U.S. patent application number 15/363385 was filed with the patent office on 2018-05-31 for systems and methods for a preemptive controller for photonic line systems.
The applicant listed for this patent is Ciena Corporation. Invention is credited to Choudhury A. AL SAYEED, Dave C. BOWNASS, Scott KOHLERT.
Application Number | 20180152241 15/363385 |
Document ID | / |
Family ID | 62166030 |
Filed Date | 2018-05-31 |
United States Patent
Application |
20180152241 |
Kind Code |
A1 |
AL SAYEED; Choudhury A. ; et
al. |
May 31, 2018 |
SYSTEMS AND METHODS FOR A PREEMPTIVE CONTROLLER FOR PHOTONIC LINE
SYSTEMS
Abstract
Systems and methods for an optical controller in an optical
network for photonic control include monitoring signal output of
the optical controller for fluctuations on a signal, wherein the
optical controller may operate with one or more upstream optical
controllers in the optical network which may cause the
fluctuations; and initiating a control loop by the optical
controller on the signal based on the monitored fluctuations. The
initiating is based on one or more thresholds used to differentiate
actions of the one or more upstream optical controllers of the
plurality of optical controllers from one or more of fiber faults,
natural fluctuations in fiber plant, and natural fluctuations in
hardware.
Inventors: |
AL SAYEED; Choudhury A.;
(Gloucester, CA) ; BOWNASS; Dave C.; (Ottawa,
CA) ; KOHLERT; Scott; (Nepean, CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Ciena Corporation |
Hanover |
MD |
US |
|
|
Family ID: |
62166030 |
Appl. No.: |
15/363385 |
Filed: |
November 29, 2016 |
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
H04B 10/0791 20130101;
H04J 14/021 20130101 |
International
Class: |
H04B 10/079 20060101
H04B010/079; H04J 14/02 20060101 H04J014/02; H04B 10/70 20060101
H04B010/70 |
Claims
1. A method implemented by an optical controller in an optical
network for photonic control, the method comprising: monitoring
signal output of the optical controller for fluctuations on a
signal, wherein the optical controller may operate with one or more
upstream optical controllers in the optical network which may cause
the fluctuations; and initiating a control loop by the optical
controller on the signal based on the monitored fluctuations,
wherein the monitoring and the initiating are based on a plurality
of thresholds comprising a Monitor Start Threshold (MST) used to
start the monitoring, a Perturbation Acceptance Threshold (PAT)
used to determine the initiating where the PAT is indicative of an
ability to converge despite the fluctuations in the signal, and a
Controller Convergence Threshold (CCT) used to declare convergence
of the control loop.
2. The method of claim 1, wherein the fluctuations are based on one
or more of fiber faults, natural fluctuations in fiber plant,
natural fluctuations in hardware, and due to actions of the one or
more upstream optical controllers.
3. The method of claim 1, wherein the initiating is based on one or
more thresholds used to differentiate actions of the one or more
upstream optical controllers from one or more of fiber faults,
natural fluctuations in fiber plant, and natural fluctuations in
hardware.
4. (canceled)
5. The method of claim 1, wherein CCT.ltoreq.PAT and
PAT.ltoreq.MST.
6. The method of claim 1, wherein the CCT is adjusted by the
optical controller dynamically based on measurements when the
fluctuations are less than or equal to the PAT but greater than the
CCT.
7. The method of claim 1, wherein the initiating occurs when the
fluctuations are less than the PAT, and wherein the control loop is
stopped when the fluctuations exceed the PAT.
8. The method of claim 1, wherein the optical controller and the
one or more upstream optical controllers operate independently of
one another without messaging for coordination.
9. The method of claim 1, further comprising: subsequent to the
initiating, performing a subsequent monitoring of the fluctuations
after a first control cycle of the control loop; and one of
continuing the control loop based on the subsequent check and
delaying the control loop based on the fluctuations in the
subsequent monitoring.
10. The method of claim 9, wherein the subsequent monitoring is
performed for a random period of time determined by control cycles
of the optical controller.
11. The method of claim 1, wherein the monitoring detects an
increase of the fluctuations over time, and the initiating waits
for a period of time while the fluctuations settle down.
12. An optical controller adapted to perform photonic control in an
optical network, the optical controller comprising: a processor; a
communications interface communicatively coupled to the processor;
and memory storing instructions that, when executed, cause the
processor to monitor signal output of the optical controller for
fluctuations on a signal, wherein the optical controller operates
with one or more upstream optical controllers in the optical
network which may cause the fluctuations, and initiate a control
loop by the optical controller on the signal based on the monitored
fluctuations, wherein the signal output is monitored and the
control loop is initiated based on a plurality of thresholds
comprising a Monitor Start Threshold (MST) used to start the
monitor, a Perturbation Acceptance Threshold (PAT) used to initiate
the control loop where the PAT is indicative of an ability to
converge despite the fluctuations in the signal, and a Controller
Convergence Threshold (CCT) used to declare convergence of the
control loop.
13. The optical controller of claim 12, wherein the fluctuations
are based on one or more of fiber faults, natural fluctuations in
fiber plant, natural fluctuations in hardware, and due to actions
of the one or more upstream optical controllers of the plurality of
optical controllers.
14. The optical controller of claim 12, wherein the control loop is
initiated based on one or more thresholds used to differentiate
actions of the one or more upstream optical controllers from one or
more of fiber faults, natural fluctuations in fiber plant, and
natural fluctuations in hardware.
15. (canceled)
16. The optical controller of claim 12, wherein the CCT is adjusted
by the optical controller dynamically based on measurements when
the fluctuations are less than or equal to the PAT but greater than
the CCT.
17. The optical controller of claim 12, wherein the control loop is
initiated when the fluctuations are less than the PAT, and wherein
the control loop is stopped when the fluctuations exceed the
PAT.
18. The optical controller of claim 12, wherein the memory storing
instructions that, when executed, further cause the processor to
subsequent to the control loop being initiated, perform a
subsequent monitor of the fluctuations after a first control cycle
of the control loop, and one of continue the control loop based on
the subsequent monitor and delay the control loop based on the
fluctuations in the subsequent check.
