of the accompanying
drawings, wherein like reference numerals refer to like elements
throughout.
[0034] FIGS. 7A through 7C are diagrams illustrating the
methodology of spectrum division multiplexing (SDM) and the
operation of a versatile interleaver, referred to herein as an
Optical Spectrum Synthesizer (OSS), according to embodiments of the
present invention. Moreover, hereinafter, the terms "spectrum
filter", "asymmetric interleaver", "1/n interleaver", "spectrum
splitter" are used interchangeably to describe various embodiments
of the present invention.
[0035] In FIG. 7A, a systematical way of organizing a collection of
signal channels into smaller groups is disclosed. In the particular
example illustrated in FIG. 7A, sixteen signal channels are divided
into four groups of channels, referred to herein as four Spectra.
Moreover, hereinafter, the terms "spectrum", and "spectra" are used
interchangeably with "group of signal channels", and "groups of
signal channels" to describe various embodiments of the present
invention.
[0036] In FIG. 7B, an OSS preferably has two outputs. One output
has a group of broader periodic pass bands with a predetermined
bandwidth and period as depicted in FIG. 7C. The other output has a
group of narrower periodic pass bands, which complements that of
output 1. The labels of output 1 and 2 are not critical and the
outputs can also be labeled as N and B for narrow and broad output.
When the bandwidth of the N output is set to be identical to that
of the output B, the device becomes a conventional interleaver as
displayed in FIGS. 3A through 3D.
[0037] Referring now to FIG. 8A, a preferred embodiment of an OSS
(800) comprises a 50% broadband non-polarizing beam splitter (803)
and two Nano-T GT mirrors (805 and 807). The thickness of each
optical cavity of the Nano-T GT mirrors is predetermined to obtain
desired output spectra with proper channel spacing and/or
wavelength separation. The incoming light (802), preferably a
parallel beam with a small angular divergence, is directed to the
beam splitter (803) at a predetermined incident angle with respect
to the surface normal of 803. The branched beam then enters two
Nano-T GT mirrors where upon exiting, the phases of the light beams
are modified. The two light beams are then recombined and
re-branched to form two outputs. These input and output light beams
are interfaced/coupled to optical fibers through lenses. A
preferred type of lens is a graded index lens known as a GRIN lens.
In one preferred embodiment, the reflective surfaces of the Nano-T
GT mirrors have reflectivities of approximately 18%, and 99.5%
respectively.
[0038] In order to match the center frequencies of the pass bands
of output 1 and 2 to that of a standard communication grid (e.g.,
ITU grid), the incident angles and/or environment temperature(s) of
the OSS are adjusted. In addition, one or both of the optical
cavities may be constructed with piezoelectric materials such that
the free-spectra-range of each of the optical cavities may be
controlled. Another preferred way to adjust the free-spectra-range
of the "air-spaced" GT mirror is to set and control the gas mixture
and the pressure of the "air-spaced" cavity. The temperature
environment may also be controlled in a way to enhance the
performance of the OSS. One or more electrical heaters and coolers
are placed close (within a few decimeters) to the two optical
cavities to ensure best performance. The temperature sensitivity of
the etalon can be reduced by using material with low thermal
expansion. Temperature is important because typically a 1 degree C.
change in temperature can have an effect on the critical product of
width and index of refraction comparable to the required precision
to achieve the desired outputs.
[0039] Referring now to FIG. 8B, a preferred embodiment of a
modified OSS (810) comprises a 50% broadband non-polarizing beam
splitter (813), two Nano-T GT mirrors (815 and 817). The thickness
of each optical cavity of the Nano-T GT mirrors is predetermined to
obtain desired output spectra with proper channel spacing and/or
wavelength separation. Two incoming light beams (811, 812),
preferably parallel beams with small angular divergences, are
directed to the beam splitter (813) at predetermined incident
angles with respect to the surface normal. The branched beams then
enter two Nano-T GT mirrors where upon exiting, the phase of the
light beams are modified. The two light beams are then recombined
and re-branched to form two outputs. These input and output light
beams are interfaced/coupled to optical fibers through lenses. A
preferred type of lens is a graded index lens known as a GRIN lens.
