FI104665B - Hierarchical synchronization method - Google Patents

Description

104665

Hierarchical synchronization method

Field of the Invention

The invention relates generally to synchronization of telecommunication networks, and more particularly to improving the interference immunity of network synchronization in a telecommunication network using message synchronization.

BACKGROUND OF THE INVENTION In this presentation, the junction of the transmission links 10 of the telecommunications network is called a node. The node can be any device or apparatus, e.g. a branching or cross-coupling device.

The nodes of the message-based synchronization system are interconnected by the transmission connections they use for data transmission. The connections used also transmit the sender's clock frequency to the recipient. Each of the 15 nodes selects as the source of its own clock frequency either the frequency of a signal coming from its neighboring nodes, the frequency of its own internal clock source, or the frequency imported from the external clock source via a separate synchronization input. In order to make all the nodes of the system operate at the same clock frequency, it is generally sought to make the system synchronize to one clock-20 source, the so-called clock. the master source. This way, all nodes directly connected to the selected master source in the system are synchronized to this master source and to these: connected but without direct connection to the master source is synchronized to these nodes adjacent to the master source. Similarly, always v: nodes farther from the main source are always synchronized to those nodes that are one connection closer to the main source.

: In order to build a synchronization hierarchy like the one described above, the system nodes transmit synchronization messages to each other. These messages contain information that allows individual nodes>: to select the source of the timing. The system nodes are prioritized and the system tries to synchronize with the clock frequency of the node that has the highest priority level. Normally there is only »· · one node in the same priority level. Normally, the synchronization messages contain information '' about the origin of the clock frequency of the sending node, which is the current one.

node priority, and a value representing the quality of the clock signal. Thus, a single! '·' 35 node may select as the source of its own clock frequency the clock rate of the 2 104665 node from the desired node which is of the highest quality. During the system startup phase, each node selects its own internal clock source as its clock frequency source, since no incoming synchronization message has been processed. When the first incoming sync-5 climbing messages have been processed, the clock rate of the neighbor with the highest priority is selected as the source of its own clock frequency. Once the system has reached a stable state of synchronization with all messages spread to the system, the system is hierarchically synchronized to the master source clock frequency.

Figure 1 shows a system MS using message-based synchronization in a stabilized situation. Priorities assigned to nodes are indicated by numbers inside the circles describing the nodes. The lower the number, the higher the node priority. The synchronization messages sent by node n (n = 1 to 6) are denoted by the reference number MSGn. The synchronization message sent by each node 15 is generally different and depends on the message-based synchronization method used. The propagation of the clock frequency from the master clock (node 1) to other nodes in the system is shown by solid lines. Dashed-line inter-node connections are not normally used for system synchronization, but are available for modification-20 instances.

A simple principle in message-based synchronization is that v: the user defines a node synchronization hierarchy by giving each call

«I

me a unique identifier that tells the level of the node in the hierarchy, and the system: 'ΐ'ί is synchronized to the specified master clock independently, using j' \: 25 if necessary, taking advantage of all existing inter-node connections. If the connection: //: to the master clock is disconnected and there is no alternate connection, or the master clock fails, the system will be synchronized to the next highest level node. The change response in synchronization occurs between nodes. through messaging. When timing arrives at a node, the · · · 30 synchronization hierarchy is rebuilt from the breakpoint forward (away from the main system source). Usually the end result is an original hierarchical »I f Y,. * Structure where the failed connection is replaced by a working connection,

| J

· 'With almost no changes in structure.

Message-based synchronization methods such as those described above are f ·]: 35 described, e.g., in U.S. Patent Nos. 2,986,723 and 4,837,850, from which the reader desires, 104665

O

I'll get a more detailed description. The messages used in one of the known message-based synchronization methods (SOMS) are described in more detail below in connection with Figures 2 and 3.

As will be seen from the above, in message-based synchronization 5, synchronization has traditionally been built from the master node along the shortest path to other nodes in the network. All connections between the nodes have been equivalent in terms of synchronization. Each connection can be enabled for synchronization as long as it only meets the quality requirements at that time. However, if a connection is known to be susceptible to failure, it may still be used for synchronization in normal situations, as it currently meets the quality requirements for synchronization and is also located in a location in the network where the message-based synchronization method wants to use it. When this failure-sensitive connection used for synchronization fails, the message-based synchronization method used has to trigger its recovery method to at least partially rebuild the synchronization in the network. This may temporarily reduce the quality of the synchronization and bit slippage.