19. An optical network, comprising: a plurality of interconnected
Optical Add/Drop Multiplexer (OADM) nodes; a plurality of links
interconnecting the OADM nodes; and a plurality of optical
controllers configured to perform photonic control, wherein each of
the plurality of optical controllers are adapted to monitor signal
output of an associated optical controller for fluctuations on a
signal, and initiate a control loop by the associated optical
controller on the signal based on the monitored fluctuations,
wherein the signal output is monitored and the control loop is
initiated based on a plurality of thresholds comprising a Monitor
Start Threshold (MST) used to start the monitor, a Perturbation
Acceptance Threshold (PAT) used to initiate the control loop where
the PAT is indicative of an ability to converge despite the
fluctuations in the signal, and a Controller Convergence Threshold
(CCT) used to declare convergence of the control loop.
20. The optical network of claim 19, wherein the control loop is
initiated, at the associated optical controller, based on one or
more thresholds used to differentiate actions of one or more
upstream optical controllers of the plurality of optical
controllers from one or more of fiber faults, natural fluctuations
in fiber plant, and natural fluctuations in hardware.
Description
FIELD OF THE DISCLOSURE
[0001] The present disclosure generally relates to optical
networking systems and methods. More particularly, the present
disclosure relates to systems and methods for a preemptive
controller for photonic line systems.
BACKGROUND OF THE DISCLOSURE
[0002] In optical networks with Optical Add/Drop Multiplexers
(OADMs) and interconnected mesh topologies, various photonic
controllers operate concurrently, such as on a per optical section
basis (OADM to adjacent OADM), to maintain spectral shape and to
obtain a desired level of performance. Such controllers,
categorized either as per channel or spectrum controllers, are
analog controllers that work on an analog optical signal as input,
use optical channel monitors located either at the start or at the
end of an optical section to get necessary feedback, and generate
an analog response, which is then used to make proportionate
actuator changes to achieve a desired level of analog optical
signal at the output. As described herein, a controller is a
processing device, apparatus, service, etc. which is implementing a
control loop. There can be multiple controllers operating
concurrently throughout the optical network, and the technical
challenge is maintaining stability since any adjustments can affect
in-service traffic carrying channels. To address the concurrent
operation, conventional approaches rely on messaging between
controllers for coordination or sequential operation. The messaging
approach is complex, especially for large mesh networks. The
sequential operation is slow to converge and difficult to
coordinate in large mesh networks. Other approaches have looked to
avoid messaging while preserving parallel operation using damped
responses in the associated control loops, the damped responses
slow convergence allowing upstream controllers to converge first
without coordination, although such method cannot stop ringing
effect or amplifying the existing output fluctuations before
achieving stability. The challenge for photonic line systems is how
to manage the operation of a plurality of controllers operating
control loops concurrently and maintaining stability avoiding the
limitations of conventional approaches, namely complex messaging,
which does not scale in large mesh networks, forced sequential
operation which is difficult to coordinate, and damped responses
that cannot minimize existing signal fluctuations at the controller
output.
BRIEF SUMMARY OF THE DISCLOSURE
[0003] In an exemplary embodiment, a method implemented by an
optical controller in an optical network for photonic control
includes monitoring signal output of the optical controller for
fluctuations on a signal, wherein the optical controller may
operate with one or more upstream optical controllers in the
optical network which may cause the fluctuations; and initiating a
control loop by the optical controller on the signal based on the
monitored fluctuations. The fluctuations can be based on one or
more of fiber faults, natural fluctuations in fiber plant, natural
fluctuations in hardware, and due to actions of the one or more
upstream optical controllers. The initiating can be based on one or
more thresholds used to differentiate actions of the one or more
upstream optical controllers from one or more of fiber faults,
natural fluctuations in fiber plant, and natural fluctuations in
hardware. The monitoring is based on a plurality of thresholds
which can include a Monitor Start Threshold (MST) used to start the
monitoring, a Perturbation Acceptance Threshold (PAT) used to
determine the initiating where the PAT is indicative of an ability
to converge despite the fluctuations in the signal, and a
Controller Convergence Threshold (CCT) used to declare convergence
of the control loop. The CCT.ltoreq.PAT and PAT.ltoreq.MST. The CCT
can be adjusted by the optical controller dynamically based on
measurements when the fluctuations are less than or equal to the
PAT but greater than the CCT.
[0004] The initiating can occur when the fluctuations are less than
the PAT, and wherein the control loop is stopped when the
fluctuations exceed the PAT. The optical controller and the one or
more upstream optical controllers operate independently of one
another without messaging for coordination. The method can further
include, subsequent to the initiating, performing a subsequent
monitoring of the fluctuations after a first control cycle of the
control loop; and one of continuing the control loop based on the
subsequent check and delaying the control loop based on the
fluctuations in the subsequent monitoring. The subsequent
monitoring can be performed for a random period of time determined
by control cycles of the optical controller. The monitoring can
detect an increase of the fluctuations over time, and the
initiating can wait for a period of time while the fluctuations
settle down.
[0005] In another exemplary embodiment, an optical controller
adapted to perform photonic control in an optical network includes
a processor; a communications interface communicatively coupled to
the processor; and memory storing instructions that, when executed,
cause the processor to monitor signal output of the optical
controller for fluctuations on a signal, wherein the optical
controller operates with one or more upstream optical controllers
in the optical network which may cause the fluctuations, and
initiate a control loop by the optical controller on the signal
based on the monitored fluctuations. The fluctuations can be based
on one or more of fiber faults, natural fluctuations in fiber
plant, natural fluctuations in hardware, and due to actions of the
one or more upstream optical controllers of the plurality of
optical controllers. The control loop can be initiated based on one
or more thresholds used to differentiate actions of the one or more
upstream optical controllers from one or more of fiber faults,
natural fluctuations in fiber plant, and natural fluctuations in
hardware.