In one preferred embodiment, the reflective surfaces of the Nano-T
GT mirrors have reflectivities of approximately 18%, and 99.5%
respectively.
[0040] In order to match the center frequencies of the pass bands
of output 1 and 2 to that of a standard communication grid (e.g.,
ITU grid), the incident angles and/or environment temperature(s) of
the OSS are adjusted. In addition, one or both of the optical
cavities may be constructed with piezoelectric materials such that
the free-spectra-range of each of the optical cavities may be
controlled. Another preferred way to adjust the free-spectra-range
of the "air-spaced" GT mirror is to set and control the gas mixture
and the pressure of the "air-spaced" cavity. The temperature
environment may also be controlled in a way to enhance the
performance of the OSS. One or more electrical heaters and coolers
are placed close (within a few decimeters) to the two optical
cavities to ensure best performance. The temperature sensitivity of
the GT mirrors can be reduced by using material with low thermal
expansion. Temperature is important because typically a 1 degree C.
change in temperature can have an effect on the critical product of
width and index of refraction comparable to the required precision
to achieve the desired outputs.
[0041] Referring now to FIG. 8C, a preferred embodiment of a
modified OSS (820) comprises a 50% broadband non-polarizing beam
splitter (823) and two Nano-T GT mirrors (825 and 827). The
thickness of each optical cavity of the Nano-T GT mirrors is
predetermined to obtain desired output spectra with proper channel
spacing and/or wavelength separation. Two incoming light beams
(821, 822), preferably parallel beams with small angular
divergences, are directed to the beam splitter (823) at
predetermined incident angles with respect to the surface normal.
The branched beams then enter two Nano-T GT mirrors where upon
exiting, the phase of the light beams are modified. The two light
beams are then recombined and re-branched to form two outputs 824
and 826. These input and output light beams are interfaced/coupled
to optical fibers through lenses. A preferred type of lens is a
graded index lens known as a GRIN lens. In one preferred
embodiment, the reflective surfaces of the Nano-T GT mirrors have
reflectivities of approximately 18%, and 99.5% respectively.
[0042] In order to match the center frequencies of the passing
bands of output 1 and 2 to that of a standard communication grid
(e.g., ITU grid), the incident angles and/or environment
temperature(s) of the OSS are adjusted. In addition, one or both of
the optical cavities may be constructed with piezoelectric
materials such that the free-spectra-range of each of the optical
cavities may be controlled. Another preferred way to adjust the
free-spectra-range of the "air-spaced" GT mirror is to set and
control the gas mixture and the pressure of the "air-spaced"
cavity. The temperature environment may also be controlled in a way
to enhance the performance of the OSS. One or more electrical
heaters and coolers are placed close (within a few decimeters) to
the two optical cavities to ensure best performance. The
temperature sensitivity of the GT mirrors can be reduced by using
material with low thermal expansion. Temperature is important
because typically a 1 degree C. change in temperature can have an
effect on the critical product of width and index of refraction
comparable to the required precision to achieve the desired
outputs.
[0043] Referring now to FIG. 9A, the modified OSS disclosed above
may be used to exchange a portion of the signal channels based upon
another preferred embodiment 900 of the present invention. A
periodic passband contained in input signal 1 (901) will be
directed to outputI (904) whereas the complementary periodic
passband will be directed to output 2 (906). For signal channels
contained in input signal 2 (902), the corresponding periodic
passbands will be directed to output 1 and 2, to fill the vacated
regions of the full spectra. The net effect is that the two spectra
contained in inputs 1 and 2 are exchanged at the outputs 1 and
2.
[0044] Referring now to FIG. 9B, the modified OSS disclosed above
may also be used to perform add/drop function in a single step
according to a preferred embodiment 910 of the present invention. A
periodic passband contained in input signal (912) will be dropped
to outputl (914) whereas the complementary periodic passband will
be directed to output (916). The added signal channels are sent
through the spectrum add port (911). These periodic passbands will
be directed to the output (916) fiber.
