CN1232585A - Layered Synchronization System - Google Patents

Abstract

The invention relates to hierarchical synchronization method for use in a communications system utilizing a message-based synchronization method, said communications system comprising a plurality of nodes interconnected by a plurality of parallel transmission links. The nodes interchange signals incorporating synchronization messages which include a synchronization signature capable of indicating the priority level of the transmitted signal in the internal synchronization hierarchy of the system. The node performs comparison of the received synchronization signatures in order to select the signature of the highest priority as the synchronization source of the node. To provide the system operator with a facility of controlling the network synchronization to use a given one of the parallel links simultaneously retaining the possibility of using any of the parallel links in a rapid and flexible manner during, e.g., fault situations, at least one node of the network performs tagging of the parallel links, which interconnect the node with a given neighbouring node, with parallel signatures indicating the mutual quality ranking of the parallel links, whereby said parallel signatures are used in the comparison process to determine the mutual priority ranking of the synchronization signatures received over said parallel transmission links.

Description

Layered Synchronization System

The present invention relates generally to synchronization between communication networks, and in particular to improvements in network synchronization in a communication network using message-based synchronization.

In the context of the present invention, the term node is used to denote a connection point of transmitting parts in a communication network. Accordingly, any means and equipment may be used to form a node, such as a branching unit or a jumper.

In a system using message-based synchronization, the nodes of the system are interconnected by transmission links that serve the information transmission needs of the system. The same links also convey the clock frequency of the sending node to the receiving node. Each node selects as its own clock frequency source either the frequency from an adjacent node, or a frequency brought into the node from an external source through a separate clock input. In order to force all nodes of a system to operate on the same clock frequency, in most cases the system is preferentially synchronized to a single clock source called the master clock. To achieve this, all nodes in the system that are directly connected to the selected master clock will be synchronized to said master clock, while more nodes that are not directly connected to the master clock but are connected to the direct Nodes remote from the ground-connected nodes will be synchronized to the nodes located closer to the master clock. Similarly, nodes further away from the master clock will be synchronized to those nodes located closer to a link from the master clock.

In order to establish a synchronization hierarchy of the above scheme in a communication system, the nodes of the system exchange synchronization messages with each other. These messages convey information that enables individual nodes to select the best clock source for their clock synchronization. The nodes of the system are prioritized, and the system tends to synchronize itself with the clock frequency of the node having the highest priority. Under normal circumstances, a given priority can only be assigned to one node of the system. The synchronization message usually includes information about the clock source used by the sending node, the priority of the sending node, and a parameter value indicating the quality of the clock signal. Therefore, any individual node can select as a synchronization source for its own clock signal the clock frequency of an adjacent node which is derived from a desired node and has the highest quality. When the system is started, no incoming synchronization message has been processed. At this time, each node uses its internal clock as the clock frequency source. Once the first set of incoming sync messages has been processed, the node selects the clock frequency of the highest priority neighbor node as its clock frequency source. After all synchronization messages required have been propagated through the system to bring the system to a synchronized steady state, the system operates hierarchically synchronized to the clock frequency of the master (clock frequency) source.

A system MS using message-based synchronization in steady state is shown in FIG. 1 . The priorities assigned to the nodes are indicated by numbers marked inside the circles representing the nodes. The lower the value of this number, the higher the priority of the node. The synchronization message sent by node n (n=1-6) is marked with reference symbol MSG _n . In general, the synchronization messages sent by the individual nodes are mutually distinct and have a format which depends on the message-based synchronization method used in the system. The distribution of the clock frequency from the master clock (node 1) to the other nodes of the system is shown in solid lines. The inter-node links indicated by dashed lines are not used for system synchronization under normal conditions, they are only useful in case of state changes.

The basic concept of message-based synchronization is that a synchronization hierarchy of nodes is defined by the system operator by assigning each node a dedicated label denoting the node's rank in the hierarchy and allows the system Synchronizes itself in a self-contained manner to a defined master clock as required using all existing inter-node links. If the synchronization chain connected to the master clock is broken, and there is not a redundant link among the links between nodes, or the master clock fails, the system will instead synchronize to the next highest level node. Such a reaction to changes in the synchronization of the system takes place via the exchange of messages between the nodes. In the event that the timing information received by a node is corrupted, the synchronization hierarchy is rebuilt from the point at which the discontinuity of synchronization began onwards (ie, hierarchically outward from the system's master clock node). In general, the end result is a hierarchy that approximates the original situation, with the remainder of the system configuration remaining more or less the same except for the replacement of the failed link with a functional link.

