U.S. patent application number 12/624594 was filed with the patent office on 2010-04-22 for method and apparatus for automatic port interconnection discovery in an optical network.
This patent application is currently assigned to MERITON NETWORKS US INC.. Invention is credited to Murali Krishnaswamy, Xiaowen Mang, Jefferson L. Wagener.
Application Number | 20100098410 12/624594 |
Document ID | / |
Family ID | 25425836 |
Filed Date | 2010-04-22 |
United States Patent
Application |
20100098410 |
Kind Code |
A1 |
Krishnaswamy; Murali ; et
al. |
April 22, 2010 |
METHOD AND APPARATUS FOR AUTOMATIC PORT INTERCONNECTION DISCOVERY
IN AN OPTICAL NETWORK
Abstract
To automatically discover the port connections for all nodes in
a network, a master node generates a predetermined optical signal
and transmits the predetermined optical signal to a neighboring
node, which signal identifies the port on which the master node
transmitted the predetermined signal. The recipient transmits a
reply signal to the predecessor node and to the master node via a
control channel, which identifies a port on which the predetermined
optical signal was received by the neighboring node. By
successively repeating this process in a methodical manner, the
master node can discover all of the port interconnections in the
optical network. Also each node can discover all its port
interconnections to its neighbors. Moreover, by selecting
controlling the state (e.g., terminate, open) of the ports of the
non-master nodes in the network, the master node can control which
nodes receive the predetermined signal, thereby ensuring proper
port discovery.
Inventors: |
Krishnaswamy; Murali;
(Piscataway, NJ) ; Wagener; Jefferson L.;
(Morristown, NJ) ; Mang; Xiaowen; (Manalapan,
NJ) |
Correspondence
Address: |
MAYER & WILLIAMS PC
251 NORTH AVENUE WEST, 2ND FLOOR
WESTFIELD
NJ
07090
US
|
Assignee: |
MERITON NETWORKS US INC.
Wilmington
DE
|
Family ID: |
25425836 |
Appl. No.: |
12/624594 |
Filed: |
November 24, 2009 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
09908459 |
Jul 18, 2001 |
|
|
|
12624594 |
|
|
|
|
Current U.S.
Class: |
398/16 |
Current CPC
Class: |
H04Q 2011/0088 20130101;
H04Q 11/0062 20130101 |
Class at
Publication: |
398/16 |
International
Class: |
H04B 10/08 20060101
H04B010/08 |
Claims
1. A method for automatically discovering port interconnections in
an optical network comprising: transmitting a predetermined optical
signal from a first available port of a first node in the optical
network to a second node in the optical network, said predetermined
optical signal including node origination information and port
origination information; receiving the predetermined optical signal
at the second node in the optical network; determining on which
port the predetermined optical signal was received by the second
node in the optical network; and forwarding a reply signal to the
predecessor node and to the first node from the second node over a
control channel of the optical network, said reply signal including
node termination information and port termination information.
2. The method according to claim 1 wherein said first node is a
master node and said second node is a non-master node.
3. The method according to claim 2, further comprising updating a
port adjacency table in the master node upon receipt of the reply
signal.
4. The method according to claim 2, further comprising transmitting
a second predetermined optical signal over a second available port,
wherein the second predetermined optical signal includes node
origination information and port origination information.
5. The method according to claim 2, further comprising receiving a
connection discovery request signal from a non-master node in the
optical network at the master node in the optical network and
transmitting the first predetermined optical signal in response to
said discovery request signal.
6. The method according to claim 2, further comprising designating
a new node that comes online in the optical network as a secondary
master node, and limiting a port interconnection discovery
capability of the secondary master node to only immediate neighbors
of the secondary master node.
7. The method according to claim 6, further comprising transmitting
a second predetermined optical signal from a first available port
of a secondary master node in the optical network to a non-master
node in the optical network, said predetermined optical signal
including node origination information and port origination
information.