In the past, attempts have been made to avoid such situations by prohibiting the use of a connection for synchronization completely if its failure sensitivity is found to be too high. However, this approach has the disadvantage of not being able to take advantage of the alternate synchronization paths provided by the network in the event of a failure, and in the event of a failure, the connection to the master clock <RTIgt;

I · »

4 I

. I »v: Summary of the Invention * r f 25 The object of the invention is to overcome the above disadvantages, *, ·,! d, and provide a message-based synchronization method, i '; *: where connections known to be fail-safe cannot cause changes in network synchronization upon failure, and where all connections remain, ί Γ: however, the best possible use of the synchronization method.

jT: 30 This object is achieved by the solution defined in the independent claims.

•

The idea of the invention is to explicitly define certain connections for backup * · Ί 'as back-up paths so that the node devices mark the synchronization identifiers passed through those «» * connections with a special backup contact; Sun 35 Sat. (Note that connections are defined as backup connections only for 4,104,665 synchronization, not for normal data transfer.) Two different synchronization tags, one coming from a backup path (a route that includes a wireless connection) and one from a regular connection, but both originate from the same main clock, the one who travels the usual route always has a high-5 amp priority. Each node that synchronizes with a signal with a synchronization tag passed through a backup path or connection must retain information about the use of the backup path in the outgoing synchronization tag.

Thanks to the solution of the invention, network synchronization can be controlled by preventing the use of certain paths for synchronization in normal situations, even if the message-based synchronization method used would like to use this method. routes. However, in the event of a failure, these backup routes may be used if no other routes to the main clock are found.

The solution of the invention makes it possible to clearly distinguish between the prioritization between synchronization sources and the use of the backup path, since the use of the backup path does not affect the value of the synchronization tag itself. Therefore, the synchronization behavior is always clearly known and it is always known whether the backup path is enabled or not.

BRIEF DESCRIPTION OF THE DRAWINGS In the following, the invention and preferred embodiments thereof will be further described with reference to the examples in the accompanying drawings, in which:

4 4 I

. '; : Figure 1 shows a message-based synchronization system * I 4 • ,,.! when it is synchronized to the clock frequency of the master source, V '· Fig. 2 shows a self-guided slave synchronization (SOMS) network in its initial state, »5! Fig. 3 shows the network of Fig. 2 in a stable state, f · 0 »T: Fig. 4a ... 4d illustrate a method according to the invention« showing a message-based synchronization network in four different j situations, vi *; * t ': Fig. 5 is a flow chart which Fig. 6a shows a device implementing the method according to the invention; 6b illustrates an alternative structure of a node apparatus, '35 Fig. 7a illustrates an example of the structure of a synchronization message, and Fig. 7b illustrates one way in which the synchronization identifier and backup path information are transmitted in the synchronization message of Fig. 7a.

DETAILED DESCRIPTION OF THE INVENTION Figure 2 shows a self-directed slave synchronization (SOMS,

Self-Organizing Master-Slave Synchronization (a message-based synchronization method known per se), which in this case comprises five nodes (or devices) designated by reference numerals 1 to 5 according to their hierarchical level. (The master node on the network has a minimum of 10 SOMS addresses.) The nodes forward to each other messages containing the above SOMS addresses. This allows the nodes to identify each other with these address numbers and build a synchronization hierarchy so that the entire network can synchronize to the master node.

As mentioned above, synchronization messages that are continuously transmitted over the network are dependent on the message-based synchronization method used. In addition, the messages are unique to each sending node. In a SOMS network, a synchronization message has three components: a frame structure, an identifier, and a checksum. The SOMS tag is the most important part of the SOMS message. It consists of three consecutive digits 20 D1 ... D3: D1 is the origin of the synchronization frequency of the sending node of the SOMS message, i.e. the SOMS address of the node that appears to the sending node as the master node.

. v. D2 is a parameter describing the quality of the connection, which is typically a distance i «i. ·. September to a node expressed by D1. This distance is expressed as the number of · · 25 nodes in between.