Various message-based synchronization methods of the above-mentioned kind have been disclosed, for example, in US Patent Nos. 2,986,723 and 4,837,850, and more details on this type of synchronization scheme can be found. A message-based master-slave synchronization method (SOMS) in the prior art will be described in more detail below in conjunction with the discussion of FIG. 2 and FIG. 3 .

As can be seen from the above description, in message-based synchronization methods, synchronization is usually established on the shortest path from the node running under the master clock to other nodes in the network. According to this concept, all links between nodes are considered equal as far as synchronization is concerned. Therefore, any link can be used for synchronization purposes, provided it meets the specified quality criteria at all times.

However, in communication systems between two nodes using parallel links to implement various connections between nodes, the system operator does not know which of the parallel links is selected for use at any time. The synchronization system. Accordingly, the control options available to the system operator have been limited to the most basic means of prohibiting the use of one or more of the parallel links for synchronization. However, this solution has the disadvantage that said links are completely removed from the pool of available synchronous connections, so they cannot be used for special cases such as system failures.

It is an object of the present invention to overcome the above-mentioned disadvantages of conventional techniques and to provide a message-based synchronization method in which the system operator can control network synchronization to use One of the parallel links, while retaining the possibility to use any one of the parallel links in a fast and flexible manner in all cases.

The object of the invention is achieved by the solution described in the appended independent claims.

The invention is based on the concept of allowing the system operator to set the priorities of the different parallel links according to an operator-defined priority classification method. In a normal situation where all parallel links are available, the synchronization flag of the signal received through the highest priority link is selected. The other parallel links are only used if the highest priority link fails and the synchronization method used still prefers a flag sent over the same internode path.

On the condition that there are multiple parallel links between two nodes and without compromising the possibility of utilizing any parallel links in special cases, by means of the arrangement according to the invention, the linking performed by the synchronization method Choices may be affected. Thus, the synchronization method can be controlled so that a given preferred link is always used under normal conditions, while allowing the functional parallel links to be used for synchronization in the event of a failure. As a result, for example, an unstable link can be prohibited from being used for synchronization when other parallel links with higher reliability are available. Furthermore, the prioritization of parallel links can be utilized in an easy and controlled manner to change the preferred synchronous link when, for example, reconfiguring the network.

In the following, the invention and its preferred embodiments will be examined in more detail with the aid of examples illustrated in the accompanying drawings, in which

Figure 1 shows a communication system using a message-based synchronization (method) when the system is synchronized to the clock frequency of the master (clock frequency) source;

Figure 2 represents a network using an ad hoc master-slave synchronization scheme in its initial state;

Figure 3 represents the network of Figure 2 in a steady state;

Fig. 4 represents a flow chart of the comparison process of synchronization marks according to the present invention;

Fig. 5 a represents a kind of device, is suitable for carrying out the method according to the present invention in a single node of this network;

Figure 5b represents a flexible embodiment of the node device;

Figure 6a represents an example of the structure of the synchronization message; and

Fig. 6b shows an exemplary embodiment illustrating the transmission of the synchronization flag shown in Fig. 6a and the alternate path information in the synchronization message.

Referring to Fig. 2, there is shown a system utilizing self-organizing master-slave (SOMS) synchronization (which is a known message-based synchronization technique), in the illustrated case, the system Five nodes (or devices) are included, denoted by reference numerals 1-5, respectively, according to their assigned levels in the synchronization hierarchy. (In this context, the master node of the network has the lowest numerical SOMS address.) Messages exchanged by the nodes include said SOMS address. Thus, by means of these address numbers the nodes can be mutually identified and a synchronization hierarchy is established so that the entire network is synchronized to the master clock source node.

As mentioned above, the synchronization messages exchanged successively in the network depend on the type of message-based synchronization method used. Moreover, each sending node sends a group of messages with a specific content. In a network using SOMS synchronization, the synchronization message consists of 3 different parts: a frame structure, a flag and a checksum. The SOMS token is the most important part of the SOMS message. The mark consists of 3 consecutive numbers D1-D3:

D1 indicates the origin of the synchronous clock frequency used by the sending node, ie, the SOMS address of the node that acts as the master clock source node for the sending node.

D2 is a connection quality parameter which is typically given a value proportional to the distance of the receiving node from the node indicated by D1. The distance is expressed as the number of nodes between the source node and the receiving node.

