Note: Descriptions are shown in the official language in which they were submitted.
~2375~2
COMMUNICATIONS SYSTEM WITH PROTECTION SWITCHING
AND CHANNEL IDENTITIES
This invention relates to communications sytems, and is
particularly concerned with the control of so-called protection
switching in a communications system.
Reference is directed to our copending patent application
filed simultaneously herewith and entitled "Communications System
with Protection Switching using Individual Selectors" which is
directed to other features of the system described herein.
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-cal1ed 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.
Several problems arise in achieving effective protection
switching in such a system. Firstly, it is necessary for the
protection switching to take place rapidly in the event of a fault,
so that information is not unnecessarily lost. At the same time, it
is desirable for the various channels to be able to have different
priorities, so that in the event of faults occurring on more than one
channel the protection channel is awarded to the highest priority
faulty channel. The resolution of priorities generally involves
software procedures implemented by processors at the respective
locations, which processors must communicate with one another. This
communication involves delays which conflict with the requirement for
rapid switching.
Secondly, the system may include an arbitrary number of
intermediate sites between end locations, individual channels being
optionally repeatered or dropped and inserted at each intermediate
site. The protection switching should desirably accommodate all
possible situations in an efficient manner, the protection channel
being used only insofar as it is required For routing traffic around
faulty parts of normal channels. In this manner, a single protection
1~37~;02
channel can be used to protect traffic from faults on a plurality of
normal channels in different parts of the system.
The protection channel may be used for the transmission of
additional information in the event that protection switching is not
currently required. In any event, it is also necessary for each site
at which traffic on the protection channel may be received to be able
to route this traffic to the intended destination.
An object of this invention is to provide an improved method
of controlling routing of traffic to the protection channel in a
communications system.
According to one aspect of this invention there is provided,
in a communications system comprising communications channels
extending between first and second locations, the channels comprising
at least one forward channel for carrying traffic in normal operation
from the first location to the second location, a protection channel
for carrying the traffic of a forward channel in the event of a fault
on the forward channel, and at least one reverse channel for carrying
signals from the second location to the first location, a method of
controlling routing of traffic to the protection channel comprising
the steps of: detecting at the second location a fault on a forward
channel; transmitting an indication of the fault via a reverse
channel from the second location to the first location; in response
to the fault indication, supplying traffic of the forward channel
having the fault to the protection channel, the traffic supplied to
the protection channel including an identity of the forward channel
having the fault; and at the second location, replacing the traffic
of the forward channel having the fault with the traffic supplied via
the protection channel in dependence upon the channel identity in the
traffic on the protection channel.
The transmission of the fault indication on the reverse
channel, coupled with the insertion of the channel identity in the
traffic consequently routed via the protection channel, and
recognition of this at the second location whereby this traffic can
be correctly directed to its intended destination, enable protection
switching to be accomplished in a particularly rapid and expedient
manner.
Preferably there is a plurality of forward channels and a
lX3750Z
corresponding number of reverse channels, and the indication of a
fault on one of the forward channels is transmitted from the second
location to the first location on a corresponding one of the reverse
channels. Preferably the forward and reverse channels constitute
respectively reverse and forward channels for the opposite direction
of transmission, from the second location to the first location, a
reverse protection channel being provided for protecting traffic
carried in said opposite direction of transmission. Thus a fully
bidirectional and well protected communications system can be readily
provided using the method of the invention.
In order to avoid delays in protection switching due to
resolution of priorities for the different channels, preferably the
step of supplying traffic of a forward channel having a fault to the
protection channel comprises controlling a selector of the forward
channel having the fault to supply the traffic of the forward channel
having the fault to the protection channel, and inhibiting a selector
of each other channel from supplying traffic to the protection
channel. Thus protection switching takes place rapidly under
hardware control. To resolve priorities where the channels have
respective priorities, the method preferably includes the step
of enabling, following the inhibition of the selector of each other
channel, the selector of each channel having a higher priority than
that of the channel having the fault, to supply traffic to the
protection channel in response to a fault indication indicating a
fault on the higher priority channel. Thus after a rapid protection
switch, software controls can be applied to prepare the selectors of
the channels for a subsequent protection switch in dependence upon
the priority scheme which is in effect.
The traffic on each channel is conveniently transmitted in a
time division multiplexed frame structure including bits for the
channel identity and the indication of a fault, the channel identity
being part of the traffic on each forward channel.