• I (D3 is the SOMS address of the sending node.

. Each node (or device) continually compares incoming • · · ...

'.Y.' SOMS tags with each other and selects the smallest one. In the tag, parts D1, D2, and D3 are directly linked into a single number by putting them in sequence • · · v: 30 (D1D2D3) (for the sake of clarity, a hyphen will be written to separate the various parts;

Ml D1-D2-D3). This will make the smallest address the primary selection criterion. for previous nodes, the SOMS address (D1) of the node visible as the master node, i.e., node I, tends to synchronize with a signal originally originating from a node with the lowest address. Then, in a stable situation, the entire 35 network is synchronized to the same master node (because the master 104665 6 nodes of the whole network have the smallest SOMS address).

If two or more of the incoming signals are synchronized to the same master node, the one which is the shortest route (D2 smallest) is selected. The last selection criterion remains the SOMS address (D3) of the node 5 transmitting the SOMS message, based on which the selection is made, unless otherwise there is no difference between the incoming signals.

Once a node has accepted one of its neighboring nodes as a new synchronization source based on an incoming SOMS identifier, the node will have to recreate its own SOMS identifier. The new SOMS identifier 10 can be derived from the selected smallest SOMS identifier as follows: the first part (D1) is left untouched, the second part (D2) is incremented by one and the third part (D3) is replaced by the node's own SOMS address.

Each node also has its own internal SOMS identifier, X-O-X, where X is the respective identifier. the SOMS address of the node. If none of the incoming SOMS messages contains an identifier that is smaller than the internal identifier, the node uses its own internal oscillator or possibly a separate synchronization input as its clock frequency source. The outgoing SOMS message will of course use the internal SOMS identifier.

Nodes continuously send SOMS messages in each direction to: ·. The 20 changed information in the SOMS tags would spread as quickly as possible and the operational status of the neighboring nodes would be constantly monitored. Before SOMS tags can be compared, incoming SOMS messages * ·. '. ·. must be accepted and SOMS tags separated.

). The first time a · · 25 SOMS message is received from a particular transmission connection, its SOMS identifier is immediately accepted for comparisons if the message was • error-free. When the incoming bearer has. approved SOMS identifier, and an error message containing the same identifier is constantly appearing, the situation remains unchanged. If a · · · * \ * SOMS message is found to be invalid, the old v: 30 SOMS tag is still retained until three consecutive SOMS messages are received invalid. This will no longer accept the code. SOMS tag for comparison.

. · * ··. Waiting for three consecutive SOMS messages is intended to eliminate •: · ''. momentary disturbances.

If there is no SOMS message coming from the connection, even if the connection is otherwise 35, three consecutive SOMS messages will be waited until 7 104665 discards the current SOMS identifier. If the connection is completely lost, the SOMS tag is immediately discarded. If, due to interferences in the incoming signal, a valid SOMS identifier is not available for comparison, the associated SOMS tag is rejected. the SOMS identifier for the connection. In this case, the comparison is used. 5 SOMS identifiers for the inbound transmission, which are standard values identifiers, where each component (D1, D2 and D3) gets its maximum value (MAX-MAX-MAX).

When a new changed SOMS identifier is detected in an incoming SOMS message, it is immediately accepted for comparison if the message was error-free. This avoids unnecessary delays in network changes.

10 Initially, each node uses its own internal synchronization source, whereby it sends its other SOMS identifier X-O-X to the other nodes. This tag is also compared to incoming SOMS tags. If none of the inbound tags is smaller than the internal tag, continue with that tag. knot the use of your own internal timing.

Figure 2 shows a SOMS network in an initial state, whereby no node (or device) has had time to process incoming SOMS messages. For each node, the highest priority is given to the internal SOMS identifier, because the others have not yet been processed. In Figure 2, the incoming SOMS identifiers are marked with each node, and the selected identifier is written inside 20 frames (in the initial state of Fig. 2, all nodes use. ": Their internal timing source). The connections used for synchronization are drawn« · <. ·:, dashed connections, dashed connections (dashed line in Figure 2. v. initial situation, all connections are docked).