8. The method according to claim 7, further comprising: receiving
the second predetermined optical signal at the non-master node in
the optical network; and determining on which port the
predetermined optical signal was received by the non-master node in
the optical network.
9. The method according to claim 8, further comprising forwarding a
reply signal to the secondary master node from the non-master node
over a control channel of the optical network, said reply signal
including node termination information and port termination
information.
10. A method for automatically discovering connections in an
optical network comprising: transmitting a predetermined signal
from a first available port of a first node in the optical network
to a second node in the optical network, said predetermined signal
including node origination information and port origination
information; receiving the predetermined signal at the second node
in the optical network; determining on which port the predetermined
signal was received by the second node in the optical network; and
forwarding a reply signal to the predecessor node and to the first
node from the second node over a control channel of the optical
network, said reply signal including node termination information
and port termination information.
11. The method according to claim 10 wherein said first node is a
master node and said second node is a non-master node.
12. The method according to claim 11, further comprising updating a
port adjacency table in the master node upon receipt of the reply
signal.
13. The method according to claim 10, further comprising
transmitting a second predetermined optical signal over a second
available port, wherein the second predetermined optical signal
includes node origination information and port origination
information.
14. The method according to claim 10, further comprising receiving
a connection discovery request signal from a non-master node in the
optical network at the master node in the optical network and
transmitting the first predetermined optical signal in response to
said discovery request signal.
15. The method according to claim 10, further comprising
designating a new node that comes online in the optical network as
a secondary master node, and limiting a port interconnection
discovery capability of the secondary master node to only immediate
neighbors of the secondary master node.
16. The method according to claim 15, further comprising
transmitting a second predetermined signal from a first available
port of a secondary master node in the optical network to a
non-master node in the optical network, said predetermined signal
including node origination information and port origination
information.
17. The method according to claim 16, further comprising: receiving
the second predetermined signal at the non-master node in the
optical network; and determining on which port the predetermined
signal was received by the non-master node in the optical
network.
18. The method according to claim 17, further comprising forwarding
a reply signal to the secondary master node from the non-master
node over a control channel of the optical network, said reply
signal including node termination information and port termination
information.
Description
STATEMENT OF RELATED APPLICATION
[0001] This application is a continuation of U.S. patent
application Ser. No. 09/908,459, filed Jul. 18, 2001, entitled
"Method and Apparatus For Automatic Port Interconnection Discovery
in an Optical Network." The prior application is incorporated
herein by reference in its entirety.
BACKGROUND OF THE INVENTION
[0002] The present invention relates generally to methods and
apparatuses for provisioning communications channels in
telecommunication or computer networks, and more particularly to a
method and apparatus for provisioning communications channels in a
telecommunications or computer network that operates optically.
[0003] Communications networks, including optical communications
networks, generally comprise many nodes at which the data stream is
coupled from one incoming port to an outgoing port to route the
data stream to the desired destination. An example of an optical
node is an optical cross-connect or optical add-drop
multiplexer.
[0004] In an optical network, in addition to the payload channels,
each network node is sometimes connected to its adjacent or
neighboring nodes and a master node by a control channel. Each
network node may have many incoming and outgoing ports, e.g., 4, 8,
16, 64, etc. Each port is connected to a port on a neighboring
node. Multiple ports one on node may be connected to the same node
on their terminal side. However, each port on a node is terminated
in only one other port.
[0005] When provisioning connections in any network, including
optical networks, it is necessary to identify the port
interconnections between adjacent nodes (e.g., ONNs) before
attempting to setup channel connections, such as optical channel
connections in an optical network cross-connect. Manually
identifying the node interconnections and populating a port
adjacency table in each node is possible, but cumbersome, even if
the table is subsequently dynamically updated as and when
connections are made and released. This is particularly problematic
when the numbers of ports per node are large. Moreover, as data
networks are ever increasing in size, manual techniques for
identifying the ports will become increasingly tedious.