[0006] The signal output is monitored based on a plurality of
thresholds which can include a Monitor Start Threshold (MST) used
to start the monitor, a Perturbation Acceptance Threshold (PAT)
used to initiate the control loop where the PAT is indicative of an
ability to converge despite the fluctuations in the signal, and a
Controller Convergence Threshold (CCT) used to declare convergence
of the control loop. The CCT can be adjusted by the optical
controller dynamically based on measurements when the fluctuations
are less than or equal to the PAT but greater than the CCT. The
control loop can be initiated when the fluctuations are less than
the PAT, and wherein the control loop is stopped when the
fluctuations exceed the PAT. The memory storing instructions that,
when executed, can further cause the processor to, subsequent to
the control loop being initiated, perform a subsequent monitor of
the fluctuations after a first control cycle of the control loop,
and one of continue the control loop based on the subsequent
monitor and delay the control loop based on the fluctuations in the
subsequent check.
[0007] In a further exemplary embodiment, an optical network
includes a plurality of interconnected Optical Add/Drop Multiplexer
(OADM) nodes; a plurality of links interconnecting the OADM nodes;
and a plurality of optical controllers configured to perform
photonic control, wherein each of the plurality of optical
controllers are adapted to monitor signal output of an associated
optical controller for fluctuations on a signal, and initiate a
control loop by the associated optical controller on the signal
based on the monitored fluctuations. The control loop can be
initiated, at the associated optical controller, based on one or
more thresholds used to differentiate actions of one or more
upstream optical controllers of the plurality of optical
controllers from one or more of fiber faults, natural fluctuations
in fiber plant, and natural fluctuations in hardware.
BRIEF DESCRIPTION OF THE DRAWINGS
[0008] The present disclosure is illustrated and described herein
with reference to the various drawings, in which like reference
numbers are used to denote like system components/method steps, as
appropriate, and in which:
[0009] FIG. 1 is a network diagram of an exemplary optical network
with interconnected OADM nodes;
[0010] FIG. 2 is a network diagram of two exemplary optical
sections in the optical network of FIG. 1 with associated
controllers and photonic hardware;
[0011] FIG. 3 is a network diagram of a mesh optical network with a
plurality of OADM nodes;
[0012] FIG. 4 is a block diagram and graphs of responses of optical
controllers when acting on already fluctuating input signals
without knowing the status of upstream dependent controllers;
[0013] FIG. 5 is a block diagram and graph of a preemptive
controller which can be used for the optical controllers;
[0014] FIGS. 6(A), 6(B), and 6(C) are graphs of average
fluctuations measured on output signals over time;
[0015] FIG. 7 is a network diagram of an optical network with a
network-wide centralized preemptive controller;
[0016] FIGS. 8 and 9 are a block diagram and graphs of dependencies
amongst an upper layer preemptive controllers with associated local
preemptive controllers under a span of control of an associated
upper layer preemptive controller; and
[0017] FIG. 10 is a block diagram of a processing device to
implement the various controllers described herein.
DETAILED DESCRIPTION OF THE DISCLOSURE
[0018] Again, in various exemplary embodiments, the present
disclosure relates to systems and methods for a preemptive
controller for photonic line systems. The systems and methods
utilize fluctuating conditions or perturbation amount on existing
signals to determine whether to initiate control loops by a
controller. Typical controllers act on the output first and make
adjustments (e.g., attenuator adjustments, etc.) before realizing
at some point that the controller is not converging for a certain
duration. The typical controller then either continues an attempt
at convergence or terminates at some point. In the systems and
methods described herein, the proposed controller takes the
signals' fluctuating conditions into account before initiating any
control loops and hence adjustments (e.g., attenuator adjustments,
etc.). Thus, the controller minimizes the occurrences of running
control actions simultaneously, while other upstream controllers
are potentially acting on the same set or a fractional set of input
signals, without requiring coordination or messaging between the
controllers. That is, the preemptive controllers described herein
do not rely on explicit coordination between one another such as
via messaging, do not require sequential operation, and do not
dampen responses. The systems and methods define the decision
points of the controllers' start of actions based on fluctuations
and perturbations. Thus, the preemptive controllers can operate
concurrently with other same layer controllers without requiring
the aforementioned conventional approaches. It has been determined
that using fluctuations and perturbations to determine whether to
start a controller plays a significant role in maintaining
stability in large mesh optical networks. Thus, the systems and
methods achieve stability in the multiple controllers by defining
specific start conditions based on monitored fluctuations and
perturbations.
Exemplary Optical Network
[0019] Referring to FIG. 1, in an exemplary embodiment, a network
diagram illustrates an exemplary optical network 10 with
interconnected OADM nodes 12 (labeled as 12A-12J). Note, as
described herein, the terms photonic network and optical network
are synonymous and can refer to a Dense Wave Division Multiplexing
(DWDM) network using fixed grid spectrum, flexible grid spectrum,
etc. The OADM nodes 12 are interconnected by a plurality of links
14 and the OADM nodes 12 can include a Reconfigurable OADM or the
like such as realized with Wavelength Selective Switches (WSSs) or
the like. The OADM nodes 12 provide the photonic layer (e.g., Layer
0) and various functionality associated therewith (e.g.,
multiplexing, amplification, optical routing, wavelength
conversion/regeneration, local add/drop, wavelength switching,
etc.) including photonic control 20. The photonic layer and the
photonic control 20 operating thereon can also include intermediate
amplifiers 22 and/or regenerators (which are omitted for
illustration purposes) on the links 14. The optical network 10 is
illustrated, for example, as an interconnected mesh network, and
those of ordinary skill in the art will recognize the network 10
can include other architectures, with OADM nodes 12 or with fewer
OADM nodes 12, with additional amplifiers 22, with additional
network elements and hardware, etc. Those of ordinary skill in the
art will recognize the systems and methods described herein can be
used in any optical networking scenario for the optical network 10,
and the optical network 10 is merely presented for illustration
purposes.