D3 is the SOMS address of the sending node.

Each node (or device) continuously compares incoming SOMS tokens and selects the one with the lowest value. At the same time, the fields D1, D2 and D3 are combined directly in the markup into a single number formed by the direct combination (D1D2D3) of the fields (fields) (however, in the text a hyphen is used to separate the different parts of the marker, eg, D1-D2-D3). Therefore, the main criterion for selecting the address with the lowest value will be based on the SOMS address (D1) of the node that appears to be the master clock source to previous nodes, meaning that any node that wishes to synchronize itself to a signal whose The clock frequency source can be traced to a node with the lowest possible (numerical) address. Then, the entire network in steady state will run under the condition of being synchronized to the same master node (since the master node of the entire network has the lowest numerical SOMS address).

If two or more incoming signals are synchronized to the same master node, the receiving node will select the master clock frequency from the sending node via the shortest path to the master node (i.e., the value of D2 is the smallest) . The final selection criterion is based on the SOMS address (D3) of the sending node of the SOMS message. If the incoming signals cannot be sorted out in the previous address selection steps, then the minimum value of this address can be selected.

After a node has been verified as a new source of synchronization from among its neighbors by comparing the SOMS signatures of incoming signals, the node must reconstruct its own SOMS signature. The new value of the SOMS tag can be derived from the selected minimum value SOMS tag as follows: field 1 (D1) remains unchanged, field 2 (D2) is increased by 1, and field 3 (D3) is replaced by the Replaced by the node's own SOMS address.

In addition, each node has an internal SOMS token in the format X-0-X, where X is the node's own SOMS address. If none of the incoming signals has a group of SOMS messages with a value less than the node's internal SOMS flag, the node will select the node's internal oscillator, or possibly the one present at the node's external clock signal input. A synchronization signal used to synchronize the clocks of this node. Obviously, the node's internal SOMS token will then be used for the outgoing SOMS message.

The node sends the SOMS message continuously in each direction to ensure maximum rapid transmission of the changed synchronization information in the SOMS label and to keep the neighboring nodes informed of each other's (node) working status. Before SOMS messages can be compared with each other, the incoming SOMS message must be accepted and the SOMS signature must be separated from it.

When a given transmission link first submits a group of SOMS messages, if the messages are error-free, the SOMS flags of the messages are immediately accepted for comparison. This state remains unchanged when the incoming signal from the transmit link has an acceptable SOMS signature and the received message conveying the same, unchanged signature is error-free. If an error is found in the received SOMS message, the current SOMS flag is kept valid until 3 consecutive sets of erroneous SOMS messages have been received. The SOMS marker is then excluded from the comparison process. This waiting for 3 consecutive SOMS messages can be used to eliminate temporary disturbances.

Even if a link is performing acceptably, if the link is unable to deliver any SOMS messages, the comparison process activates a SOMS message corresponding to 3 consecutive SOMS messages before the currently selected SOMS flag is considered invalid. text delay. If the link fails completely, the SOMS marker is discarded immediately. Equally, when the interference superimposed on the incoming signal makes it impossible to extract a SOMS signature of sufficient quality for comparison, the SOMS signature of the link is discarded. Instead, the comparison is performed using the default (default) SOMS flag with all fields (D1, D2 and D3) set to their maximum value (MAX-MAX-MAX) for the incoming link process.

When an incoming message is detected with a new, changed SOMS signature, the signature is directly included in the comparison process, assuming that the message is error-free. This way, any changes in the network configuration are processed without delay.

Initially, each node uses its own internal synchronization frequency source, whereby the node sends internal SOMS tokens in the format X-0-X to the other nodes. This token is also compared to other SOMS tokens that are coming. If none of the incoming tokens has a value less than the node's internal token, the node will continue to use its internal synchronization source.

Still referring to Fig. 2, when so far no node (or device) has time to process incoming SOMS messages, the SOMS-synchronized network shown in the figure is in its initial state. All nodes give highest priority to the node's internal SOMS label since no other labels have been processed so far. In Fig. 2, the coming SOMS mark of each node is all marked next to the node, and the selected mark is marked in the frame (here it must be pointed out that in the initial state shown in Fig. 2, All nodes use their internal synchronization sources). The links used for synchronization are drawn with solid lines, and the spare links are shown with dashed lines (it must thus be noted that in the initial state shown in Figure 2, all connected lines represent spare links) .