[0006] Currently, an automatic method for port interconnection
discovery in all-optical cross-connect-based network does not
exist. Although such a scheme may be feasible in an
optical-electrical cross-connect-based network, in which
optical-electrical-optical conversion takes place in each optical
network node (ONN), techniques for automated port interconnection
discovery are limited due to the lack of optical to electrical
conversion in an all-optical cross-connect. Moreover, in an optical
communication network, some nodes are not capable of generating an
information-bearing signal that can be transmitted over one of the
optical communications payload channels. This is because the node
lacks the ability to convert a signal from the electrical domain to
the optical domain. It is the precise ability of the node to
operate completely optically that allows the node to operate at
extremely high speeds. Consequently, requiring an electrical
conversion in the process would unduly limit the operating speed or
increase the cost. As a result, techniques for automatically
provisioning nodes in an optical/electrical communications network
are not possible in all-optical communications networks.
[0007] The present invention is therefore directed to the problem
of developing a method and apparatus for automated port
interconnection discovery in an optical network employing optical
cross-connects that operate completely optically (i.e.,
all-optical).
SUMMARY OF THE INVENTION
[0008] The present invention solves this and other problems by,
inter alia, generating a predetermined optical signal at a first
node, transmitting the predetermined optical signal from the first
node to a neighboring node, which signal identifies the port on
which the first node transmitted the predetermined signal, and
transmitting a reply signal from the neighboring node to its
predecessor node and to the first node via a control channel, which
identifies a port on which the predetermined optical signal was
received by the neighboring node. By successively repeating this
process in a methodical manner, the first node can discover all of
the port interconnections in the optical network. Also each node
can discover all its port interconnections to its neighbors.
Moreover, by selecting controlling the state (e.g., terminate,
open) of the ports of the secondary nodes in the network, the first
node can control which nodes receive the predetermined signal,
thereby ensuring proper port discovery.
[0009] In one embodiment of the invention, the first node is a
master node and the secondary nodes are non-master nodes.
BRIEF DESCRIPTION OF THE DRAWINGS
[0010] FIG. 1 depicts an exemplary embodiment of an optical network
to which various embodiments of a method according to the present
invention are applicable.
[0011] FIG. 2 depicts another exemplary embodiment of an optical
network to which various embodiments of a method according to the
present invention are applicable.
[0012] FIG. 3 depicts an exemplary embodiment of a method according
to one aspect of the present invention.
[0013] FIG. 4 depicts an exemplary embodiment of a packet structure
for use in a variable modulation signal according to another aspect
of the present invention.
DETAILED DESCRIPTION
[0014] It is worthy to note that any reference herein to "one
embodiment" or "an embodiment" means that a particular feature,
structure, or characteristic described in connection with the
embodiment is included in at least one embodiment of the invention.
The appearances of the phrase "in one embodiment" in various places
in the specification are not necessarily all referring to the same
embodiment.
[0015] The present invention provides, inter alia, a technique for
an Optical Network node (ONN) in an optical data network to
automatically discover its port interconnections without requiring
a signal to be generated with an optical-cross connect. In this
document, ONN and node are used interchangeably. As used herein, an
optical network node is any node having one or more incoming ports
and one or more outgoing ports. Examples of optical network nodes
include: optical cross-connects, optical add-drop multiplexers,
optical terminal multiplexers, etc.
[0016] Moreover, herein the terms port and channel are used
interchangeably. However, a channel is more accurately two ports
coupled together by a communication medium. Thus, a port is either
the input or the output of a channel, although, herein the terms
are often used interchangeably.
[0017] Turning to FIG. 1, if one needs to set up an optical channel
connection (e.g., a light path) between node A 101 and node B 102,
the only two possible ports on node A 101 are port 3 and port 6. Of
these two ports, port 3 is unavailable (see the ONN A--Port
Adjacency Table below).
