WO1995033321A1

WO1995033321A1 - Securing of routing in a digital cross connect equipment

Info

Publication number: WO1995033321A1
Application number: PCT/FI1995/000294
Authority: WO
Inventors: Ove Strandberg; Sami KÄRNÄ
Original assignee: Nokia Telecommunications Oy
Priority date: 1994-05-26
Filing date: 1995-05-26
Publication date: 1995-12-07
Also published as: FI942466A0; FI96469B; FI96469C; FI942466A; AU2568195A

Abstract

Path protection in a TST cross connect is implemented in such way that less than half of the capacity is selected as the number of signals to be connected at the front end in the transmission direction. The signal to be path-protected is duplicated in the first time switch through the software, after which each of the duplicated signals (W, P) is routed through the TST using an ordinary TST routing algorithm for discrete signals. At the receiving end, both signals (W, P) are routed separately through the TS portion up to the last time switch, in which a selection is made through the software.

Description

Securing of routing in a digital cross connect equipment

This invention relates to a method, according to the preamble of claim 1, for implementing path protection in a digital TST cross connect. The invention also relates to a method of implementing broadcasting in a digital cross connect (DXC) .

The recommendations CCITT G.707 define the Synchronous Transport Module (STM-1) signals of the first level of SDH signals. Other defined levels are STM-4 and STM-16. The recommendation CCITT G.708 defines the frame structure STM-N (in which N = 1, 4, 16) . The STM-1 frame enables the transmission of 63 subsystem containers (e.g., TU-12, Tributary Unit, which can contain a two Mbps signal of an ordinary 30- channel PCM system) . The STM-N frames are assembled from several STM-1 signals, for example, the STM-4 signal is composed of four STM-1 signals.

Digital Cross Connect systems have been defined for the SDH; the CCITT draft recommendations G.sdxc-1...-3. SDH DXC define (abridging freely) : "A digital SDH cross connect is a cross connect device having two or more interfaces at SDH rates (G.707) and being at least able to terminate a transmission section and to controllably, transparently connect and reconnect virtual containers (VC) between the interface ports. "

An SDH DXC can transmit traffic between different SDH levels and connect traffic between different signals. The use of the cross connect also includes a possibility for remote control of routing, path protection, initialization of reserve routes, connection from one signal to several signals (broadcasting) , and so on. The connections are usually bidirectional.

Cross connects can be implemented with a number of architectures. Known are the TS structure (Time-Space) and the TST structure (Time-Space-Time) , which quite well fulfils the conditions of non-blocking properties and feasibility. The TST cross connect is also adapted to very large cross connects, though in this case certain problems arise when expanding the system. It has been attempted to eliminate these problems, among other things, through a double capacity TST architecture.

The starting point in this invention is a digital cross connect that is known per se, i.e. a DXC unit in which TST architecture is used. It is known that blocking can occur in a cross connect when the traffic need exceeds the routing possibilities of the switch. In an ordinary TST architecture, blocking can occur, for example, in cases in which broadcasting and/or path protection are needed. The problem could be solved with a different kind of architecture. For example, by means of the n(TS)T architecture presented in our parallel patent application , non-blocking broadcasting can be accomplished through simple measures and at a moderate cost because one path protection possibility that exists in a TST node is to connect the parallel T switch next to the real T switch of the input side. In this case, the incoming signal is copied to both T switches. These two T switches, which have the same content, could route the broadcasts 1 -> 2 in the same way as in the proposed n(TS)T architecture. It is nevertheless a condition that for the second T switch there is available an extra input port in the S switch. Now, however, connection of the 2 Mbps inputs has to be restricted, i.e. only half of the two Mbps interfaces, of which originally there were 2^*63, can be used.