After the nodes have had some time to process incoming SOMS messages, node 1 continues to synchronize to its internal clock source, node 2 and node 4 will synchronize to node 1 on the basis of their markers 1-0-1, Node 3 will be synchronized to node 2 (label 2-0-2), and node 5 will be synchronized to node 3 (label 3-0-3). Also, the nodes will rewrite their own new SOMS tags in the above style and add this tag to the SOMS message being sent out. After the network has entered a steady state, its configuration will be as shown in Figure 3. Here, all nodes are synchronized to the master node 1 via the shortest possible path.

As mentioned above, in the prior art, the system operator has not always had the possibility to control the synchronization behavior when two or more parallel links already exist between two nodes, thus, for example, prohibiting the The least reliable link used for synchronization was ever necessary. According to the present invention, direction has parallel flags (i.e., parallel priorities, which are used in the comparison process of sync flags in order to arrange the synchronization flags received in parallel links relative to each other. The nodes in the order of quality) provide a set of parallel links connecting a node to an adjacent node,

Figure 4 shows a flow chart illustrating the comparison of synchronization flags which takes place at each node in a network using a message-based SOMS synchronization method. In the procedure shown in the figure, it is assumed that a "new" sync-flag received from one interface is compared with a so-called "old" sync-flag received in the figure from some other interface.

In the first step of the process (step 41), the values of the flag parameters are compared. If the value of parameter D1 in the new signature is superior (with a smaller value) to the value in the old signature, it is immediately determined that the new signature has a higher quality. Similarly, if the parameter D1 of the old flag is better, it is directly determined that the old flag is prioritized. In contrast, if the values of the parameters D1 are equal, then at step 42 the process compares the parameters D2 and selects the tag having the better value in the parameter D2. In the event that the values of the parameters D2 are equal, then the process compares the parameters D3 in step 43 . Such a comparison step is carried out in the manner described above, whereby the marker having the better value among the parameters being compared is directly selected. If even the comparison of the parameters D3 fails to find a qualitative difference between the marks, then in step 44 the process compares the parallel priorities. Here, the tag with the higher tie priority is selected. Thus, comparisons of parallel priorities are made only if the "SOMS parts" (D1-D2-D3) of the tags are equal. This is the result of both tags passing from the same node on parallel links. In fact, this embodiment according to the present invention improves the synchronization flag comparison process of a conventional SOMS method by adding to the comparison process a step 44 of performing a comparison of parallel priorities.

Figures 5a and 5b show functional block diagrams of means suitable for implementing the method described above in a single node of a network. In its generalized configuration, the node may comprise, for example, a plurality of parallel interface units IU1, IU2,..., IUN, each of which communicates with at least one adjacent node, and a The common control unit CU, whereby the control unit carries out the decisions related to the synchronization of the nodes. The control unit and the individual interface units can communicate with each other eg via a set of internal buses CBUS of the node device.

As an example, these two figures represent nodes of the system communicating with neighboring nodes via two incoming links _A1 and _A2 , each of which terminates in its own interface unit . Typically, the connections are formed by two sets of 2 Mbit/s PCM digital signals compatible with ITU CCITT Recommendations G.703 and G.704 or SDH signals compatible with ITU CCITT Recommendations G.708 and G.709. The signals in these connections are likely to carry synchronization messages in various ways. One of the exemplary embodiments will be described below. Each interface unit IU may have one or more interfaces via which the node is respectively connected to one or more neighboring nodes. In general, a node can be characterized as using M interfaces to form N interface units, the total number of which is (M ³ N).

In FIGS. 5a and 5b, reference symbols specific to an interface or interface unit are indicated with a subscript, whereas components common to all interface units have no subscript.

Incoming signals of each connection pass through a signal transmission/reception block 13 _i (i=1,2,...), in which the actual signal processing takes place. The sync telegram extracted from block 13 _i is also passed to a sync telegram send/receive block 16 _i directly connected to it. Among other various operations, the synchronization message sending/receiving block _16i checks the integrity of the synchronization message and sends the confirmed message to a centralized decision of the node via the common bus CBUS Box 20. The signal transmission/reception block also monitors the quality of the received signal and stores the results in the interface-specific error database _14i . In this way, each isochronous message send/receive block can retrieve error data from its dedicated error database. In each signal transmit/receive block, various conventional methods are used to monitor a set of signals transmitted from a transmit link for errors or changes.