The method preferably also includes the steps of:
transmitting information via the protection channel when traffic from
a forward channel is not being supplied thereto; supplying traffic of
a forward channel having a fault to the protection channel with an
indication that the traffic is protection traffic; and, at the second
4 lZ3750~
location: detecting said indication in the traffic on the protection
channel to produce a repeater control signal; producing an inhibit
signal in response to detection of a fault on a forward channel; and
repeating the traffic on the protection channel for transmission from
the second location to a third location in response to the repeater
control signal in the absence of the inhibit signal. In this manner
additional traffic can be transmitted via the protection channel, the
indication on this channel also being constituted by the state of one
bit in the frame structure. This also permits different spans of the
protection channel to be used efficiently for protecting different
channels simultaneously, in the manner described in detail below.
According to another aspect this invention provides a
communications system comprising: communications channels extending
between first and second locations, the channels comprising at least
one forward channel for carrying traffic in normal operation from the
first location to the second location, a protection channel for
carrying the traffic of a forward channel in the event of a fault on
the forward channel, and at least one reverse channel for carrying
signals from the second location to the first location; means at the
second location for detecting a fault on each forward channel and for
producing a fault indication in response to such detection; means for
transmitting the fault indication via a reverse channel from the
second location to the first location; means at the first location
responsive to the fault indication for supplying traffic of the
forward channel having the fault to the protection channel with an
identity of the forward channel having the fault; and means at the
second location for replacing the traffic of the forward channel
having the fault with the traffic supplied via the protection channel
in dependence upon the channel identity in the traffic on the
protection channel.
The invention will be further understood from the following
description with reference to the accompanying drawings, in which:
Fig. 1 schematically illustrates parts of a first terminal of
an optical fiber communications system incorporating an embodiment of
the invention;
Fig. 2 schematically illustrates parts of a second terminal
of the optical fiber communications system;
~237502
Fig. 3 is a schematic block diagram illustrating parts of a
demultiplexer unit of the system;
Figs. 4 and 5 are state diagrams with reference to which the
operation of control logic units of the system are described; and
Fig. 6 schematically illustrates parts of an intermediate
unit of the optical fiber communications system.
The invention is described in the context of an optical fiber
communications system in which d plurality of optical fiber channels,
each transmitting digital signals at a bit rate of about 570Mb/s,
extend from a first terminal, optionally via intermediate units, to a
second terminal, and back again whereby signals can be transmitted in
both directions. For clarity and simplicity in the description, the
first terminal is referred to as the head end and the second terminal
is referred to as the tail end of the system, the channels providing
transmission from the head end to the tail end are referred to as
forward channels or simply as channels, and the channels providing
transmission from the tail end to the head end are referred to as
reverse channels. The system is fully bidirectional and the same
comments apply equally to both transmission directions. However, for
simplicity the description refers to the reverse channels only where
it is necessary to do so for a complete understanding of this
embodiment of the invention.
Each 570Mb/s optical fiber channel provides for the
transmission of twelve DS-3 bit rate channels (about 45Mb/s each) and
overhead information in a time division multiplexed frame format with
a frame period of 1.21~s and master frames each comprising twelve
frames and having a period of 14.5~s. Individual bits in each master
frame are used for purposes related to protection signalling in the
manner described in detail below.
The optical fiber channels comprise one or more working
channels, each of which carries normal traffic in the manner
described above, and a protection or P channel on which traffic is
transmitted in the event of a failure on one of the working channels,
the routing of traffic onto the P channel being referred to as a
protection switch. The P channel can also be used for transmitting
other information, for example more DS-3 channels of a relatively low
priority, when it is not being used for protection purposes (i.e.
6 1237502
transmission of traffic on a failed channel). The following
description relates to two working channels 1 and 2, in addition to
the protection channel P. Other working channels may be provided in
a similar manner.
The system may include an arbitrary number of intermediate
units between the head end and the tail end of the system. Each such
intermediate unit, which is provided at a respective location or
intermediate site, may provide glass-through (i.e. the optical fibers
are interconnected without any intervening electronics), repeater, or
drop-and-insert (i.e. DS-3 channels may be tapped) capabilities for
the optical fiber channels in arbitrary combinations. The
communications links between the head and tail end and an adjacent
intermediate unit, or between consecutive intermediate units, are
referred to as spans.
In this embodiment of the invention, the P channel can be
used to provide protection for faults on a plurality of working
channels where these faults occur on different spans. For example,
with two working channels 1 and 2 passing on two spans from the head
end via one intermediate unit, providing drop-and-insert capabilities
for each of the working channels, to the tail end, a fault on the
first span of channel 1 can be protected by the first span of the P
channel while a simultaneously existing fault on the second span of
channel 2 can be protected by the second span of the P channel.