[0020] The OADM nodes 12 are connected with one another optically
over the links 14. OADM nodes 12 can be network elements which
include a plurality of ingress and egress ports forming the links
14. As described herein, a port may be formed by a transceiver
module or optical modem to provide an optical connection between
the OADM nodes 12. That is, in addition to the OADM nodes 12, there
can be network elements with optical transceivers or modems
included therein, such as switches, routers, terminals, etc.
operating any upper layer protocols. The optical network 10 can
include the photonic control 20 which can be viewed as a control
algorithm/loop for managing wavelengths from a physical perspective
at Layer 0. As described herein, sectional controllers, spectrum
controllers, and per channel controllers are implemented in the
photonic control 20. In one aspect, the photonic control 20 is
configured to add/remove wavelengths from the links in a controlled
manner to minimize impacts to existing, in-service,
traffic-carrying channels. For example, the photonic control 20 can
adjust modem launch powers, optical amplifier gain, variable
optical attenuator (VOA) settings, WSS parameters, etc. The
photonic control 20 can also be adapted to perform network
optimization on the links 14. This optimization can also include
re-optimization where appropriate.
Optical Sections
[0021] Referring to FIG. 2, in an exemplary embodiment, a network
diagram illustrates two exemplary optical sections 30 (labeled as
30A, 30B) in the optical network 10 with associated controllers and
photonic hardware. As described herein, an optical section 30 is
between two OADM nodes 12 which are channel access points for
optical channels. Thus, each optical section 30 has the same number
of channels at ingress and egress, i.e., typically in each optical
section 30, spectral loading remains the same at a given time. In
the example of FIG. 2, there are three OADM nodes 12A, 12B, 12C
with the optical section 30A between the OADM nodes 12A, 12B and
the optical section 30B between the OADM nodes 12B, 12C. FIG. 3
illustrates the optical sections 30A, 30B from a unidirectional
perspective from the OADM node 12A to the OADM node 12B to the OADM
node 12C, i.e., right to left for channel flow. Of course, the
unidirectional perspective is for simplicity of illustration and
actual deployments can include a second set of equipment for
bidirectional communication.
[0022] The photonic hardware includes a multiplexer 32 (labeled as
a multiplexer 32A at the OADM node 12A and a multiplexer 32B at the
OADM node 12B), a demultiplexer 34 (labeled as a demultiplexer 34A
at the OADM node 12B and a demultiplexer 34B at the OADM node 12C),
optical amplifiers 36, and Optical Channel Monitors (OCM) 38
(labeled as OCMs 38A, 38B, 38C, 38D). The multiplexers 32 and the
demultiplexers 34 can be WSSs or the like. The multiplexers 32
include a plurality of input ports and a common output port, and
the multiplexers 32 have an ability to control optical power of
individual channels as well as a composite signal on the common
output port. The demultiplexers 34 includes a common input port and
a plurality of output ports, and the demultiplexers 34 have an
ability to control optical power of individual channels as well as
a composite signal on the common input port.
[0023] For illustration purposes, optical links 14A, 14B between
the OADM nodes 12A, 12B, 12C are shown only with pre and post
amplifiers 36. Of course, there can be zero or more intermediate
amplifiers 36 on the links 14A, 14B which are omitted for
illustration purposes. The OCMs 38 are configured to receive a tap
off an input or output to each OADM node 12 (note, the inputs and
outputs to each OADM node 12 can be referred to as degrees). The
OCMs 38 use the tap to determine per channel and overall power
profiles of optical channels.
[0024] In operation, the multiplexer 32 receives channels 40A from
upstream degrees as well as locally added channels 42A. The
channels 40A, 42A are added together by the multiplexer 32A and
provided to the OADM node 12B over the optical link 14A. The OADM
node 12B receives the channels 40A, 42A, provides some channels 40B
downstream to the OADM node 12C, some channels 42B for local drop,
and some channels 42C for local add. The multiplexer 32B combines
the channels 40B, 42C and provides them to the OADM node 12C over
the optical link 14B. The demultiplexer 34B receives the channels
40B, 42C and provides some channels 40C to downstream degrees (not
shown) and some channels 42D for local drop.
[0025] Again, the photonic control 20 is configured to implement
various controllers over the optical sections 30A, 30B. The
photonic control 20 can operate on a per section basis. Typically,
in each optical section 30A, 30B, where the spectral loading at a
given time remain unchanged, multiple controllers 50, 52, 54 can
co-exist to control channel powers, signal to noise ratios (SNRs),
and to maintain the spectrum shape to obtain a desired level of
performance. Specifically, the controllers can include a sectional
controller 50 (labeled as sectional controller 50A for the optical
section 30A and as sectional controller 50B for the optical section
30B). The sectional controller 50 can include a per channel
controller 52 (labeled as per channel controller 52A for the
optical section 30A and as per channel controller 52B for the
optical section 30B) and a spectrum controller 54 (labeled as
spectrum controller 54A for the optical section 30A and as spectrum
controller 54B for the optical section 30B). Such controllers 52,
54, categorized either as per channel controllers 52 or spectrum
controllers 54, are analog controllers that work on an analog
optical signal as input, use the OCMs 38 located either at the
start or at the end of the optical section 30 to get necessary
feedback, and generate an analog response, which is then used to
make proportionate actuator changes, such as via the multiplexer
32, the demultiplexer 34, etc., to achieve a desired level of
analog optical signal at the output. Note, while the controllers
50, 52, 54 are shown as separate, these can be the same devices and
this separation is shown for logical functionality, namely the per
channel controllers 52 operate on individual channels while the
spectrum controllers 54 operate on the overall spectrum, i.e., all
channels.
[0026] The controllers 50, 52, 54 can be implemented by a
processing device, server, controller, etc. to receive input,
perform optimization and provide an output related to the
appropriate settings of the various components. The input can
include optical spectrum at the start of the optical section 30,
source ports of each signal in the optical spectrum, destination
fiber degree of each signal in the optical spectrum, and signal
characteristics of each signal. The signal characteristics can
include, for example, control frequency or center frequency of the
signal, the signal's spectral shape, required bandwidth or spectral
spacing compared to neighboring signals, total signal power or
power spectral density of the incoming signal at the source port,
target launch power for each signal at the destination fiber span,
each signal's modulation characteristics, and Signal-to-Noise Ratio
(SNR) bias preferences compared to neighboring signals.