The decision block 20 of the control unit CU stores the synchronization flags received from the interfaces in the memory area 21, compares the stored synchronization flags, and on the basis of the comparison, selects the one with the highest priority The flags of the clock serve as a synchronization source for its own clock. On the basis of the tokens received, the decision block forms a priority list in the storage area 22, in which the available synchronization sources are in descending order according to the priorities resolved from the contents of their synchronization tokens The ordering is done such that the highest (priority) in the list is set as the clock frequency source currently selected for the node's synchronization, and its signature is used as a template for forming the outgoing signature of the node. The decision block also obtains from the interface units error information related to the incoming signals of the selected synchronization source, either as a set of synchronization messages, or as individual error data.

The decision block also saves the currently valid value of the outgoing sync flag in the memory area 24, whereby the flag is assigned to the interface-specific sync telegram send/receive block 16 _i .

In the node, depending on where the priority information is stored, two different methods can be used to define parallel priorities.

In Figure 5a, a first alternative arrangement is shown, in which the priority definitions made by the system operator for the parallel links are individually for each interface communicating with the same node storage (in the storage area _17i ), whereby each storage is adapted in connection with the respective synchronization message transmission/reception block. When the synchronization message receiving block receives a group of messages through a link of the network and finds a changed synchronization flag, it will The parallel priority of the route is added to the tail of the received message information. This added priority information is then utilized in the manner shown in FIG. 4 during the tag comparison process performed by the decision block. In a SOMS based system, this scheme works well since the decision block can tell directly from the messages which are received from the same node (ie have equal D3 values). As a result, there is no need to store separate information telling which interfaces are connected to mutually parallel links. In contrast, these synchronization methods, which receive the same synchronization flags from different nodes, inevitably require means for storing information about these parallel links, or conversely, the flags or messages transmitted in the network It should be supplemented with information about the originating node, whereby the receiving nodes can independently detect which links are formed by parallel links.

Figure 5b shows an alternative arrangement in which the priority definitions of the parallel links entered by the operator are stored in a centralized manner in one location (the storage area 23 of the decision block), for this node to share. Now, when the decision block receives a new synchronization marker from some parallel link, it can identify the priority of the marker transmitting link and use it in the comparison process shown in FIG. 4 . (It must be pointed out here that if the tags to be compared are not transmitted from the parallel links, the comparison step of the parallel tags will not be entered.)

Typically, the signals sent from another node of the communication system to the signal processing blocks of the interface unit IU are, for example, a set of 2048 kbit/s signals conforming to ITU CCITT G.703/G.704 recommendations , the signal frame includes 32 time slots (TS0-TS31), and each multiframe (multiframe) is formed by 16 frames. In the frame structure of this signal, the synchronization message can be transmitted such that, for example, in certain time slots of the frame structure, two bits are reserved for the synchronization message, preferably from time slot TS0 (here it must be pointed out that the frame alignment signal occupies the bits of time slot TS0 of every other frame, and every other frame leaves bits 4-8 reserved for national use, whereby they can be used to transmit the synchronization message). If the bits of time slot TS0 are used to transmit the synchronization message, then at most 3 bits are reserved for other purposes, such as the traffic channel. Obviously, the bits required for the synchronization message could also be reserved from some other time slot, but this arrangement is penalized in that it steals the necessary transmission capacity from the capacity reserved for the (communication) load.

After reserving the aforementioned two bits for the synchronization message from an appropriate time slot of the frame structure, the message is sent in a "segmented" fashion (two bits in each frame) sent on the selected channel.

The generalized structure of the synchronization message may be as shown in FIG. 6a, for example, including 8 consecutive bytes. In serial transmission, the actual message is sent starting from the first "0" following the string of 8 consecutive "1"s (messages are sent consecutively without delay). After the first byte, the highest bit (bit 8) of each byte is set to "0", in order to avoid the message byte containing 8 "1", otherwise it will be in the message byte and the message byte Scrambling occurs between start bytes. In byte 1 of the actual message, 6 bits (bits 2-7) are reserved for header information and the last bit (x) is reserved for user data. The following 5 bytes each use its bits 1-7 to convey user data, while bit 8 is set to "0". Bits 1-7 of the last byte contain the checksum of the message.

In this type of synchronization message, the SOMS flag fields D1-D2-D3 may be transmitted in the manner shown for example in Fig. 6b. For example, bytes 2-4 are used to transfer field D3, bytes 4 and 5 are used to transfer field D2, and bytes 6 and 7 are used to transfer field D1.