The system is initially described with reference to Figs. 1
and 2, illustrating the head end and tail end respectively, which can
be placed side by side with Fig. 1 on the left in order to show
clearly the signal flows therebetween.
Each of the working channels 1 and 2 includes, for the
forward channel from an optical transmitter lD via an optical fiber
30 12 to an optical receiver 14, a first multiplexer (MUX) 16, which is
referred to as the A MUX, at the head end (Fig. 1) for multiplexing
twelve incoming synchronized DS-3 channels and overhead information
in the manner already described above and supplying the multiplexed
signal to the transmitter 10, and a first demultiplexer (DEMUX) 18,
which is referred to as the A DEMUX, at the tail end (Fig. 2) for
demultiplexing the signal from the receiver to produce twelve
outgoing DS-3 channels and overhead information. At the tail end the
123~0~
outgoing DS-3 channels are derived via a selector 20 controlled by a
control logic unit 22 in the manner described below. Figs. 1 and 2
also show, for the reverse channels 1 and 2 on reverse channel
optical fibers 24, the corresponding A MUX 26 and optical transmitter
28 at the tail end and the optical receiver 30 and A DEMUX 32 at the
head end for transmitting traffic in the opposite direction of
transmission.
In order to protect against a failure anywhere on a working
channel between the A MUX 16 and the A DEMUX 18, including the MUX
16 and DEMUX 18 themselves, traffic (the multiplexed DS-3 channels
and the overhead information) can be routed via a second multiplexer
34, which is referred to as the B MUX, at the head end, an optical
transmitter 40, forward channel fiber 42, and optical receiver 44 of
the protection channel P, and a second demultiplexer 46, which is
referred to as the B DEMUX, and the selector 20 at the tail end.
Corresponding protection paths, not shown, exist for the reverse
direction of transmission. In order that the protection channel P
can carry optional extra traffic when it is not being used to carry
traffic from a failed working channel, the channel P also includes an
A MUX 16 (but no B MUX), a selector 36, and a control logic unit 39
at the head end and an A DEMUX 18 (but no B DEMUX) at the tail end.
It will be noted that the output of the receiver 44 of the P
channel is coupled to the input of the A DEMUX 18 of the P channel
and to the B DEMUX 46 of each working channel. In order to enable
the traffic which is being carried on the P channel to be identified,
each channel is allocated a respective channel identity, represented
by ID0 to ID2 for the channels P, 1, and 2 respectively, which is
transmitted as part of the overhead information in the multiplexed
signals. To this end, for each channel the channel identity is
supplied to the respective multiplexers 16 and 34 to be multiplexed
into the overhead information, and to the respective demultiplexers
18 and 46 to be compared with the incoming received and demultiplexed
channel identity. The result of this comparison is supplied via a
respective line 48 to the respective control logic unit 22.
Fig. 3 illustrates the form of each demultiplexer, showing
the comparison of the channel identity which is assumed here to
consist of a 6-bit number. In this respect, it is observed that
8 ~ 237~0~
in this embodiment of the invention it is not essential for the A
DEMUX 18 in each working channel to include such a comparison, but
for convenience all of the demultiplexers have the same form.
Similarly, all of the multiplexers have the same form, even though
it is not essential in this embodiment of the invention for the
channel identity to be inserted in the traffic on each working
channel.
Referring to Fig. 3, each A DEMUX 18 or B DEMUX 46 includes a
frame find circuit 50 and a demultiplexer unit 52 which operate in
generally known manner on the incoming multiplexed signal from the
respective receiver to produce the demultiplexed DS-3 channels, the
6-bit channel identity from the multiplexed signal, and other
(overhead) information. A 6-bit comparator 54 compares the
demultiplexed channel identity with the locally supplied channel
identity ID and produces an output signal dependent on the result of
this comparison, which output signal is gated in an AND gate 56 with
a 'frame found' signal produced by the frame find circuit 50 on a
line 58 to produce the signal on the line 48. Although not shown in
Fig. 3, the demultiplexer also uses the signal on the line 48 to
squelch, or completely suppress, its DS-3 channel and other
information outputs in the absence of frame synchronization or in the
case of a channel identity mismatch.
Before describing the manner in which a protection switch is
effected, the various states which can be adopted by each of the
control logic units 22 and 38, for controlling the selectors 20 and
36 respectively, are described with reference to simplified state
diagrams in Figs. 4 and 5 respectively.