TABLE-US-00001 Port Type Port Available Oc48/92/na 0 = No, Adjacent
Node Local Port Remote Port 1/2/8 1 = Yes B 3 2 8 0 B 6 11 8 1 C 8
7 8 0 None 4 None 1 1
[0018] Hence, port 6 is the only port of choice. Similar to node A
101, node B 102 also maintains its own Port Adjacency Table, hence
node B knows port A6 is connected to port B 11.
[0019] The Port Adjacency Table on each node may be manually
configured during start-up or initialization. At this time,
probably all the ports would be marked as "Available". When a new
optical channel connection needs to be setup, e.g., between port A4
and port D7 through node B, then the problem is one of determining
the right ports on each of node A 101, node B 102 and node D
104
[0020] For this example, assume there is already another traffic
flowing in channel B6-D3; hence, channel B6-D3 is unavailable.
Consequently, the only possible connection path between node B 102
and node D 104 is channel B4-D5. As a result, the full connection
path between port A4 and port D7 is: path A4-A6-B11-B4-D5-D7.
[0021] Note that in the above case only the source and destination
information (A4 and D7) and the intermediate node B 102 are known.
All other intermediate ports (A6 . . . D5) need to be discovered
automatically.
[0022] To set up a connection between ports A4 and D7 through node
B 102, ONN A 101 sends a connection request to node D 104 along
node B 102. For example, the request can be a message such as the
following:
Setup a Connection Between Ports A4 and D7 through Node B 102.
[0023] The connection request flows along the control channel 105
only. Control channel 105 is terminated on each node (101-104) and
the request/other information in the control channel 105 is
interpreted by each node (101-104). Hence, the selection of the
ports between node B 102 and node D 104 will be done by node B 102
and node A 101 is unaware of this selection. After node B 102
selects the ports between node B 102 and node D 104, it can signal
the information back to node A over the control channel 105.
[0024] Note that all requests for connection (and tear down) flow
ONLY over the control channel 105. The other port connections are
called optical channel connections (also known as data connections
or bearer channels) and these are purely light paths traveling all
the way from the source CIC card 106 (via port A4) to destination
CIC card 107 (via port D7). It is through the Customer Interface
Cards (CICs) present in an optical node that a customer sends and
receives his/her data traffic over the network. Using only the
control channel 105, one can keep track of this optical channel
connection.
[0025] Based on this connection path, optical cross-connects are
made on node A 101 (A4-A6), node B 102 (B11-B4) and node D 104
(D5-D7), thereby forming an end-to-end optical channel connection.
After this, the Port Adjacency Tables in node A 101, node B 102 and
node D 104 are updated and ports A4, A6, B11, B4, D5 and D7 are
marked as unavailable (in other words, a channel exists from
A4-A6-B11-B4-D5-D7). When this connection (A4-A6-B11-B4-D5-D7) is
torn down, the above set of ports are remarked as available for new
connections.
[0026] Manual mapping is cumbersome--even for a small network--to
accurately compute and populate the information to all the nodes.
This manual process may be acceptable for a 16 node, 16-channel
(port) network. However, scaling this to a large network with as
few as 64 channels becomes cumbersome. If mistakes are made in the
map table or in the physical connections during this manual
process, diagnosing the errors is rather difficult.
High Level View
[0027] According to one aspect of the present invention, the CIC
card is enabled to send various special signals (e.g., optical
patterns) over the optical channels, but not the control channel.
The special signals are termed Variable Modulation (VM). VM is
encoded in such a way as to represent a node name and port number
(e.g. A6), which information is termed a Label, and which Label is
automatically generated. The ONNs have optical monitors on each of
its ports, known as Port Monitors (PM). The PM detects the VM and
the ONN deciphers the Label. When a node receives a Label from its
neighbor, (the Label has the neighbor's node name and the port over
which it was sent) it correlates the Label with its own node name
and the port on which it was received using the PM. Thus, the node
discovers to what neighboring node and port (on that neighbor) to
which that this particular receiving port is connected. The node
then transfers this Port Adjacency information to that neighbor,
and the master node. Thus, two neighboring nodes maintain identical
Port Interconnection details.