[0045] In FIG. 10A, an OSS preferably has two outputs. One output
has a group of broader periodic pass bands with a predetermined
bandwidth and period as depicted in FIG. 10B. The other output has
a group of narrower periodic pass bands, which complements that of
output 1. The labels of output 1 and 2 are not critical and the
output can be better labeled as N and B outputs. When the bandwidth
of the N output is set to be identical to that of B, the device
becomes a symmetrical interleaver as displayed in FIGS. 3A through
3D. In a preferred embodiment, a phase correction element and
spectrum filter element may also be introduced to each output to
enhance the OSS performance. In a preferred embodiment of the
present invention, one of the outputs is used to carry channels
with one OC protocol, e.g., OC-192, the other output is used to
carry channels of a different OC protocol to best utilize the
frequency space and maximize the data throughput rate. In a
different embodiment of the present invention, different OC
protocols may be carried in one or both of the outputs.
[0046] FIG. 11A is a diagram illustrating a one to four Spectrum
De-Multiplexer (SDEMUX) constructed with three OSS devices. In a
preferred embodiment of the present invention, three OSS devices
are in a cascade with appropriate spectrum filters and/or phase
correction elements to form a SDEMUX. In a preferred embodiment of
the present invention, the optical spectrum is evenly divided into
four complementary spectra with the same pass channel bandwidths.
In a different preferred embodiment of the present invention, the
optical spectrum is divided into four complementary spectra having
different pass channel bandwidths. In another preferred embodiment
of the present invention, the number of the optical spectra or
output groups, n, is greater than one. When n is equal to two, the
SDEMUX is simply an OSS, whereas when n is equal to four, the
SDEMUX device is as illustrated in FIG. 11A. In additional
embodiments of the present invention, a particular SDEMUX can be
used in the reverse direction as a SMUX. In these cases, n
different and complementary spectra are combined through a SMUX to
form a single composite output signal. FIG. 11B illustrates a
proposed symbol 1100 for a SDEMUX.
[0047] Referring now to FIGS. 12A through 12B, a group of three 1/3
DEMUX and a group of three 1/3 MUX are illustrated, respectively.
According to a preferred embodiment of the present invention, n 1/n
DEMUX devices and n 1/n MUX devices are constructed for a SDEMUX or
SMUX device. Each 1/n DEMUX (and 1/n MUX) carries a subgroup
consisting of 1/n of the total number of channels. In a different
preferred embodiment of the present invention, each 1/n DEMUX (and
1/n MUX) carries a spectrum, which uses a fraction of the whole
frequency space, and in certain cases this fraction may be set to
1/n. The 1/n DEMUX inputs 1201, 1222 and 1242 produce outputs 1204,
1224 and 1244, respectively. The 1/n MUX inputs 1212, 1232 and 1252
produce outputs 1214, 1234 and 1254, respectively.
[0048] FIG. 13A is a diagram illustrating a long haul system
according to a preferred embodiment 1300 of the present invention
wherein grouped input signals 1362, 1372, 1382 and 1392 are
transported to grouped output signals 1364, 1374, 1384 and 1394. In
this case, a long haul system is formed using a SDEMUX 1304, n
optical fibers 1310, 1320, 1330 and 1340, EDFAs (Erbium Doped Fiber
Amplifiers) and a SMUX 1306. Due to a much larger channel spacing
compared with a conventional long haul system using only one
optical fiber or several optical fibers with broadband filters,
nonlinear effects are significantly reduced. A much higher optical
power can therefore be lunched into each of the n fibers thereby
significantly increasing the distances between amplification and/or
recondition stations. FIG. 13B is a diagram illustrating a long
haul system according to a preferred embodiment of the present
invention. In this case, a long haul system is assembled using n
1/n-MUX, n optical fibers 1360, 1370, 1380 and 1390, EDFAs and n
1/n-DEMUX devices. Due to a much larger channel spacing compared
with a conventional system using fewer optical fibers, nonlinear
effects are significantly reduced. A much higher optical power can
therefore be launched into each of the n fibers thereby
significantly increasing the distances between amplification and/or
recondition stations. In a different embodiment of the present
invention, a combination of conventional DWDM devices, SMUX,
SDEMUX, 1/n MUX, 1/n DEMUX and EDFA devices are arranged in a way
to achieve a long haul transport system consisting of more than one
fiber to transport the composed signal spectrum with a larger
channel spacing in each of the fibers.