Referring to Figs. 2 and 4, each of the control logic units
22 has five possible states 61 to 65 referred to as READY, ARMED,
S~ITCHING, DONE, and LOCKOUT respectively. Each unit 22 receives
signals on two lines 48 from the respective A DEMUX and B DEMUX;
these signals are referred to as QA and QB respectively, representing
the quality of the signal received by the respective demultiplexer.
Each unit 22 produces three signals UB, PR, and BI. The signal BI is
not shown in Fig. 2 as it is used only at intermediate units, and is
described below with reference to Fig. 6. The signal UB constitutes
a control signal to the respective selector 20; the signal UB=1
9 123~502
constitutes a control command to use the output of the B DEMUX for
the outgoing DS-3 channels. The signal PR is a protection request
signal which is supplied by the unit 22 to the A MUX (and the B MUX)
of the reverse channel for transmission, as one bit in the overhead
information in each master frame, back to the head end.
With reference to Figs. 1 and 5, each of the control logic
units 38 has four possible states 71 to 74 referred to as READY,
BRIDGING, DONE, and OFF respectively. Each unit 38 receives the
protection request signal PR, derived from an output of the
respective A DEMUX 32 of the reverse channel, and a signal UIN from
the next lower (as illustrated in Fig. 1) channel. Each unit 38
produces a bridging signal BR which controls the respective selector
36, and a signal UOUT which is passed on to the next higher (as
illustrated in Fig. 1) channel for which it constitutes the signal
UIN. The signals UIN and UOUT are thus chained through the control
logic units 38, the signal UOUT produced by the highest (in Fig. 1)
working channel, namely channel 1, being supplied as a control signal
to the control logic unit 39 for the P channel selector 36. Each
signal UOUT is also supplied as an input to the B MUX 34 (and
optionally the A MUX 16) of the respective channel, in which
multiplexer it is incorporated as one bit in the overhead information
in each master frame and thus is transmitted on the P channel in the
event of a protection switch, for the purpose described below with
reference to Fig. 6.
Various conditions for transitions between the states shown
in Figs. 4 and 5 are shown in these drawings and are described below.
Some of these transitions are effected solely under software control,
and any of the states can be forced by software control. Such
software control is provided by processors which are not shown in the
drawings but are provided throughout the system, generally one at
each terminal (head end, tail end, or intermediate si-te) for each
channel. The processors co-operate with the control logic units not
only to control their states in accordance with the state diagrams
but also so that they are aware of the state of the system, and in
particular of each protection switch which takes place. The
processors communicate with one another via a communications network
comprising some of the overhead information on the communications
1237~50~
system itself, in a manner which forms the subject of a copending
patent application entitled "Communications Network" to which
reference is directed in this respect. In response to each
protection switch, the processors control the states of the
control logic units as described below so that they are prepared for
making a subsequent protection switch.
The manner in which a protection switch is effected is as
follows. Initially, each control logic unit 38 is in the READY state
71, and each control logic unit 22 is in the READY state 61, with
traffic being routed for each working channel via its A MUX 16,
transmitter 10, receiver 14, and A DEMUX 18. While this traffic is
being correctly received, the A DEMUX 18 produces the signal QA=1 and
(in the absence of a software command forcing a change) the unit 22
remains in the READY state 61, producing the signals UB=O, causing
the selector 20 to select the output of the A DEMUX, and PR=O, which
is fed via the reverse channel to the unit 38 of the respective
channel.
In this normal operating situation, each control logic unit
38 receives, in addition to the signal PR=O, the signal UIN=O from
the unit 38 of the adjacent channel, or from a ground point in the
case of the final working channel, and passes this signal to the unit
38 in the next adjacent channel as the signal UOUT. The signal UOUT
from the unit 38 of the working channel 1 is supplied to the control
logic unit 39 of the protection channel P to control its selector 36,
whereby the P channel transmitter 40 receives traffic from the P
channel A MUX 16 rather than from a protection traffic line 80. The
line 80 passes successively between the selectors 36 of the working
channels, each of which selectors 36 is controlled by the signal BR=O
from the respective unit 38 to link the line 80 through the selector
and thereby isolate the output of the respective B MUX 34.
In the event of a fault, for example on the working channel
2, the A DEMUX 18 detects a loss of frame synchronization or a
mismatched channel identity and produces the signal QA=O. In
consequence, the control logic unit 22 of this channel adopts the
ARMED state 62, in which it produces the protection request signal
PR=1. This signal is transmitted via the reverse channel and
received by the control logic unit 38 of the respective channel.