In existing DXC nodes, the reconfiguration of the hardware of the switch stages for broadcasting and path protection proves to be too expensive, in terms of both costs and the installation time required for the changes. The task of the present invention is thus to propose a solution by means of which non-blocking routing and path protection can be provided for an existing DXC node. As regards path protection, the problem presented is solved by means of the invention by virtue of the characteristic features set forth in claim 1. Embodiments of the invention are presented in the dependent claims. Briefly, the invention can be described in such a way that the number of signals accessing a time switch is selected to be smaller than half of the capacity and that the signal to be path-protected is duplicated, through the software, i.e. the signal is not multiplied by connecting it to parallel time switches, whereby the signals that are duplicated according to the invention are routed with an ordinary routing algorithm for discrete signals. The signals to be path-protected are advantageously 2 Mbps signals.

The method in accordance with the invention can advantageously be applied in a node of a mesh network, in an interconnecting access node and especially in an access node that is located in a protected SDH ring, in a so-called self-healing ring, SHR. In nodes of this kind the method in accordance with the invention can also be used for broadcasting.

The advantage of the solution by way of this invention is that no hardware changes or other software changes need be made to the existing TST architecture because with the new routing algorithm alone, both path protection and broadcasting can be handled, in which case, of course, the numbers of signals accessing the time switches must be adjusted to the capacity of the switch.

The configuration and algorithm in accordance with the invention provide non-blocking routing in a DXC node. In the following the invention is discussed by means of embodiments with reference to the accompanying drawing in which

Figure 1 shows the typical interfaces of the node of an STM-4 ring;

Figure 2 shows the configuration of a non-blocking node in which there are 63 path-protected containers TU-12; Figure 3 shows an example of an SHR node in which the path- protected configuration is used;

Figure 4 shows an example of an access node between networks;

Figure 5 shows an example of an access node of a mesh network; and

Figure 6 is a schematic representation of the card slots of a DXC subrack that is known per se and the equipping alternatives for the card slots.

The DXC node that we have taken as an example in Figure 6 can comprise the subrack in accordance with a Synfonet system that is known per se, which subrack has slots for 19 circuit board units. The Synfonet system is manufactured by Nokia Telecommunications, Finland. The slots in the subrack can be equipped in a manner specified separately and in a specified order, of which certain of the most common alternatives are shown in Figure 6. The subrack is shown schematically as a rectangle in which the card slots are marked 1...19 in the bottommost row. On the left-hand side of the rack are shown alternative cards, or circuit boards, that can be installed in this subrack of the system. In the row of each card, at the different card slots, there is a marking indicating whether installation of the card in this slot is allowed (@) , not allowed and mechanically prevented (!) or whether it is a prohibited slot (empty) in which the card will not function. Alternative cards are a CU (control unit) , SS (space switch; S) , TS 1 (time switch; T) , TSWO (time switch for 64 kbit/s signals) , STM interfaces according to the standard (synchronous transport modules for STM-N, as well as 2 Mbps interfaces 2M, of which "2M" switches 16 bidirectional 2 Mbps links; 2MTA comprises a 2M basic card to which is added a T switch stage for connection to the S stage of the AU4 level) , SU and SPU. If only a small number of 2 Mbps signals are connected to the DXC node, the largest size of the node, 16^*16, or 16^*16 SMT-1 signals, can be achieved. If, however, all the inputs of the

16^*16 space switch (S) are not used for the STM-N signals, then the S switch can offer spare capacity. This is illustrated in Figure 1, in which the example is a typical configuration of signal interfaces in an STM-4 ring. This kind of situation occurs often in applying the SDH (Synchronous Digital Hierarchy) ring concept. The spare capacity is used in accordance with the invention to guarantee non-blocking operation. The path protection according to the invention will be examined in respect of the source or front end and the target or back end. In a typical connection situation, most of the path protections are implemented near the customer, i.e. near the terminal connected to the DXC network. Path protection can be considered a special case of broadcasting, in which n = 2. This means that the source and target of the path protection are a 2 Mbps interface. This can also be expressed in such a way that in broadcasting, the 1 -> 2 source is a 2 Mbps port and in the selection situation 2 -> 1 the target is also a 2 Mbps port.