Mesh Optical Network
[0027] Referring to FIG. 3, in an exemplary embodiment, a network
diagram illustrates a mesh optical network 10A with a plurality of
OADM nodes 12 (labeled as A-J). A key assumption for typical
photonic controllers 50, 52, 54 is that the input signal fed into
the controller 50, 52, 54 is stable or has minimal breathing (i.e.,
very small fluctuations such as .ltoreq.0.4 dB peak-to-peak) that
typically falls under most controllers' negligible error terms or
dead zone. Again, conventionally, such stability on the input
signal is typically ensured by establishing a messaging
infrastructure among multiple photonic controllers so that, at any
single time event, more than one controller does not act or makes
active changes on the optical signal. Typically for linear
networks, where controllers are linearly dependent with each other,
such message-based framework can be scaled to maintain stability
among controllers. However, when the number of dependent
controllers, who will be acting on a common optical signal or a
group of optical signals, are stretched in a mesh network
environment or even in a long-haul linear network environment with
meshed channel add/drops, such as in the optical network 10A, the
message-based infrastructure cannot scale to maintain the sequence
of events among the controllers.
[0028] In the mesh optical network 10A, a plurality of photonic
controllers, each acting on a group of optical signals in a
specific fiber direction, are optically dependent on each other,
and when two and more controllers in such large setup act
simultaneously, they create conflict between controllers and create
unusual perturbations on the optical signals leading to
instability. In such complex mesh optical networks 10A, which is
typical in metro network deployments, establishing a message-based
framework among controllers to sequence their action items is going
to be difficult to achieve. In some network examples, instead of
peer-to-peer messaging, it is also possible to establish a
centralized controller to act as a master to control the sequence
of action events among all photonic controllers. Since in such
centralized control setup, each individual photonic controller
cannot take its own decisions in terms of control loop initiations,
the overall performance recovery of a network following a fiber
fault or capacity change could be very slow. In addition, if any of
the photonic controllers loses communication with the centralized
master, the network cannot recover autonomously out of fault.
[0029] Referring to FIG. 4, in an exemplary embodiment, a block
diagram and graphs illustrate responses of optical controllers 50,
52, 54 when acting on already fluctuating input signals without
knowing the status of upstream dependent controllers. In operation,
the controller 50, 52, 54 receives an input 60 which is combined
with a measured output 62 to form an error signal 64. The
controller 50, 52, 54 performs a control loop which can result in
actuator 66 adjustments which are combined with an output 68 of the
controller 50, 52, 54 to provide the measured output 62.
Fluctuations 70 on the input signals 60 can either come from fiber
faults, from natural fluctuations in fiber plant (e.g.,
Polarization Dependent Loss (PDL)) or hardware (e.g., dither), or
simply due to actions of one or more upstream controllers 50, 52,
54. As the example illustrates, depending on the fluctuation 70
amount and frequency of the input signal 60, the controller 50, 52,
54 can make wrong decisions on actuator adjustments that in turn,
widens the fluctuation amount (shown on output signal 72). When a
downstream controller 50, 52, 54 further reacts, it amplifies the
fluctuation even more (shown in graphs 74, 76), ultimately leading
to instability in the network.
Preemptive Controllers
[0030] As described herein, stability between multiple photonic
controllers 50, 52, 54 running in mesh optical networks 10, 10A is
a long-standing issue. In a large mesh optical network 10, 10A,
where optical channels add/drop over a short distance between the
OADM nodes 12, there are complex optical dependencies among the
controllers 50, 52, 54. In these highly meshed optical networks 10,
10A, maintaining a message-based solution often creates deadlock
conditions. Using a timer-based solution, where at some point,
multiple controllers wake up at the same time and make actuator
changes at different parts of the network, can also lead to the
large ringing of optical signals.
[0031] To ensure stability in the mesh optical networks 10, 10A,
while maintaining the scalability of adapting the controllers 50,
52, 54, a preemptive controller is proposed that can be adapted
both for per channel controllers 52 or for spectrum controllers 54
in the mesh optical networks 10, 10A. The key concept of the
preemptive controller is to take the existing optical signals'
perturbation into account at the output, i.e., at the feedback
monitoring point such as via the OCM 38 before even initiating its
own control loops. The preemptive controller decides whether to
initiate the control loop in the current fluctuating state at the
output and, if initiated, whether to adapt the controller
convergence threshold criteria dynamically to match the existing
fluctuating condition.
[0032] Referring to FIG. 5, in an exemplary embodiment, a block
diagram and graph illustrate a preemptive controller 100 which can
be used for the controllers 50, 52, 54. The preemptive controller
100, similar to FIG. 4, has the input 60 combined with the measured
output 62 to provide the error signal 64 to the preemptive
controller 100 which implements a control loop providing the output
68 which is combined with the actuator 66 and associated
adjustments thereon based on the control loop. The output 102 is
the same as the measured output 62 and an example graph 104
illustrates the output 102.
[0033] The primary objective of the proposed preemptive controller
100 is to initiate control loop actions only when it is ensured to
be safe without disrupting the stability of the optical networks
10, 10A. The preemptive controller 100 works based on three key
thresholds 110, 112, 114 which are monitored on the output signal
102, such as based on the OCM 38. Note, with the preemptive
controller 100 inoperable, the output signal 102 is substantially
the same as the input 60. The three key thresholds 110, 112, 114
are used to decide if fluctuations exist on the signals at the
point of control, whether to initiate the control loop in its
current monitoring state and if initiated, whether to adapt the
convergence threshold criteria dynamically to match the existing
fluctuating condition. The three key threshold points 110, 112, 114
are Monitor Start Threshold (MST) 110, Controller Convergence
Threshold (CCT) 112, and Perturbation Acceptance Threshold (PAT)
114.
