Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
20~55~8
COMMUNICATIONS SYSTEM WITH A SINGLE PROTECTION LOOP
This invention relates to optical communications
systems, and is particularly concerned with providing a
protection channel for so-called protection switching.
Backqround of the Invention
It is known to provide a plurality of
communications channels, for example optical fiber
transmission channels on which digital signals are
transmitted in time division multiplexed frames, between
different locations. In order to maintain transmission in
the event of a fault on one of the channels, it is also
known to provide a so-called protection channel via which
the traffic of a faulty channel is transmitted. The
routing of traffic from a faulty channel onto the
protection channel is referred to as protection switching.
The SONET (Synchronous Optical Network) standard
(ANSI Tl.105 and Bellcore TA-TSY-000253) defines a physical
interface, optical line rates known as Optical Carrier (OC)
signals, a frame format, and an OAM&P protocol. The OC
signals have electrical equivalents call Synchronous
Transport Signals (STS). A base rate of 51.84 Mbit/s (OC-
1/STS-1) is defined, with higher rates being integer
multiples of the base rate. SONET also defines
requirements for protection switching, a hierarchy of
failure conditions to be used and a signalling scheme.
The signalling uses Automatic Protection Switching (APS)
bytes within the line overhead (LOH) of the SONET frame.
It is desirable to provide an optical
communications system that takes advantage of the inherent
efficiency of the SONET standard with respect to the use of
automatic protection switching (APS) bytes defined therein
for APS signalling. It is also desirable to provide a
protection switching arrangement that reduces the
duplication of hardware required for implementation.
206~558
Summary of the Invention
An object of the present invention is to provide
an improved communications system with a single protection
loop.
In accordance with one aspect of the present
invention there is provided in a communications system
comprising SONET communications channels extending between
first and second locations, the channels comprising a
plurality of forward and reverse channels for carrying
o traffic in normal operation between first and second
locations and a protection channel for carrying traffic of
one channel in the event of a fault, each pair of the
plurality of forward and reverse channels is provided on a
shelf at each location, the shelves having the same
relative position at both locations and the protection
channel is provided on a protection shelf, the protection
channel including, between the first and second locations,
a forward optical link and a reverse optical link and at
each of the first and second locations, a single protection
loop, coupling the forward and reverse optical links, that
forms the protection channel, a method comprising the steps
of providing, at each location, a local virtual protection
loop for indicating the protection requirements and status
of the respective location, and providing, at each
location, a remote virtual protection loop for indicating
the protection requirements and status of the location
remote from the respective location.
In accordance with another aspect of the present
invention there is provided a communications system
comprising SONET communications channels extending between
first and second locations, the channels comprising a
plurality of forward and reverse channels for carrying
traffic in normal operation between first and second
locations and a protection channel for carrying traffic of
one channel in the event of a fault on said one channel,
the protection channel including, between the first and
second locations, a forward optical link and a reverse
2065558
_ 3
optical link and at each of the first and second locations,
a single protection loop, coupling the forward and reverse
optical links, that forms the protection channel.
An advantage of the present invention is by using
a single protection loop, a reduction in equipment is
achieved over either dual loop or bus arrangements. A
further advantage is achieved by using standard formats for
the signalling used to implement virtual protection loops,
in that the same hardware is used for both the protection
0 loop and the normal communications channels. This reduces
the number of different circuit cards required thus
allowing reductions in spacing. An advantage in using
virtual protection loops is that separate circuitry between
shelves is not required to provide the switching status and
request information.
Brief Description of the Drawinqs
The present invention will be further understood
from the following description with reference to the
drawings in which:
Fig. 1 illustrates a known SONET frame format;
Fig. 2 illustrates a block diagram of a
communications system in accordance with an embodiment of
the present invention;
Fig. 3 illustrates detail of the protection and
working shelves of the system of Fig. 2;
Figs. 4a, 4b, and 4c, illustrates the protection
loops in accordance with the present invention;
Fig. 5 illustrates the protection loops of Fig. 4
with the system in an idle state;
Fig. 6 illustrates the protection loops of Fig. 4
with the system in a manual switched state; and
Fig. 7 illustrates the protection loops of Fig. 4
with the system responding to a request to be protection
switched.