I When nodes have time to process incoming SOMS messages, node 1 · · · 25 stays on internal timing, nodes 2 and 4 synchronize based on ring 1 identifier 1-0-1, node 3 synchronizes to node 2 (2-0-2) and node 5 to node 3 (3-0-3). At the same time, the nodes create their own new · · ::: SOMS identifiers as described above and change their outgoing • · · SOMS message with a new identifier. The state of the network after it has stabilized is ♦ ♦ ·: 30 shown in Figure 3. All nodes are synchronized to master node 1 in the shortest possible * * · path.

. ···. As mentioned above, one or more of the network transmission connections may be · · · more susceptible to failure, e.g. the connection (s) is different in type from other connections. These fail-sensitive connections 35 are designated as backup connections for synchronization according to the invention by adding to the synchronization message passing through (or passing through) such connection that the corresponding synchronization identifier has passed through the backup connection.

Figure 4a-4d illustrates the effect of the method on the synchronization behavior of the network 5 by showing a network of eleven nodes in four different situations. In all situations, the main network clock is a node 1, the circle of which is oblique. It is assumed that the message-based synchronization method used is SOMS, although the invention can also be applied to other message-based synchronization methods in which the synchronization identifier to be transmitted contains the master clock identity or the like. The fixed lines indicate, similarly to Figures 2 and 3, the connections along which synchronization takes place.

Figure 4a shows an initial situation where synchronization runs along paths 1-> 2- »4-> 7-» 9, (1->) 2-> 5 and 1- »3-> 6-> 8-» 10 and ( 1-> 3 »6 ->) 8-» 11. 15 Assume that other connections in the network are, for example, fiber-optic connections, and two of the connections (denoted by the reference symbols X and Z) are radio links which are inherently more susceptible to interference. These connections are therefore marked as backup for synchronization purposes. Thereafter, synchronization is performed as shown in Figure 4b, i.e. along the paths 1-> 2-> 4-> 7- »9, 20 (1->) 2-> 5-> 6-> 3, (1-> 2-» 5- > 6 ->) 8-> 10 and (1 -> 2-> 5-> 6 ->) 8-> 11. No connection to X. . so it's no longer used for syncing.

If, for example, the connection between nodes 6 and 8 were to fail, it would move

. v. a network in terms of synchronization to the state of FIG. 4c with the backup connection Z

• «] is enabled for synchronization, otherwise nodes 8, 10 and 11 would not have • · · · *: ·. 25 obtained the master clock frequency. The synchronization now takes place along the paths t · · 1- »2-» 4- »7-» 9- »10-» 8- * 11 and (1- »2) -» 5- »6-> 3. If after this. still, for example, the connection between nodes 1 and 2 fails, the synchronization changes to figure 4d. In this case, the backup connection X is also in use by · »\ * n to spread the master clock frequency to the network.

Therefore, defining back - up connections to the network did not in any way increase the network 's resilience, but significantly improved the network' s resilience. (assuming connections X and Z are susceptible to malfunction or failure). If backup connections were not used, interruptions on X and Z would also cause changes in synchronization and thus affect the wider network whenever change-35 situations occur. On the other hand, when X and Z connections are defined as stand-by connections 104665 alone, interference with them does not in any way affect network synchronization.

FIG. 5 is a flowchart illustrating comparison of synchronization identifiers at a single node of the network when the message-5 based synchronization method is SOMS. The figure assumes that a "new" synchronization identifier received from one interface of a node is compared to an identifier received from another interface referred to as an "old" synchronization identifier.

In the first step (step 51), a comparison of the parameters D110 is performed. If the new identifier has a better (smaller) parameter D1, it is directly decided that the new identifier is better, regardless of whether the new or old identifier has come through the backup connection. Similarly, if the old identifier has a better parameter D1, it is directly decided that the old identifier is better, regardless of whether the new or old identifier has come through backup link 15 or not. If, on the other hand, the parameters D1 are equal, we proceed to look at the backup connection information associated with the synchronization identifiers. In step 52, it is first tested whether the new identifier has come through (one or more) backup connections. If the new identifier has passed through the backup connection, the next step (step 53) is to determine whether the old identifier originated from the backup connection. 20 If the old identifier does not originate from the backup path but the new one is, it is decided that. ""; the old tag is better. Similarly, if it is found in step 54 that the new I · · identifier does not originate from the backup path, but the old one is, it is determined that the new identifier is better. If neither identifier originates from the backup path or • ·. both are from the fallback route, going to step 55 for comparison, ···. 25 parameters D2. After that, backup connection information will no longer be used.