[0034] The MST 110 is the threshold level at which point if the
measured error 64, which is the delta between measured output 62
and a target at any point in time becomes higher than the given
level, the preemptive controller 100 wakes up and starts monitoring
the output signal 102. The CCT 112 is the threshold that the
preemptive controller 100 uses to declare convergence or exit
control loops if the measured error 64 at any given time or for a
duration of time falls within this defined threshold. The PAT 114
is the threshold where the preemptive controller 100 determines it
can converge successfully even if the input signal 60 is
fluctuating within that threshold. All these three threshold values
110, 112, 114 can be different from controller to controller based
on the configuration where the preemptive controller 100 is running
and the corresponding signal(s) and target(s). The threshold values
110, 112, 114 can be bounded between a positive and negative value
with a check indicative of the output signal 102 being with the
bound.
[0035] Both the MST 110 and the PAT 114 values can be set during
the design phase of the preemptive controller 100, while the CCT
112 value can be adjusted by the preemptive controller 100
dynamically based on real-time measurements of signals. There can
be a validity check on the preemptive controller 100 that always
ensures that, CCT 112.ltoreq.PAT 114 and PAT 114.ltoreq.MST 110.
Before starting the control on a signal, the output signal 102 is
monitored at the output monitoring point at a sample
rate.gtoreq.controller iteration cycle and for a
duration.gtoreq.controller iteration cycle.
[0036] If the output 102 is fluctuating by an amount greater than
the CCT 112, but less than or equal to the PAT 114, the CCT 112 can
be moved to match the (average fluctuation amount+.epsilon.), where
.epsilon. is small error offset to overcome the measurement and
actuator setting errors. If the output signal 102 fluctuation is
greater than the PAT 114, then the preemptive controller 100
freezes (does not make any actuator adjustments), and keeps
monitoring the output 102 until fluctuation becomes lower than or
equal to the PAT 114.
[0037] The preemptive controller 100 can be the per channel
controller 52 where controller monitors power or Optical Signal to
Noise Ratio (OSNR) at its output monitoring point. The preemptive
controller 100 can be the spectrum controller 54 as well, where the
controller monitors power or OSNR of the plurality of channels at
its output monitoring point.
[0038] Further, the preemptive controller 100, before initiating
the control loops, tracks the rate of change of the average amount
of fluctuations in the output signal(s) 102 over time that helps to
differentiate fluctuations due to natural causalities versus
interactions due to upstream controllers. This also helps the
preemptive controller 100 to decide when and where to re-adjust the
convergence threshold before initiating the control loops.
Referring to FIGS. 6(A), 6(B), and 6(C), in exemplary embodiments,
graphs illustrate average fluctuations measured on output signals
102 over time. As shown in FIG. 6(A), the preemptive controller 100
takes samples of output signal values over certain durations, where
a delta between output signal values (power or OSNR) collected over
n number of samples are further averaged to get an average
fluctuation amount at any given time t.sub.n. If the average
fluctuation amount measured over certain duration remain within
+/-X dB of each other, where X is small finite number (e.g.,
X.ltoreq.0.4 dB), the preemptive controller 100 considers that
condition as a "natural state", at which point, the convergence
threshold is re-adjusted to match (average
fluctuation+X/2+.epsilon.), where .epsilon. is a small error offset
to overcome the measurement and actuator setting errors (e.g.,
.epsilon..ltoreq.0.1 dB).
[0039] However, as shown in FIG. 6(B), if the average fluctuation
amount grows up over a certain amount of time, the preemptive
controller 100 considers the fluctuation is coming due to
interactions of upstream controllers. At this point, the preemptive
controller 100 does not re-adjust the CCT 112 and does not initiate
control loops, even if the average fluctuation amount is less than
the PAT 114. Similarly, as shown in FIG. 6(C), if the average
fluctuation amount is reduced over time, the preemptive controller
100 considers the state as settling down for upstream controllers.
The preemptive controller 100 does not re-adjust the CCT 112 and
does not initiate control loops in this state, even if the average
fluctuation amount is less than the PAT 114. In both cases, the CCT
112 is re-adjusted only when the measured average fluctuation
amount reaches the "natural state."
[0040] If analog optical signals are perturbing beyond the CCT 112,
a typical controller will make things worse when applying the
control on top of already perturbed signals. The key objective of
the preemptive controller 100 is to maintain the stability of the
optical network 10, 10A while achieving the objective of per
channel power or OSNR target or maintaining power or OSNR
equalization over the spectrum at targeted points in an optical
section 30. There are some controllers that initiate their control
loops simultaneously and independently without any messaging from
upstream controllers, and take the input power variations into
account to adapt the control loop parameters at least after running
one or more iterations to make active changes to actuators.
Compared to that, the preemptive controller 100 is designed to take
the existing signals' perturbation into account at the output or
feedback monitoring point before even starting its own control
loops and decide even whether to initiate to control loop in the
current state, and if initiated, whether to adapt the convergence
threshold criteria dynamically to match the existing fluctuating
condition.
[0041] Also, the preemptive controller 100 works independently
without any messaging to coordinate actions between upstream or
downstream controllers, making the preemptive controller 100
perfectly scalable for large linear or mesh networks 10, 10A. By
monitoring the signals in its own control loop and feedback
mechanism ahead of time, the preemptive controller 100 minimizes
any occurrences of running control actions simultaneously while
other upstream controllers are potentially acting on the same set
or a fractional set of input signals. In this way, the fluctuation
on the optical signals does not get amplified when traversing
through a single or plurality of controllers over the optical
network 10, 10A--all of which can potentially react on the same
optical signal input without any messaging or external coordination
between them.
[0042] Referring to FIG. 7, in an exemplary embodiment, a network
diagram illustrates an optical network 10B with a network-wide
centralized preemptive controller 100A. Here, the preemptive
controller 100 can monitor controlled signals' fluctuations at a
plurality of monitoring points (for example, at the start and at
the end of each OADM section, via the OCMs 38), and based on the
fluctuation amount, either preempt local node level or device
controllers 50, 52, 54, 100 not to initiate their control loop
actions, or start their control loops with re-adjusted CCT 112 so
that the overall signal fluctuation amount originated primarily due
to natural causalities does not grow up with controller initiated
actions.