Similar references are used in different figures
to denote similar components.
~;, toS5S~
-
By the nature of the SONET standard STS and OC
signals are interchangeable. Thus, reference herein to one
or the other are in respect to the specific embodiment
described and it is to be understood that one skilled in
the art would recognize that those elements designated as
OC could as easily be STS and vice versa.
Detailed Descri~tion
Referring to Fig. 1, there is illustrated the frame
format of the SONET standard. The frame 10 is defined as
90 bytes by 9 bytes and includes a transport overhead (TOH)
12, of 3 bytes by 9 bytes and a payload 14, of 87 bytes by
9 bytes. The TOH 12 iS further subdivided into a section
overhead (SOH) 16, of 3 bytes by 3 bytes and a line
overhead (LOH) 18, of 3 bytes by 6 bytes. The first column
of nine bytes in the payload 14 forms the path overhead
(POH) 20. Each overhead byte has an assigned purpose and
label as shown in the expanded view in Fig. 1. For example
the first and second bytes are for framing and are labelled
Al and A2. The fifth and sixth bytes of the LOH are the
APS bytes Kl and K2 of STS-l #1. The Kl byte is used to
indicate requests. The K2 byte is used primarily to
indicate status.
The defined values of the Kl Byte are given in the
SONET standard. Bits 1-3 are used to indicate the
protection condition. Bit 4 is used for priority for
automatic requests only. Bits 5-8 contain the channel
number of the channel asserting the protection condition.
If channel number 0 is selected then the condition
bits show the protection channel status. If channel 15 is
selected then the condition bits show the priority assigned
to extra traffic.
Reverse request is used in a bi-directional
switching mode to request the switch in the reverse
direction. Reverse request is implicitly prioritized at
the level of the request it is acknowledging.
The defined values of the K2 Byte are given in the
SONET standard. Bits 1-4 are used to indicate the channel
~o ~5S6
number of the traffic carried on the P-Channel.. Bit 5 is
used to indicate a l:N or 1+1 protection scheme.. Bits 6-8
have several defined and yet to be defined uses..
States other than '111' and '110' for bits 6-8 are
not defined in the current version of TA-TSY-000253. The
two states listed as Request Blocked and Protection State
Change are anticipated for use in nested span switching.
All other states are reserved for future use.
Thus, it is well known to provide communications
0 systems with protection loops. For example Reid et al., in
U.S. Patent No. 4,646,286, issued Feb. 24, 1987 to Northern
Telecom Limited, teach separate protection paths, at each
location, for the transmit and receive paths.
Referring to Fig. 2, there is illustrated, in a
functional block diagram, a site level view of a l:N
terminal configuration for an optical communications system
in accordance with an embodiment of the present invention.
Each terminal includes a protection shelf 30. Each
protection shelf 30 includes transmit and receive units, 32
and 34, respectively, coupled to optical fibers 36 and 38,
respectively. Each protection shelf 30 also includes a
redundant set of transmit and receive units, 40 and 42,
respectively. Each terminal includes N working shelves 60
(represented by a single shelf in Fig. 2) . Each working
shelf 60 includes transmit and receive units 62 and 64,
respectively, coupled to optical fibers 66 and 68,
respectively. Each working shelf 60 also includes a
redundant set of transmit and receive units 70 and 72,
respectively. For simplicity, connection of lïnes from
demultiplex units and to transmit units are not shown in
Fig. 2.
On working shelves 60, an A set of units,
consisting of the transmit A unit 62, the receive A unit
64, and a DEMUX A unit 74, iS provided to support optical
communication on an optical channel via optical fibers 66
and 68. A second set of units, the B set, consisting of
the transmit B unit 70, the receive B unit 72 , and a DEMUX
sss~
B unit 76, is provided to support a protection loop (P-
Loop) 80.
The protection shelf 30, also has A and B sets of
units. The A set of units consist of the transmit A unit
40, the receive A unit 34, and a DEMUX A unit 44. The
receive unit 34 is coupled to the optical fiber 38. The
DEMUX A unit 44 couples the receive unit A 34 to the
transmit A unit 40. The transmit A unit 40 is coupled to
the P-Loop 80. The B set of units consist of the transmit
B unit 32, the receive B unit 42, and a DEMUX B unit 46.