a a.a Based on the comparison in step 55, it is decided which of the identifiers is better with parameter D2. If the parameters D2 are equal to I · ·, we proceed to step 56 to compare the parameters D3 and select the better identifier ♦ · · v * 30 according to which has the better identifier D3. If also »» t 1 ...: the parameters D3 are equally good, then it is decided that the identifiers are equally good.

Compared to the comparison of synchronization tags in the known SOMS method, the block according to the invention has been added a block designated by the reference sign NB to examine the spare 35 path information contained in the tags.

10 104665

As can be seen in Figure 5, two synchronization identifiers indicating the same original synchronization source always attempt to select one whose synchronization identifier information indicates that the signal has not passed through a backup connection. Instead, of the two synchronization tag-5 paths whose original synchronization sources are different sources, the one whose parent synchronization source is at a higher level in the synchronization hierarchy is always selected. In this case, it is therefore irrelevant whether the selected identifier has passed through the fallback route. This way, in normal situations, synchronization is obtained from the best source.

10 The above are the preferred selection rules for the SOMS method. Selection rules may vary depending on the selected synchronization method and how it behaves.

Figures 6a and 6b illustrate, in a functional block diagram, those elements implementing the method described above at individual nodes of the network. The general structure of the node is, for example, such that it comprises a plurality of parallel units from the interface unit IU1, IU2 ... IUN, each of which communicates with at least one neighboring node, and a common control unit CU for all interface units. The control unit and the various interface units are connected to each other, for example, via the CBUS of the internal bus 20.

: * '': The figures show, by way of example, two neighbors to the system node ': transmission links from the nodes, A1 and A2, each connected to its own interface unit. Transmission connections are typically 2 Mbit / s PCM connections according to ITU-T Recommendations i G.703 and G.704 or SDH connections according to G.708 and G. · 25970709, respectively. In such signals, synchronization messages can be transmitted in different ways, one example being described below. Together. . ·. the interface unit IU may have one or more interfaces through which the node is connected ···. correspondingly swinging to one or more neighboring nodes. In general, • · t

[· "Thus, it can be stated that the node has N interface units with a total of M

* · * * 30 connections (M> N).

Figures 6a and 6b also show an index for the interface or interface unit specific reference numbers, and parts common to all interface units are shown without index.

Each transmission line is coupled to a signal transmitting and receiving part 35 13, (i = 1.2, ...), which performs physical signal processing. Part 13j forwards the 11 104665 synchronization message to the transmission and reception part 16, of the synchronization message connected thereto. The transmitting and receiving portions 16j perform e.g. message error checking and forwarding the message to the centralized node synchronization decision section 20 via the CBUS bus. The transmit and receive portions of the Sig-5 signal also monitor the quality of the signal it receives and store these data in interface-specific fault databases 14j. Each transmission and reception section of a synchronization message receives fault information from its corresponding database. The detection of a transmission fault / change in the transmit and receive portions of the signal is known per se.

The decision unit 20 of the control unit CU stores the synchronization identifiers received from the interfaces in the memory area 21, performs a comparison of the synchronization identifiers and, based on the comparison, selects the best identifier as the source of synchronization of the node. Based on the received identifiers, the decision part creates a priority list 15 in the memory area 22 having different synchronization sources in the order of priority assigned by their synchronization identifiers, with the top currently selected as the source of synchronization, based on its identifier. The decision-making part also receives from the interface units fault information for the corresponding signal, either in the form of a synchronization message. ' i ': 20 dos or as separate fault data.

The decision-making part maintains the current, outgoing; : ·. a synchronization tag in memory area 24, from which it provides a synchronization identifier. ·,, For interface-specific synchronization message transmitting and receiving portions 16j.

j. ·. An incoming or outgoing connection can be marked on the node in synchronization ···. There are two ways to make a backup connection, depending on where the backup contact information is stored.