In accordance with the invention, the solution to the problem stated is handled by means of the software, whereby without hardware modifications nearly the same effect can be achieved as with the above-mentioned n(TS)T solution. When for purposes of path protection less than half of the maximum amount of 2 Mbps signals (63/2 => 31) are attached to the second T switch, the duplication can be implemented in the first, i.e. the real, T stage. The duplication, or multiplication, is then done through the software, i.e. by means of the program, and not by duplicating interfaces, as was done in the n(TS)T example. The fundamental routing problem remains: now two identical signals must be routed from the same input port through TS instead of these two signals being routed from two parallel input ports. In the solution according to the invention, an ordinary TST routing algorithm that is known per se can nevertheless be used. It is to be noted that the duplication (multiplication) according to the invention is needed only in the ports for the 2 Mbps signals and not in the STM-1 ports. On the other hand, it can be said that an architecture resembling n(TS)T is implemented by means of software in the 2 Mbps part of the TST. As an example illustrating this, Figure 2 shows an STM-4 ring access node in which 63 (31 + 31 + 1) TTJ-12 signals access the T switches at any given time. The abbreviation TTJ-12 (tributary unit) in Figure 2 refers to the 2 Mbps signal according to the standard. The node in the figure is non-blocking in accordance with the invention and there are 63 path-protected links.

In path protection, in accordance with the SDH standard (Synchronous Digital Hierarchy, Sub Network Connection

Protection) , use is made of parallel W and P signals, which are physically routed by different paths from the source to the target. Accordingly, the abbreviation W for the signals means working and the abbreviation P means protecting; practically speaking, it makes no difference which of the signals is selected as the "working" one. In the case presented, the parallel W and P signals are created in the transmission direction in the time switch (T) that is the source by means of duplication. Thereafter these signals are routed through the TST in the manner of normal point-to-point signals and thus a conventional TST routing problem arises. In the receiving direction the W and P signals are routed to the same target T switch of the output side, in which the final selection is made (W or P is selected) . This means that the same number of TU-12 signals as in the original case calls for more S switch capacity in order to form path-protected links.

It can be seen from Figure 2 that one TU-12 signal per input port (marked with the number 63) remains unprotected and thus will be left unused for path protection purposes. The 63rd signal could nevertheless be routed without protection, but the benefit of this may be dubious because the routing procedures become complex and the customer would have a lot of trouble finding out which interface signals can be protected and which cannot. As could be seen from Figure 6, the DXC subrack in our example has eight card slots that permit the installation of 2 Mbps interface cards (2M) . In the original configuration, two S switch ports of the AU-4 level (AU, Administrative Unit) are used in these card slots. Thus, it is possible to switch a total of 2^*63 2 Mbps signals, but non-blocking path protection cannot always be guaranteed for these signals. The configuration specification is for one 2MTA per 63 Mbps channel, in which case 3 2M cards use the same time switch that is on the 2MTA card. Since the time switch has three ports, they can be utilized by replacing part of the 2M cards with 2MTA cards, whereby, for example, only 31 2 Mbps signals are connected to one time switch on a 2MTA card, and now non- blocking path protection can be guaranteed for these signals. A different number of 2 Mbps signals can be provided with non- blocking protection by selecting a suitable number of 2MTA cards to be put into use. If 4 2MTA cards are selected, a maximum of 124 2 Mbps signals can be protected. The original configuration specifies 2 2MTA cards, in which case a maximum of 62 signals can be protected. If the number of 2MTA units can be doubled, there is nothing preventing an even greater increase in this amount. If in the subrack in the example, the number of 2MTA units selected is the number of available card slots, then the maximum 2MTA number is 8. With this amount, the number of connected path-protected 2 Mbps signals can be raised to the maximum value of 126. It should be noted that the maximum amount (126) of 2 Mbps signals can be achieved already with six 2MTAs. The advantage obtained from using eight 2MTA units does not show up until some degree of broadcast (1 -> n) has to be implemented. The differences are summed up in Table

1, which compares the normal configuration of a DXC node with a configuration in accordance with the invention. Table 1: Configurations in different systems