[0028] The mechanisms of generating the right Label, directing it
to the right destination, coordinating orderly discovery of the
Port Interconnections between neighbors over an entire network are
described in subsequent paragraphs.
Variable Modulation
[0029] Variable Modulation is a low frequency signal, compared to
the high bit-rate of the optical channels, which low frequency
signal can be generated by a CIC card. As the name modulation
indicates, VM is a user definable signal or pattern. A CIC card can
generate VM only when it is not generating its normal payload
traffic, i.e., when it is idle. In other words, the system does not
mix VM with real traffic.
[0030] FIG. 4 depicts the packet format of the VM signal. One
possible embodiment of the VM signal is a 4.times.32-bit packet.
The first 2.times.32 bits consist of framing to enable the
recipient to determine the start of a valid VM packet. The next 32
bits comprise the label, e.g., node ID and port ID. A 16-bit cyclic
redundancy code is appended to the end to enable error detection.
In this embodiment, a remaining 16 bits is unused.
[0031] Port (ONN) Modes
[0032] The ports on the ONN can be put in OPEN and TERMINATE modes.
In the OPEN mode, the ports allow light to pass through them to
another port--this can be its connected port on its neighbor or can
be another port on the same node to which it is cross-connected. In
the TERMINATE mode, as the name indicates, no light is allowed to
pass through.
Port Interconnection Discovery Mechanism
[0033] Turning to FIG. 2, one (or more) ONN that has at least one
CIC card is configured as the Master. The information as to which
node is the Master is configured in all the ONNs. All the ports in
the Master are in the OPEN mode. All other ONN ports are in
TERMINATE mode.
[0034] Master A has an CIC card attached to port 1 and uses this
CIC card to launch VM signals to the network. Master A launches VM
signals through each of its other ports successively (2, 3 and
4).
[0035] First Master A launches the VM signal through port 2 (i.e.,
Master A does a cross-connect in A, A1-A2). The Label (i.e., VM) is
A2 (this encoding is automatically done by software). The signal A2
is received by the neighboring ONN B on port 1, which is detected
by the port monitor on B. Hence, node B knows that B1 is connected
to A2. Node B then informs this A2-B1 interconnection information
to node A, through the control channel, not shown. Next, node B
puts both its ports 1 and 3 in the OPEN mode and cross-connects 1
and 3. Next, a new signal is launched by A1 and it goes to C1
through A1-A2-B1-B3. The Label in this case is encoded as B3. Node
C on receiving this on port 1 knows that B3-C1 is interconnected
and informs this to B.
[0036] Master node A then sends a signal from A1 to A3 and thereon
to B2 for identifying the A3-B2 interconnection. In this case, the
Label sent is A3. Then Node A probes A4-D1 and so on. The Label
information is software configurable and it is always a destination
node's predecessor node ID and port number.
[0037] Reliable Discovery Algorithm
[0038] In a large network, it is essential to send the VM signals
in an orderly manner over one node after another and one port and
another for complete discovery of all the port interconnections.
All the nodes periodically inform the Master as to which of their
ports are still to be discovered, thereby prompting the Master to
send VM signals in their direction.
[0039] Turning to FIG. 3, shown therein is a method for
automatically discovering the port interconnections in a network
whose nodes and port connections are totally unknown. As part of
the network initialization, each node is designated either a master
node or a non-master (sometimes known as slave) mode. Each network,
or network neighborhood, has a designated master node. The
remaining nodes are non-master or slave nodes. Each node knows
which node in the network or network neighborhood is the master
node and what type of node it is designated. Typically, the master
nodes have the CIC cards with the ability to generate the
above-described variable modulation signal. Moreover, each node has
a communication channel to the master node (or its neighboring
node) via the control channel.