[0049] Referring now to FIGS. 14A, 14B and 14C, an Optical Spectrum
Add-Drop module (OSAD) 1400 is assembled using two OSS based upon a
preferred embodiment of the present invention. A group of signal
channels can be added and removed simultaneously. This device can
be used to direct network data traffic in a collective way. In
another preferred embodiment of the present invention, the status
of many channels can be monitored using a SADM in a parallel way to
speed up network data management and routing. FIG. 14B depicts a
proposed symbol for this new device and FIG. 14C illustrates a long
haul implementation using the SADM. Input spectrum signal 1402 has
spectrum signal 1406 dropped and spectrum signal 1408 added to
produce output spectrum signal 1404 in FIGS. 14A and 14B.
[0050] In FIG. 14C, input spectrum signals 1453 are combined in
SMUX 1454 with input spectrum signals 1432 from 1/4 MUX 1430. The
dropped spectrum signals 1448 via 1/4 DEMUX 1440 and leave output
spectrum signals 1458 at SDEMUX 1456.
[0051] FIGS. 15A and 15B disclose a preferred construction of a
1.times.4 Spectrum Switch (SS). In this case, a 4.times.4 optical
switch 1520 follows an SDEMUX 1510 that allows flexible redirection
of subgroups of signal channels 1508. In other preferred
embodiments, lxn SS is constructed with the combination of 1 to n
SDEMUX and an nxn optical switch. FIG. 15B illustrates a proposed
symbol for the spectrum switch 1500.
[0052] Referring now to FIG. 16, a 4.times.4.times.4 Spectrum
Cross-Connect (SCC) 1600 is disclosed. The construction of this SCC
has a similar structure in comparison with a conventional optical
cross-connect where different channels in a conventional cross
connect are replaced by subgroups of channels in a SCC. According
to a preferred embodiment of the present invention, eight 1.times.4
SS are connected to form this SCC. A general n.times.n.times.m SCC
uses 2 n 1.times.m SS connected in a way similar to a conventional
n.times.n.times.m optical cross connect. Inputs 1-4 (1601, 1602,
1603 and 1604) are cross connected to become outputs 1-4 (1605,
1606, 1607 and 1608).
[0053] FIGS. 17A and 17B are diagrams illustrating a module 1700
and spectra for which overlapping spectra input 1702 are passed as
the outputs 1714, 1716, 1724 and 1726, according to a preferred
embodiment of the present invention. A wavelength insensitive
branch coupler is used to branch the original composed data into
two or several parts. An OSS is then used to split the composed
signal into two spectra. These spectra are used in a collective way
to process and pass data at a higher throughput rate than
conventional methods by allowing certain degrees of crosstalk
between adjacent channels. The crosstalk between adjacent channels
is then removed through electronic and/or optical decoding of the
original data.
[0054] In another preferred embodiment of the present invention, a
Spectrum Processor 1800 is disclosed where a flexible usage of the
frequency space is enabled. As illustrated in FIG. 18A, the
frequency space is divided to accommodate different OC protocols as
well as to provide a group of channels all within a specific
frequency window and with a different channel spacing and width.
Such a SP module can be made with a combination of OSS 1810, 1820
and 1830 and filters 1814, 1824, 1834 and 1840 as illustrated in
FIG. 18A generating the spectra of FIG. 18B.
[0055] Having thus disclosed various embodiments of the present
invention, it being understood that numerous alternative
embodiments are contemplated and that the scope of the invention is
limited only by the appended claims and their equivalents, what is
claimed is:
* * * * *