The receive B unit 42 is coupled to the DEMUX B unit 46.
The DEMUX B unit 46 couples the receive B unit 42 to the
transmit B unit 32. The transmit B unit 32 is coupled to
the optical fiber 36.
One embodiment of the P-loop 80 is a high-speed
(2.488 Gbit/s) coaxial loop that is an extension of the
P-channel optical line. The P-loop 80 originates on the
protection shelf 30 and is routed down to the last working
channel shelf 60 of the terminal. From there it propagates
through each working shelf 60 via the B set of units back
to the protection shelf.
In operation, each working shelf 60 can either
accept traffic from the P-loop 80 and bridge its own
traffic onto the P-loop 80 or pass-through the existing
traffic on the P-loop 80.
Typically, in the event of a fault affecting it, a
working shelf 60, at the head end of a span, bridges its
traffic onto the P-loop 80. Each other working shelf 60 on
the P-loop 80 between the bridged shelf and the protection
shelf 30 passes the traffic through its B set of units
72,76, and 70. The protection shelf 30 passes this traffic
out on the P-channel. At the tail end of the span, the
traffic is routed via the P-loop down from the protection
shelf to the terminal's last working channel shelf and then
passed back up through each working channel shelf until
reaching the working channel shelf that asserted the
request.
p~
4~--
20~5~8
-
Referring to Fig. 3, there is illustrated a more
detailed functional block diagram of the protection and
working shelves 30 and 60 of Fig. 2. A block 48, shown in
broken line, labelled TRIBS (tributaries) represents
optional low priority channels that may be coupled to the
protection shelf 30. The TRIBS 48 provide data to the
transmit B unit 32 via a bus 50 and receive data from the
DEMUX A unit 44 via the bus 52. When the protection
channel is needed for protection switching of one of the
o working shelves 60, the low priority data from the TRIBS 48
is dropped. The implementation of the optional tributaries
is similar to that for the working shelf 60, which is
described in detail below
For extra traffic, the payload must be terminated
on DEMUX A unit 44 of a protection shelf 30. This ensures
that extra traffic can be supported during a P-loop failure
or during site expansion (i.e. adding a new working channel
shelf).
A block 54, labelled TRIBS (tributaries)
represents channels coupled to the working shelf 60. The
DEMUX units 74 and 76 provide received clocks, frame and
data to the TRIBS 54, via busses 56 and 58, respectively.
A line 78, labelled RX USEB provides an enable signal that
informs the TRIBS 54 which of the receive lines to use.
Similarly, the transmit units 62 and 70 provide transmit
clocks and frame to the TRIBS 54 via lines 82 and 84,
respectively. A line labelled TX USEB 86 provides an
enable signal that informs the TRIBS 54 which of the
transmit clocks and frame to use. The TRIBS 54 is coupled
to the transmit units 62 and 70 via a bus 88. The transmit
unit 62 includes a MUX 90 coupled to the bus 88 for
multiplexing the data thereon for transmission via the
optical fiber 66. The transmit unit 70 includes a MUX 92
coupled to the bus 88 and a switch 94 having inputs coupled
to the P-loop 80 and to the MUX 92 and an output coupled to
the P-loop 80.
206~58
-
In operation, the TRIBS 54 transmit and receive
data via the optical channel provided by the transmit unit
62 and the optical fiber 66, and the optical fiber 68, the
receive unit 64, and the DEMUX A unit 74. Data to be
transmitted is provided to both transmit units 62 and 70.
However, the switch 94 in the transmit unit 70 is switched
to connect the P-loop 80 at its input to the P-loop 80 at
its output, thereby providing a pass-through for the data
on the P-loop 80. Thus, data from the TRIBS 54 to the
0 transmit unit 70 is blocked. In this case, the enable
signal on the TX USEs line 86 in not provided, that is a do
not use transmit unit B condition exists.
on the receive side, data received via the optical
fiber 68 by the receive unit 64 is demultiplexed by the
DEMUX A unit 74 and provided to the TRIBS 54 via the bus
56. In this case, the enable signal on the RX USEB line 78
is not provided. Thereby, the TRIBS 54 are informed to
accept the data on the bus 56. Thus, the B set of units
merely form part of the P-loop by providing a path for data
to pass through the shelf.