Normal Path- Path- protected protected 1 2

Number of 2 Mbps 126 124 126 interfaces

Number of 2 Mbps 8 8 8 card slots

Number of 2MTA units 2 4 8(6)

Number of S ports 14 12 8(10) not used for 2 Mbps interfacing

Path protection of blocking non- non- 2 Mbps interfaces possible blocking blocking

Routing algorithm "^*?" or "-" TST + TST + duplication duplication extension extension

It can be seen from the table that in the path-protected case (path protection 1) the number of ports in other use than for 2 Mbps interfaces decreases by 2 compared with the normal configuration. The path-protected DXC node according to the invention thus offers two STM-1 interfaces less than in the normal case. Usually this loss is of no significance, as is explained below in more detail.

The following discussion deals with an example of an SHR ring access node (SHR, self-healing ring) . The signals to be protected in the access node have two Mbps interfaces as sources and targets, whereas the other signals, which have other nodes as sources and targets, travel through the given node in the point-to-point mode. Figure 3 gives a schematic representation of the SHR access node in the path protection configuration. Figure 2 shows the largest numbers of interfaces in conformity with the system in the example. It should be noted that non-blocking cannot be guaranteed for a protected TU-12 signal (TU, Tributary Unit) which resides in the STM-N interface signal. Table 2: Configuration alternatives for the access node

Alternatives STM-4 STM-1 2MTA Path Number of pro¬ ports tection, used in 2 Mbps the

S switch

STM-4 SHR 2 8(6) 126 16(14)

STM-4 SHR 2 4 124 12

STM-4 SHR 2 2 62 10

STM-4 & 2 2 6 96 16 STM-1 SHR

STM-4 & 2 6 2 32 16 STM-1 SHR

STM-1 SHR 8 8(6) 126 16(14)

STM-1 SHR 14 2 32 16

It should be noted that the table shows the maximum configuration values at any given time. Usually, simple STM-4 or STM-1 SHR rings will be used, in which case eight or two ports of the STM-1 level will be made available from the S switch. In these cases, utilization of the S switch will still be low.

The fundamental idea of the invention can also be used in an interconnecting access node. The task of the node is then to deliver the traffic gathered in the SHR to the upper level network and vice versa. This is illustrated schematically in Figure 4. The task is now to implement path-protected routing between the source and the target. This can be done in a non- blocking way for the path-protected VC12 when the same principle is applied as in the previous example. In other words, less than half of the interconnecting line capacity can be used. The reason for this is that half of the T switch capacity is needed to realize the duplication and for the selection of the path-protected VC12 signals. From the standpoint of the configuration, this means that the interconnecting capacity of the S switch must be more than twice the amount of the ring capacity. An STM-1-SHR ring thus requires 3 STM-1 interconnecting lines to the other network.

The discussion next deals with the use of the DXC node in our example in other networks, which can be generally described by means of a mesh network. In a mesh network, all the interface signals are randomly distributed. The configuration of the node varies and physical limitations are imposed by the size of the subrack and the capacity of the S switch. Figure 5 presents a schematic example of the position of an access node in a mesh network.

In examining 2 Mbps signals, the above-described SHR case holds. The path-protection of these signals to any STM-N port can be accomplished in a mesh network in the same way through copying (at the front end) and selection (at the back end) in the key switch. The divergence now is that there is more competition for the capacity of the S switch. The number of STM-N signals and the amount of the 2MTA is limited by the size of the S switch, i.e. the number of the STM-1 signals (and/or the signals corresponding to them) and the 2MTA signals must not, in the structure in the example, exceed the size of the switch, 16. The figures in Table 2 also hold for the mesh network, but now more configuration possibilities are available. One must contemplate, of course, a compromise between the number of protected 2 Mbps signals and the number of STM-N signals. With a smaller amount of protected 2 Mbps signals, more STM-N signals are obtained and vice versa. The limitations imposed by the subrack, for example, the subrack in Figure 6, must be taken into account in contemplating other configurations.