Fig. 6a shows a first alternative in which each interface unit S · «knows which of its interfaces is connected to the backup connection. The user-made * S * backup connection specifications are stored for each interface (memory area 17j) in connection with the send and receive portions of the sync · v · 30 climbing message. When the transmission part of the synchronization message • · · ', ..' · receives a new outbound synchronization identifier for decision making, it adds a backup path information to it. When the receiving part of the synchronization message * ·: receives a message from the network and notices that the synchronization identifier it contains has changed, it inserts a backup path information 35 (if it does not already exist) before forwarding the identifier to the decision part of the node.

Figure 6b illustrates another alternative in which the user-defined backup connections are centrally stored in a single location (memory area 23 of the decision section) for the entire node. When the decision-5 co-component sends an outbound synchronization message to the interface units, it adds a backup contact information to those interfaces labeled as backup connections. When the decision component receives the received synchronization identifier from the interface unit, it knows whether the identifier has come from a backup connection and is therefore able to add a backup contact information before comparing the synchronization tags, if necessary.

The signal processing portions of the IU interface units receive from their other node a signal, typically 2048 kbit / s according to ITU-T Recommendations G.703 / G.704, with a frame having 32 time slots (TS0 ... TS31) and superframe with 16 frames. In the frame structure of this signal, the synchronization message may be transmitted e.g., such that the sync-15 cloning message allocates two bits of a frame structure from any time slot, preferably the TSO bits of the time slot (one frame having a TSO frame lock symbol; used to transmit a synchronization message). If the TSO bits of the time slot are used, 'j': 20 for a maximum of three bits will remain for other use, e.g. The synchronization message may also be allocated bits from another time slot, but in this case the required capacity must be taken. ': /; the capacity reserved for the payload.

, ·. . When, for example, the above two bits, \ from a suitable time slot, are reserved for the synchronization message, the message is transmitted in this selected channel • j *. 25 "at a time" (2 bits per frame).

The general structure of the synchronization message may be, for example, as shown in Fig. 7a, consisting of eight consecutive bytes. When viewed serially, the actual message starts at the first zero after eight consecutive ones (messages are sent consecutively without delay: · · · v: 30). After that, the most significant bit (# 8) of each byte is zero, so that the actual message body cannot enter eight consecutive ones, which would be confused with the opening byte. The first byte of the actual message has six bits (bits 2 to 7) reserved for the header information and the last bit I · (x) for the user data. The next five bytes contain bits 1 to 7 of user 35 data with bit 8 being zero. Bits 1-7 of the last byte have a checksum.

104665 The SOMS identifier D1-D2-D3 and the backup path information in such a synchronization message may be transmitted, e.g., as shown in FIG. 7b. Part D3 moves in bytes 2-4, part D2 in bytes 4 and 5, and part D1 in bytes 6 and 7. The backup path information Ww can be transmitted e.g. in bits 2 and 3 of the fourth byte, for example using the following combinations: 00: no synchronization tag has not passed through a backup path, 01: synchronization tag has passed through a level 1 backup path, 10: a synchronization tag has passed through a level 2 backup path, 11: a synchronization tag has passed through a level 3 backup path.

10 If you do not want to display different levels for the backup paths, using either of these bits alone (eg 0: Sync tag has not passed through the backup path, 1: Sync tag has passed through the backup path).

If backup path prioritization is used, the comparison also takes into account the level of the backup path and identifies the one with the highest priority level from tags coming from two different backup paths and from the same sync-15 source. For example, in the case of the SOMS method, the priority information thus in a sense forms a parameter D1.5, i.e. a parameter between parameters D1 and D2, which is used in the comparison after the comparison of the parameters D1 but before the comparison of the parameters D2.

-V; It is advantageous to adjust the sending buffer of the node to the length of the message (8 bytes), whereby in undisturbed operation the recipient finds the beginning of the message '/! ; always in the same place buffer, no need to go through the buffer unnecessarily: '·': to find the origin of each message.

If the communication system using the method of the invention is made up of, e.g., an SDH network in which the signals received by the node are similar to those described in ITU-T Recommendations G.707, G.708 and G.709. , the synchronization message can be transmitted in that part of the STM-N signal ·: *. (N = 1,4,16 ...) a frame specifically reserved for synchronization information (STM-1 frame header area).