Dependencies Amongst Preemptive Controllers
[0043] Referring to FIGS. 8 and 9, in an exemplary embodiment, a
block diagram and graphs illustrate dependencies amongst an upper
layer preemptive controllers 100B (labeled as 100B1, 100B2, 100B3)
with associated local preemptive controllers 100C under a span of
control by an associated upper layer preemptive controller 100B.
The upper layer controllers 100B have optical dependencies between
one another, i.e., control of optical signals by one upper layer
controller 100B has an effect on other upper layer controllers
100B, but there is no communication between the upper layer
controllers 100B. In an exemplary embodiment, every upper layer
preemptive controller 100B runs monitoring and controller cycles on
their own clocks/ticks that reduce the probability of sync up
between multiple optically dependent controllers.
[0044] Graphs 150, 152, 154 illustrate a control cycle for the
upper layer preemptive controllers 100B1, 100B2, 100B3,
respectively. At a point 160, the upper layer preemptive controller
100B1 detects the error at the controller output is large enough to
trigger controller cycles, in the graph 150. At a point 162, the
upper layer preemptive controller 100B2 monitor detects the error
at the controller output is large enough to trigger controller
cycles, in the graph 152. At a point 164, the upper layer
preemptive controller 100B3 monitor detects the error at the
controller output is large enough to trigger controller cycles, in
the graph 154.
[0045] As seen in the graphs 150, 152, 154, over N number of
controllers, it is possible that two or more optically dependent
controllers' monitoring cycle sync up allowing downstream
controller(s) to initiate control loop regardless of the initial
preemptive check. The N.sup.th controller may take a longer time to
settle down due to perturbations coming from upstream controllers,
and can generate some ringing effect on channel powers.
[0046] To avoid this issue in the graph 154, each upper layer
preemptive controller 100B can run a preemptive check before
initiating the control cycle for a random period of time limited by
a minimum of two regular controller cycle time to a maximum of m
number of controller cycle time, where m>2. Adding this extra
randomness in preemptive check time reduces further sync-up
probabilities between upper layer preemptive controllers 100B.
[0047] If for an upper layer controller 100B, all channels are
originated from local add access points, then the duration of the
preemptive check be reduced as in this case, the upper layer
controller 100B only needs to check perturbations generated from
local controllers 100C within its span of control. This is also one
reason why, for a preemptive check, each controller has to check
its own output point for stability in errors instead of checking
solely at inputs (since checking inputs does not catch issues with
perturbations coming from local controllers 100C).
[0048] In addition, each controller 100B can run at least one
additional preemptive check after the first control cycle (i.e.
following a control loop and actuator settling time) that helps to
detect perturbations coming either from upstream controllers or
from stabilization of local controllers. In this way, even if a
downstream controller starts a control cycle before the initiation
of an upstream one, the downstream controller gets a chance to
detect the perturbations in the next preemptive check. Also, to
ensure network-wide stability, in each control cycle, every
controller can only compensate for a small percentage change in the
total error accumulated on that cycle or iteration.
[0049] The primary objective of a preemptive check for controllers
100 is to make sure two or more optically dependent controllers 100
do not overlap their control actions at the same time. Here,
maintaining the networkwide stability among controllers 100 (i.e.,
minimizing the probability of controller to controller chain
reaction or ringing effect) is the primary concern and overall
convergence time for an individual controller 100 are not the
priority. The proposed preemptive check helps controllers to detect
perturbations on the output and avoid running a control loop to
ensure stability as the first priority. On the second order,
controller 100 may decide to run a control loop on condition that
the fluctuations are within tolerable range and it can still
maintain existing stability without causing any further
amplification of the existing fluctuations in errors. In addition,
the random duration for the preemptive check, running each
controller on their own clock, and doing extra preemptive check
after a first control cycle helps to avoid synchronization or
lock-up between two or more optical dependent controllers.
[0050] FIG. 9 illustrates graphs 170, 172, 174, corresponding to
the graphs 150, 152, 154, with the implementation of the preemptive
check. In the graphs 170, 172, 174, the upper layer controller
100B1, 100B2, 100B3 monitors detect the error at the controller
output is large enough to trigger controller cycles. Each
controller runs a preemptive check at the start for a random period
of time limited by a minimum of two regular controller cycle time
to a maximum of m number of controller cycle time, where m>2,
adding this extra randomness in preemptive check time reduces
further sync-up probability between controllers. In addition, each
controller does an extra preemptive check after a first control
cycle that helps to detect perturbations coming either from
upstream controllers or from stabilization of local
controllers.
Exemplary Controller
[0051] Referring to FIG. 10, in an exemplary embodiment, a block
diagram illustrates a processing device 200 to implement the
controllers 50, 52, 54, 100. The processing device 200 can be part
of common equipment at a network element, such as the OADM 12, or a
stand-alone device communicatively coupled to the network element.
The processing device 200 can include a processor 202 which is a
hardware device for executing software instructions such as
operating the control plane. The processor 202 can be any custom
made or commercially available processor, a central processing unit
(CPU), an auxiliary processor among several processors associated
with the processing device 200, a semiconductor-based
microprocessor (in the form of a microchip or chip set), or
generally any device for executing software instructions. When the
processing device 200 is in operation, the processor 202 is
configured to execute software stored within the memory, to
communicate data to and from the memory, and to processing device
control operations of the processing device 200 pursuant to the
software instructions. The processing device 200 can also include a
network interface 204, a data store 206, memory 208, an I/O
interface 210, and the like, all of which are communicatively
coupled to one another and to the processor 202.
[0052] The network interface 204 can be used to enable the
processing device 200 to communicate, such as to communicate to the
network elements, OCMs 38, etc., to a management system, and the
like. The network interface 204 can include address, control,
and/or data connections to enable appropriate communications on the
network. The data store 206 can be used to store data, such as
control plane information, provisioning data, OAM&P data, etc.