[0040] When a network is first initialized, the nodes do not know
the ports of other nodes to which their ports are connected. Upon
startup, each node sends a connection request to the master node
indicating the ports in it and requesting information as to which
other nodes' ports its ports are connected. Upon receipt of all of
these requests, the master node creates a port adjacency table for
the network. This table identifies all of the ports in the network
for each node, and includes a field for the node/port to which each
port is connected. As information is collected regarding these
connections, the master node updates this table.
[0041] Referring to FIG. 3, the process begins in step 31. First,
all ports in all non-master nodes are placed in the TERMINATE state
(step 32). This ensures that signals will only reach the nodes
neighboring the master node (also referred to as the first level
nodes).
[0042] The master node discovers all of its port connections by
successively transmitting the variable modulation signal to all of
its ports (step 33). The master node transmits successively over
each of its nodes a predetermined signal, e.g., the variable
modulation signal discussed above. The predetermined signal has
encoded thereon the node and port from which the signal originates.
The signal is then transmitted from this port to the connection at
the other end of the port.
[0043] Upon receipt of this predetermined signal, the recipient
node decodes the encoded information and replies with the port on
which the predetermined signal was received to the master node via
the control channel (step 34).
[0044] Now the master node knows which port on the recipient node
is connected to the port on which it transmitted the predetermined
signal. The master node then updates the port adjacency table (step
35). Each recipient node also updates its port adjacency table
(step 36).
[0045] The above process continues until all of the ports in the
master node's port adjacency table are identified. This is tested
throughout the process (step 37). Those ports in those nodes whose
port connections are completely known are then placed in the OPEN
state, which allows them to pass through VM signals (step 38). The
master then successively sends VM signals to the next level nodes
whose port connections remain unknown (step 38). The VM signals in
these cases are labels with the node/port identification of the
ports in the next level whose connections remain unknown. The
process returns to step 34 and repeats, whereby the port
connections of the next level from the master node should be
completely discovered. Again, it is tested whether there remain any
ports whose connections are not known. The above process continues
to the next level until all port connections are known.
[0046] Alternatively, the master node could select a port from the
port adjacency table that remains unidentified as to its
connection. The master node then addresses the predetermined signal
to that port. For example, if node B has a port 3, the connection
to which remains undiscovered, the master node transmits the
predetermined signal to node B to be transmitted from node B port
3. The recipient of the predetermined signal then informs its
predecessor node and the master node via the control channel as to
the port on which the predetermined signal encoded with "B3" was
received. The predecessor node and the master nodes then update the
port adjacency table with this information. This process continues
until all undiscovered ports are discovered.
[0047] After the initial configuration of a network if a new node
comes on-line, then this new node is configured as a Secondary
Master. Unlike the Master, the role of a Secondary Master is
limited to discovering port inter-connections to its neighbors
only. A Secondary Master cannot discover remote nodes port
interconnections. The discovery process of the Secondary Master
itself is identical to the Master except that it stops after
discovering its immediately connected neighbor ports. It should be
noted that there might be several Secondary Master nodes in a
network, as each newly joined node is configured as a Secondary
Master.
[0048] If any ports in any nodes remain undiscovered, the node
whose port connections remain unknown sends a connection request to
the master node and the port connection is thus discovered, as
described above.
[0049] Although various embodiments are specifically illustrated
and described herein, it will be appreciated that modifications and
variations of the invention are covered by the above teachings and
within the purview of the appended claims without departing from
the spirit and intended scope of the invention. For example, while
several of the embodiments depict the use of specific data formats
and protocols, any formats and protocols will suffice. Moreover,
while some of the embodiments describe specific embodiments of
ONNs, others apply. Furthermore, these examples should not be
interpreted to limit the modifications and variations of the
invention covered by the claims but are merely illustrative of
possible variations.
* * * * *