9 I I

*; '* 30 There are several options for setting up a backup route. A straightforward way is to implement the emergency route directly, as shown in the diagram in Figure 5. Alternatively, the provision of a back-up route may be allowed only after a situation has occurred in the sol-Ι''Ί for a predetermined period that the node has not received the network clock clock frequency on any other route 35 other than the backup route. Namely, due to network delays, the synchronization tag from the master clock 14 104665 may, in the event of a fault or change, become somewhat delayed along some other route. By using the aforementioned time control, unnecessary round trips are prevented, i.e., the node is not switched first to use the identifier coming from the backup path and immediately afterwards to the identifier coming from another route 5.

The transmission connection in the node can be marked as a backup connection either during transmission or reception. In addition, marking can be made at both ends of the connection or at one end only. If the entry is made only at the node at the other end of the transmission connection, the entry is made at both transmission and reception. The latter option assumes that the connection is a backup connection in both transmission directions for synchronization purposes. Namely, the transmission link can only be used as a backup path in the other transmission direction for synchronization. Generally, one connection is used for synchronization only in the other direction, as shown in the examples above.

While the invention has been described above with reference to the examples in the accompanying drawings, it is to be understood that the invention is not limited thereto, but can be modified within the scope of the inventive idea set forth above and in the appended claims. As mentioned above, for example, the selection rules for a synchronization tag that passes through the backup path may vary depending on the selected synchronization method and how it behaves. • • • • • • • • • • • ••••••••••••••••••••••••••••••••••••••••• 9 • ·

Claims (11)

1. Hierarchical synchronization method for a telecommunications system using message-based synchronization and comprising multiple nodes interconnected with transmission links, wherein the processing nodes transmit each other signals containing the synchronization messages provided with a synchronization identifier indicating the synchronization identifier Priority in the system's internal synchronization hierarchy, characterized in that some of the transmission connections are defined as backup connections for synchronization and the respective synchronization identifier (Ww) is connected to the synchronization identifier (Ww) indicating whether the synchronization identifier has gone through a backup connection. Method according to claim 1, wherein the synchronization identifiers convey information about the original synchronization source, characterized in that the intention is that for two synchronization identifiers indicating the same original synchronization source, the identifier in which the synchronization identifier does not indicate the identification identifier is selected. via a backup connection. Method according to claim 1, characterized in that by two synchronization identifiers indicating different original synchronization sources, the identifier whose original source of synchronization is always selected: ':': higher in the synchronization hierarchy. Method according to claim 1, characterized in that , ·. the connections are distributed on different levels by using more than one bit in the backup path information. 25 Method according to claim 1, characterized in that the addition of said information is carried out in the nodes at both ends of a connection defined as a backup connection. 6. A method according to claim 1, characterized in that the addition of said information is performed in the node at one end of a connection defined as a backup connection. I · Method according to claim 6, characterized in that in the node addition of information is carried out in the transmission directions of the connection in question. Method according to claim 2, characterized in that a node, when it loses its connection to the main clock of the network, waits for a predetermined time before it selects a signal whose synchronization identifier as its source has a synchronization identifier. via a backup connection. 9. Node equipment for a telecommunication system using message-based synchronization and comprises several nodes connected to each other with transmission links, in which node equipment includes a centralized controller (CU) for deciding on synchronization of the node and interface units (IU1 ... IUN) by mediating whose interfaces a node star in conjunction with the other nodes in the network, which nodes transmit signals containing synchronization messages provided with a synchronization identifier indicating the corresponding signal priority in the system's internal synchronization hierarchy, which synchronization identifier da controller (CU), characterized by node equipment including - memory means (17 ^ 172; 23) by means of which part of the node equipment interface has been defined as backup connections for synchronization, and - means (16i, 162; 20) for connection of reserve connection information mation to synchronization identifiers that are conveyed via the relevant interface. Node equipment according to Claim 9, characterized in that: the memory means and the means for adding reserve connection information are: Ύ decentralized in connection with the interfaces corresponding to the reserve connections. : 11. Node equipment according to claim 9, characterized in that the memory means and the means for adding reserve connection information are centralized in the control unit (CU) of the node equipment. • • • • • • 1