The data store 206 can include any of volatile memory elements
(e.g., random access memory (RAM, such as DRAM, SRAM, SDRAM, and
the like)), nonvolatile memory elements (e.g., ROM, hard drive,
flash drive, CDROM, and the like), and combinations thereof.
Moreover, the data store 206 can incorporate electronic, magnetic,
optical, and/or other types of storage media. The memory 208 can
include any of volatile memory elements (e.g., random access memory
(RAM, such as DRAM, SRAM, SDRAM, etc.)), nonvolatile memory
elements (e.g., ROM, hard drive, flash drive, CDROM, etc.), and
combinations thereof. Moreover, the memory 208 may incorporate
electronic, magnetic, optical, and/or other types of storage media.
Note that the memory 208 can have a distributed architecture, where
various components are situated remotely from one another, but may
be accessed by the processor 202. The I/O interface 210 includes
components for the processing device 200 to communicate with other
devices.
[0053] In an exemplary embodiment, the processing device 200 is an
optical controller adapted to perform photonic control in an
optical network. The optical controller includes a processor; a
communications interface communicatively coupled to the processor;
and memory storing instructions that, when executed, cause the
processor to monitor signal output of the optical controller for
fluctuations on a signal, wherein the optical controller operates
with one or more upstream optical controllers in the optical
network which may cause the fluctuations, and initiate a control
loop by the optical controller on the signal based on the monitored
fluctuations. The fluctuations can be based on one or more of fiber
faults, natural fluctuations in fiber plant, natural fluctuations
in hardware, and due to actions of the one or more upstream optical
controllers of the plurality of optical controllers. The control
loop can be initiated based on one or more thresholds used to
differentiate actions of the one or more upstream optical
controllers of the plurality of optical controllers from one or
more of fiber faults, natural fluctuations in fiber plant, and
natural fluctuations in hardware.
[0054] The signal output is monitored based on a plurality of
thresholds including a Monitor Start Threshold (MST) used to start
the monitor, a Perturbation Acceptance Threshold (PAT) used to
initiate the control loop where the PAT is indicative of an ability
to converge despite the fluctuations in the signal, and a
Controller Convergence Threshold (CCT) used to declare convergence
of the control loop. In an exemplary embodiment, CCT.ltoreq.PAT and
PAT.ltoreq.MST. The CCT can be adjusted by the optical controller
dynamically based on measurements when the fluctuations are less
than or equal to the PAT but greater than the CCT.
[0055] The control loop can be initiated when the fluctuations are
less than the PAT, and wherein the control loop is stopped when the
fluctuations exceed the PAT. The optical controller and the
plurality of optical controllers operate independently of one
another without messaging for coordination. The memory storing
instructions that, when executed, can further cause the processor
to, subsequent to the control loop being initiated, perform a
subsequent monitoring of the fluctuations after a first control
cycle of the control loop, and one of continue the control loop
based on the subsequent monitoring and delay the control loop based
on the fluctuations in the subsequent monitoring.
[0056] In another exemplary embodiment, an optical network includes
a plurality of interconnected Optical Add/Drop Multiplexer (OADM)
nodes; a plurality of links interconnecting the OADM nodes; and a
plurality of optical controllers configured to perform photonic
control, wherein each of the plurality of optical controllers are
adapted to monitor signal output of an associated optical
controller for fluctuations on a signal, and initiate a control
loop by the associated optical controller on the signal based on
the monitored fluctuations. The control loop is initiated, at the
associated optical controller, based on one or more thresholds used
to differentiate actions of one or more upstream optical
controllers of the plurality of optical controllers from one or
more of fiber faults, natural fluctuations in fiber plant, and
natural fluctuations in hardware.
[0057] It will be appreciated that some exemplary embodiments
described herein may include one or more generic or specialized
processors ("one or more processors") such as microprocessors;
Central Processing Units (CPUs); Digital Signal Processors (DSPs):
customized processors such as Network Processors (NPs) or Network
Processing Units (NPUs), Graphics Processing Units (GPUs), or the
like; Field Programmable Gate Arrays (FPGAs); and the like along
with unique stored program instructions (including both software
and firmware) for control thereof to implement, in conjunction with
certain non-processor circuits, some, most, or all of the functions
of the methods and/or systems described herein. Alternatively, some
or all functions may be implemented by a state machine that has no
stored program instructions, or in one or more Application Specific
Integrated Circuits (ASICs), in which each function or some
combinations of certain of the functions are implemented as custom
logic or circuitry. Of course, a combination of the aforementioned
approaches may be used. For some of the exemplary embodiments
described herein, a corresponding device in hardware and optionally
with software, firmware, and a combination thereof can be referred
to as "circuitry configured or adapted to," "logic configured or
adapted to," etc. perform a set of operations, steps, methods,
processes, algorithms, functions, techniques, etc. on digital
and/or analog signals as described herein for the various exemplary
embodiments.
[0058] Moreover, some exemplary embodiments may include a
non-transitory computer-readable storage medium having computer
readable code stored thereon for programming a computer, server,
appliance, device, processor, circuit, etc. each of which may
include a processor to perform functions as described and claimed
herein. Examples of such computer-readable storage mediums include,
but are not limited to, a hard disk, an optical storage device, a
magnetic storage device, a ROM (Read Only Memory), a PROM
(Programmable Read Only Memory), an EPROM (Erasable Programmable
Read Only Memory), an EEPROM (Electrically Erasable Programmable
Read Only Memory), Flash memory, and the like. When stored in the
non-transitory computer readable medium, software can include
instructions executable by a processor or device (e.g., any type of
programmable circuitry or logic) that, in response to such
execution, cause a processor or the device to perform a set of
operations, steps, methods, processes, algorithms, functions,
techniques, etc. as described herein for the various exemplary
embodiments.
[0059] Although the present disclosure has been illustrated and
described herein with reference to preferred embodiments and
specific examples thereof, it will be readily apparent to those of
ordinary skill in the art that other embodiments and examples may
perform similar functions and/or achieve like results. All such
equivalent embodiments and examples are within the spirit and scope
of the present disclosure, are contemplated thereby, and are
intended to be covered by the following claims.
* * * * *