12375~z
11
Assuming that this unit 38 still receives the signal UIN=0, it adopts
the BRIDGING state 72 in which it produces the signals UOUT=l and
BR=l. The former signal is propagated through the control logic
units 38 of the other working channels, in this case only the channel
1, inhibiting them from adopting the BRIDGING state, to the control
logic unit 39, which thereby controls the selector 36 to couple the
protection traffic line 80 to the transmitter 40 of the protection
channel. At the same time, the si gnal BR=l of the channel 2 being
protected causes its selector 36 to couple the output of the
respective B MUX 34 to the line 80, whereby this channel's traffic is
routed to the protection channel.
At the tail end, the A DEMUX of the channel P detects a
channel identity mismatch and squelches its output as already
described. The traffic on the protection channel P is also supplied
to the B DEMUX 46 of each working channel, which similarly detect
channel identity mismatches and squelch their outputs, except for the
B DEMUX 46 of the working channel 2 which now detects a channel
identity match and produces the signal QB=l. This signal causes the
unit 22 to adopt the SWITCHING state 63, in which it produces
the signal UB=l which controls the selector 20 to switch over to
supply the outputs of the B DEMUX 46 to the outgoing DS-3 channels,
thereby completing a protection path for the traffic on the failed
channel.
It should be noted that the protection path is set up in the
above manner without relying on any software control. In
consequence, the protection switch is accomplished very much more
quickly than would be the case if software control, and hence
communications among the processors, were required. However, the
result of the protection switching as so far described prevents
traffic on another, possibly higher priority, channel from being
protected instead if that channel fails. In order to accommodate
this, software control is now exercised to prepare the protection
arrangement to be able to make another protection switch if this
should be desired. As already explained, the software control is
asserted by processors coupled to the control logic units and
communicating with one another via the communications system. The
delay in the exercise of the software control due to delays inherent
~2375C~
12
in this communication is of little consequence, because the traffic
on the failed channel has already been protected by being transmitted
on the protection channel.
The working channels are assigned respective priorities,
which are arbitrary as far as the protection arrangements are
concerned, in dependence upon which the software control forces the
control logic units 22 and 38 to various states. More particularly,
for the failed (and now protected) channel the software control
forces the unit 22 to adopt the DONE state 64 and the unit 38 to
adopt the DONE state 73. As in the DONE state 73 the unit 38
produces the signal UOUT=UIN rather than UOUT=1, the software control
also forces the control logic unit 39 for the protection channel P
to a state in which its selector 36 couples the line 50 to the
transmitter 40, this state being effective whenever protection
switching is in effect. For any higher priority working channel, for
which the protection channel P must be made available in the event of
a failure, the software control forces the unit 22 to (remain in) the
READY state 61 and the unit 38 to (remain in) the READY state 71.
For any 1Ower priority working channel, for which the protection
channel P will not now be made available in the event of a failure,
the software control forces the unit 22 to the LOCKOUT state 65 and
the unit 38 to the OFF state 74. In these states, which are also
used for idle working channels not requiring protection, the signals
PR=0 and BR=0, so that no protection switching of these channels can
take place until the software control dictates otherwise.
If, following a protection switch and consequent exercise of
software control as described above, a higher priority working
channel fails, this is protected in a similar manner. Thus the unit
22 of the higher priority failed channel adopts the ARMED state 62
producing the signal PR=1, and the unit 38 of the channel
consequently adopts the BRIDGING state 72, producing the signals
U0UT=1 and BR=1. If the previously failed channel is above (in Fig.
1) the higher priority failed channel, the signal U0UT=1 produces a
signal UIN=1 for the previously failed channel, so that its unit 38
adopts the OFF state 74. In any event, the traffic from the B MUX 34
of the higher priority failed channel is now passed via the line 80
to the protection channel P, ultimately producing the signal QB=l for
12375i0~:
13
this channel so that its unit 22 adopts the SWITCHIHNG state 63 to
protect the traffic on this channel and with subsequent software
control, in the manner already described. Meanwhile the B DEMUX of
the previously failed channel detects a channel identity mismatch,
producing the signal QB=O and squelching its output. If at this time
the traffic on the normal route for this channel is correctly
received, the A DEMUX produces the signal QA=l, and in response to
this combination of signals the unit 22 of the previously failed
channel adopts the LOCKOUT state 65 as illustrated in Fig. 4.