When a fault occurs on the working channel, the
data from the TRIsS 54 is switched to the protection
channel. How such switching is accomplished is the subject
of U. S. Patent No. 4,646,286 mentioned above and the
Bellcore standard TA-TSY-000253 and as such is not repeated
here. Once switched, the TRIBS 54 use the B set of units
for both transmitting and receiving data. When the
protection channel is used, the switch 94 in the transmit B
unit 70 is coupled to the MUX 92 thereby providing the data
from the TRIss 54 to the P-loop 80. In this case, the TX
USEB line 86 provides an enable signal. The receive B unit
72 receives data from the P-loop 80 and supplies the data
to the DEMUX B unit 76. The DEMUX B unit 46 has two output
ports, one (not demultiplexed) for passing the P-loop data
on to the transmit B unit, the other for providing
demultiplexed data, together with receive clocks and frame
to the TRIBS 54 via the bus 58. In this case, the RX USEB
~o ~5~58
line carries an enable signal. Thus, the TRIBS 54 accept
the receive clocks, frame and data from the DEMUX B unit
76.
Referring to Fig. 4a, 4b, and 4c there is
illustrated the P-loop in its physical, virtual (local and
remote) forms. The single physical loop 100 of Fig. 4a
carries the APS bytes K1 and K2 of STS-1 #1 used for APS
signalling in accordance with the SONET standard.. Within
the terminal site, two virtual loops are established as
represented by Figs. 4b and 4c. A local virtual loop 110,
Fig. 4b, is established by providing local K1 and K2 bytes
in the El slot of STS-1 #25 and #2, respectively. A remote
virtual loop 120, Fig. 4c, is established by providing
remote K1 and K2 bytes in the El slot of STS-1 #26 and #3,
respectively.
In operation, the APS bytes (K1, K2) support all
the necessary inter-site and inter-shelf protection
switching control. The P-loop consists of a single
physical loop 100 that supports two virtual APS byte loops
within it. A local APS byte loop 110 carries the K1 and K2
bytes generated at the local site. The K1 byte uses the E1
(LOW) slot in the SOH of STS-1 #25. The K2 byte uses the
E1 slot in the SOH of STS-1 #2. A remote APS byte loop 120
carries the K1 and K2 bytes received from the remote site.
The K1 byte uses the E1 slot in the SOH of STS-1 #26. The
K2 byte uses the El slot in the SOH of STS-1 #3. The K-
bytes in the local and remote APS byte virtual loops never
leave the P-loop of the terminal site. The El slots are
always overwritten by the transport overhead multiplexer on
the P-channel shelf OCI-48TX B unit. Thus, no site outside
the P-loop will receive these virtual bytes.
By using slots in the SOH, a shelf can assert a
request (e.g. place a local K1 byte onto the P-loop)
without altering the line BIP-8 of an STS-1, thus avoiding
a Line BIP-8 re-calculation. The section BIP-8 is always
calculated for the outgoing STS-48.
206555~
These virtual loops ensure that each shelf is
given a complete set of information (Local and Remote K
bytes) upon which to base its protection decisions. In
order to differentiate the K bytes in the virtual APS byte
loops from the true K bytes of STS-1 #1, the true K-bytes
are referred to hereinbelow as the real K bytes
Referring to Figs. 5 and 6 examples are provided
of the inter-shelf routing of the APS bytes. Fig. 5
illustrates a terminal-terminal 1:3 system under idle
conditions (channel P is assumed to be idly bridged) and
Fig. 6 illustrates a bi-directional Manual switch in
effect. Both the local and remote APS byte loops are shown
a each terminal site.
The inter-shelf routing rules are as follows:
a) Remote APS byte loop, The P-channel shelf
copies the incoming K1 and K2 bytes from slots within STS-1
#1 to the E1 slot in STS-1 #26 and the E1 slot in STS-1 #
3, respectively. Recall that the incoming line is always
passed through to the P-loop, thus all shelves are capable
of extracting the received K-bytes from the E1 slots on the
P-loop. A bridged working channel must relay the Remote K-
bytes across the bridge. A working channel in pass-through
must pass the remote K1, K2 bytes.
b) Local APS byte loop. A bridged working
channel must set the local K2 byte, bits 1-4 equal to its
own channel number, as it is now the source of the traffic
on the P-loop (only when actively bridged). Relay the
local K1 byte value across the bridge if the priority of
that request is higher than its own or replace the local K1
byte with its own request if the reverse is true. Copy
both local K-bytes into real K-bytes timeslots.