In a mesh network it is not always possible to guarantee non- blocking connection of path-protected VC12 signals to STM-N signals. In this case, the path protection must be realized with a separate TST path protection algorithm in which the protection is implemented with a space switch and which involves some degree of blocking.

Certain aspects of the invention have been discussed above. In the manner proposed, the signals of a 2 Mbps interface can be path protected in a non-blocking manner. A change is made in the existing system to provide path protection through the software. The path protection somewhat limits the utilization of the capacity of the S switch of the TST node, but to a very moderate degree, which can be expressed as follows:

Table 3 : Maximum configuration of the S switch (With the example in Figure 6, cf. Table 1) :

Numbers of signals Original Path-protected according to the invention

2 Mbps 126 126

STM-1 4 2

STM-4 2 2

Certain advantageous applications that can be achieved with the invention have been discussed above. One versed in the art will understand that the idea behind the invention can also be applied to other cases. Accordingly, a wider use of the 2MTA units could be contemplated, whereby also larger 2 Mbps broadcasts (1 -> n) could be implemented. In this case, however, when making the connection from the T switch of a 2 Mbps interface, blocking may occur, this depending on the degree of broadcasting (n) and the number of 2 Mbps signals connected to the 2MTA unit.

For the sake of clarity there is reason to mention that in respect of TU-12 signals, it is not expedient to implement broadcasting and path protection in a TST in a manner according to the invention if the TU-12 signal is part of the STM-N signal. To implement non-blocking path protection also in the maximum case, in the access node more than half of the STM-N signal would have to be used for this purpose, which does not seem sensible, cf. the interconnecting node case in Figure 4. Here a more correct solution would be to use a separate TST protection algorithm.

It is also conceivable to combine the method according to the invention and the features of a separate TST protection algorithm. Accordingly, by means of copying/duplication and selection, the problems of the protection algorithm can be minimized by "removing" a portion of the path protections, i.e., the protection algorithm is used to process a number of different signals (W or P) of protected links as separate signals, which also enhances the performance of the protection algorithm.

Presented above is a way of protecting SDH signals. Of course, other signals having a corresponding structure, such as PDH signals, can use the same method of protection.

Claims

1. A method in a digital TST cross connect for implementing path protection, which at the front end of the link comprises division of the signals to be switched such that they proceed along two physically different routes and at the back end of the link it comprises selection of the second received signal as the active signal, characterized in that less than half of the amount of signals permitted by the capacity of the time switch are selected as the number of signals to be connected to the time switch of the input side of the front end, and that the signal to be path-protected is duplicated through the software in the first time switch, after which each of the duplicated signals (W, P) is routed through the TST cross- connect utilizing an ordinary TST routing algorithm for discrete signals, whereby at the back end the pair of signals that has been protected at any given time is directed to the output time switch determined by its address, in which time switch one of the signals is selected as the actively received signal, and whereby the number of path-protected signals to be connected to the time switch of the output side is smaller than half of the amount of signals made possible by the capacity of the time switch.

2. A method according to claim 1, characterized in that the signals to be path-protected and connected to the time switch of the input side and the time switch of the output side are 2 Mbps signals.

3. A method according to claim 1 or 2, characterized in that it is implemented in an SHR ring access node and/or an interconnecting access node and/or in the node of a mesh network.

4. A method according to claim 1 or 2, characterized in that it is implemented as the node of a mesh network, whereby in the routing of the signals an algorithm is used which can produce blocking.

5. A method for implementing broadcasting by means of a digital TST cross connect, which method comprises, at the front end of the link, the division of the signals to be connected such that they pass along several different routes for routing to several receivers, characterized in that the number of signals to be connected to the time switch of the input side is selected such that the number of connected signals and the sum of the duplicated broadcast signals is smaller than the number of signals made possible by the capacity of the time switch, and that the signal intended for broadcasting is duplicated through the software in the first time switch, after which each duplicated signal is routed through a TST cross connect utilizing an ordinary TST routing algorithm for discrete signals.