Having described the system head end and tail end,
intermediate units of the system will now be described with reference
to Fig. 6, which can be placed between Figs. 1 and 2 in order to
maintain the illustration of the signal flows.
As already described, at each intermediate site the
respective intermediate unit of the system may provide a
glass-through, repeater, or drop-and-insert capability for each
working channel. Fig. 6 illustrates a drop-and-insert capability for
channel 1 and a repeater capability for channel 2, a repeater and
selector being provided for the protection channel P. The protection
channel P also has a drop-and-insert capability for the optional
traffic which may be carried on this channel when protection is not
in effect, but this need not be provided. For consistency, in Fig. 6
similar references have been used to those in Figs. 1 and 2 to denote
similar elements, and some corresponding signal paths have been
omitted for clarity.
Referring to Fig. 6, for each repeatered working channel,
such as the channel 2, the intermediate unit includes an optical
receiver 14 and an optical transmitter 10 which are coupled via a
repeater interface unit (RIU) 90, the RIU consisting of a
demultiplexer and multiplexer coupled together in sequence. The RIU
90 thus permits recovery of overhead information from the respective
channel at a repeater site, and can have a form similar to that of an
A MUX 16 and A DEMUX 18 in combination.
The protection channel P similarly includes an RIU 90,
but in this case the output of the RIU is coupled to the respective
transmitter 40 via a selector 36, which is controlled by the control
logic unit 39 to supply to the transmitter either the output of the
14 1~37502
RIU 90 or traffic from the protection traffic line 80. As in Fig. 1,
the line 80 is chained through selectors 36 of the working channels,
but in this case only those working channels having drop-and-insert
capabilities have selectors 36. Such a situation is shown for
channel 1, for which the intermediate unit comprises components
substantially the same as those at the tail end (Fig.2) for dropping
DS-3 channel traffic and components substantially the same as those
at the head end (Fig. 1) for inserting DS-3 channel traffic.
Finally, Fig. 6 also shows the A DEMUX 18 and A MUX 16 for
respectively dropping and inserting via the line 80 optional extra
DS-3 channel traffic having the lowest priority.
In the event of a failure of channel 2, because this channel
is only repeatered in the intermediate unit in Fig, 6 its traffic
must be protected by routing it through the RIU 90 and selector 36 of
the protection channel, and hence via the protection channel spans on
both sides of the intermediate unit. In the event of a failure of
channel 1, for example on the span on the receive side (left as
illustrated) of the intermediate unit, however, its traffic need only
be routed via the corresponding span of the protection channel, i.e.
on only one side of the intermediate unit, leaving the protection
channel span on the other side free for carrying the optional extra
traffic or for protecting corresponding spans of working channels on
this other side of the intermediate unit.
In order to accommodate this, as already mentioned with
reference to Fig. 1 the signal UOUT from each control logic unit 38
is supplied as an input to the corresponding B MUX 34 to be
transmitted as one bit, referred to below as a signal U, in the
overhead information on the protection channel, and as also already
mentioned with reference to Fig. 4, each control logic unit 22 also
produces the signal Bl whose value is shown in Fig. 4 for the various
states which the unit 22 may assume. The signal Bl is applied from
each control logic unit 22 to a line 92 which as shown in Fig. 6 is
terminated with a pull-down resistor 94 to serve a wired-OR function
for all of the units 22, so that a signal Bl=1 produced by any of the
35 control logic units 22 produces this signal BI=1 on the line 92.
The signal U in the overhead information on the protection
channel is derived in the RIU 90 and is supplied to the control logic
1 Z 3 7 5 0 ~
unit 39 for the protection channel, as is the signal Bl on the line
92. If the signal U=1, indicating that the protection channel is
carrying protection trafFic from a failed channel, and the signal
BI=0, indicating that the failed channel is repeatered (or glassed
through) in this intermediate unit, then the unit 39 controls the
selector 36 to couple the output of the RIU 90 to the transmitter 40
to bridge the protection traffic through the intermediate unit. If,
instead, the signals U=1 and BI=1, then the unit 39 controls the P
channel selector 36 to connect the line 80 to the transmitter 40, as
the protection traffic relates to a drop-and-insert channel at this
intermediate unit and need not be bridged through the intermediate
unit on the protection channel.
While the control logic units 22 and 38 have been shown as
separate blocks in Fig. 6, it should be appreciated that they can
conveniently be combined to form a single control logic unit.
Numerous modifications, variations, and adaptations may be
made to the particular embodiment of the invention which has been
described above without departing from the scope of the invention as
defined in the claims.