Note that this also applies to a bridged P-channel for the
Demux B to OCI-48TX s path (i.e. outgoing optical line).
A working channel in pass-through must relay the
local K2 byte, however, it must be capable of replacing the
current Local K1 byte with its own, if its request is
higher in priority. Otherwise, the current local K1 byte
2065~58
11
is also passed through. This also applies to a P-channel
in pass-through for the Demux B to OCI-48TX s path.
The P-channel completes the local APS bytes loop
by relaying the local K bytes back onto the P-loop.
When a shelf is in pass-through the received SOH and LOH
are passed through untouched (i.e. transparent pass-through
mode). When a shelf wishes to insert a request (i.e. K1
byte in the local loop), it must change to the non-
transparent pass-through mode (NTP). In the NTP mode, the
10 LOH iS still passed through untouched but the virtual K
bytes in the SOH can be overwritten.
Referring to Fig. 7, there is illustrated the APS
byte signalling sequence in the event of an SF switch
request at site C, Channel 1 with an existing Manual Switch
in effect on Channel 4.
At site C, the switching span's tail end, the SF
condition is detected by demux A and subsequently
communicated to the microcontroller on demux s. The
microcontroller on demux B determines that this SF is the
highest priority switch request and consequently, initiates
the following protection switching sequence.
Step 1: Channel 1 asserts the SF request in the
outgoing K1 byte (local APS byte loop), however the K2 byte
still shows a CHID of 4 because Channel 4 is the channel
currently bridging traffic onto the P-loop at site C;
Steps 2-5, site C: the SF request propagates
around the local APS byte loop so that all shelves, above
and below the requesting shelf are aware of the condition;
Steps 3-7, site A: the request propagates up
through the span~s head end via the remote APS byte loop;
Step 8: the span's head end detects the request
and consequently bridges its traffic onto the P-loop.
Assertion of the TX uSEs control is held off (tentatively 3
ms) to allow the B OCI-48TX unit to phase lock to the OCI-
TX A unit. At this time, site A, Channel 1 also asserts aReverse Request since it is provisioned for bi-directional
switching.
206~58
12
Step 9: the outgoing real K2 byte from site A
contains a CHID of 1;
Note that via the remote APS byte loop, all
shelves at site A, above and below the bridging channel are
aware of the far end's request.
Step 10: Channel 4 at site A detects a CHIDMM and
thus releases its switch (Channel 4 being aware of the SF
request does not assert a Switch Fail condition). It also
determines that the local SF request has been honoured
o since it detects a CHID of 1 in the remote K2 byte.
Consequently, this channel must release its bridged state
after Channel 1 has bridged, that is when CHID=l appears in
the local K2 byte, to allow the traffic to reach the
requesting channel on the P-loop.
Steps 11-13: the reverse request and remote CHID
reaches the span's tail end.
Step 14: at site C, Channel 1 must bridge in
response to the Reverse Request and also in response to the
correct CHID in the Remote K2 byte indicating that the
request has been honoured. The bridging by Channel 1 is
required to cause Channel 4 at site C to release its bridge
thereby allowing the incoming traffic to reach Channel 1
via the P-loop.
Step 15: the mechanism used to cause Channel 4 to
release its bridge is the changing of the local K2 byte by
Channel 1.
Step 16: the new local K2 byte appears at the
bottom channel at both sites A and C.
Step 17: Channel 4 at site C releases its switch
reverting to the pass-through mode.
Steps 18-20: the new K2 byte, Local at site C and
remote at site A, propagates up through the respective
virtual loops. When both the remote and local K2 bytes
carry the correct CHID, the protection switch is complete.
This occurs a step 19 at site A and step 20 at site C.
Numerous modifications, variations and adaptations
may be made to the particular embodiments of the invention
2065558
13
described above without departing from the scope of the
invention, which is defined in the claims.