Preferably, the length of the sending buffer in the node is set equal to the length (8 bytes) of the message, thus allowing the receiving node to find out the same message that is usually in the buffer in a non-disturbing operating condition. The starting point of the message at one point, thus there is no need to initiate a scan of the buffer contents in order to find the starting point of each group of messages individually.

In the case where the communication system using the method according to the invention is formed by, for example, an SDH network, in which the signals appearing at the input ports of the node comply with ITUCCITT G.707, G.708 and G.709 recommendations (Regulations), the synchronization information can be transmitted in the part reserved for the synchronization information (the header part of the STM-1 frame) in the STM-N signal frame (N=1, 4, 16, ...) .

The method according to the invention can be implemented in a number of alternative ways. For example, the parallel links can be individually prioritized, i.e. each link is assigned a separate class, or conversely, the links can be divided into a given number of priority classes, For example, only two priority classes (defining primary links and secondary links) are included. Different parameterization alternatives increase the system operator's options in terms of network maintenance.

In this way, when a faulty primary link (i.e., a link with the highest parallel priority) becomes available again, the node can automatically activate its use, or conversely, the node can be scheduled to to wait for a command from the operator to activate the use of the link.

As described in the example of Figure 4, the parallel priority of a link can be used in this comparison, or conversely, a list can be written in which each different synchronization flag appears only once. After selecting the best of these labels, the interfaces receiving said labels are searched, the parallel priorities of said interfaces are checked, and on this basis, the most suitable This link to the sync source. This implementation requires that the synchronization tokens be unique and node-specific (meaning that a node can only send a given token at a time). If this is not the case, the synchronization flag must be supplemented with information indicating the integrity of the sending node (in the SOMS method, this is the parameter D3).

Classification of priority may also be based on other criteria, such as the type of connection (eg, fiber optic link versus wireless link).

It is obvious to a person skilled in the art that the invention is not limited to the exemplary embodiments described above with reference to the accompanying drawings, but can be found in the appended claims and the examples given above. Make changes within the bounds of the scope and creative spirit. For example, those parts of the links that communicate with only one neighboring node can be classified as parallel labels. Thereafter, for example, all "unclassified" links may be assigned the highest possible priority by default, and only those parallel links used when no highest priority is available are classified into Tied priorities with "lower quality".

Claims (7)

1. the method for synchronous of a layering is used for a communication system based on message, and this communication system comprises a plurality of nodes (1-6) that interconnect via all transmission links, and has many parallel transmission links between at least two adjacent nodes, in the method

-Zhu node switching contains all signals of various sync messages, and described sync message comprises a sync mark, and it can point out the priority level of signal in this internal system layer, sync aggregated(particle) structure that sent,

-this node compares all sync marks that received, so that select mark with limit priority synchronisation source as the clock of itself,

It is characterized in that

Has a node in this network at least, come to add mark with all and row labels that indicates the mutual quality order of all parallel links for all parallel links, above-mentioned all parallel links interconnect this node with a given adjacent node, described thus all and row labels is used to this comparison procedure, with the mutual priority orders of all sync marks of determining to receive by described all parallel transmission links.

2. as the defined method of claim 1, it is characterized in that the parallel transmission link of each bar all is assigned with specific, a unique and row labels.

3. as the defined method of claim 1, it is characterized in that,, described parallel transmission link is categorized as requisite number purpose priority class according to their all and row labels.

4. as the defined method of claim 3, it is characterized in that described parallel transmission link is classified as two kinds of different priority class.

5. be used for a node device that uses a kind of communication system of the method for synchronous based on message, described system comprises a plurality of nodes (1-6) that interconnect via transmission link, described node device comprises a concentrated control unit (CU), be used for the selection of synchronisation source being made a strategic decision at this node, also comprise all interface units (IU1-IUN), all interfaces wherein allow this node to communicate with other all nodes of this network, described all node switching comprise all signals of various sync messages, sync message comprises a sync mark, it can point out the priority of signal in this internal system layer, sync aggregated(particle) structure that sent, described sync mark is formed in described control unit (CU), this control unit comprises and is used for device (20) that all sync marks that received are sorted each other, so that select mark with limit priority synchronisation source as this node

It is characterized in that

Described node device comprises device (17 ₁, 17 ₂23), be used for this node is carried out mutual ordering with all parallel links that a given adjacent node couples together, so that obtain a classified order, the classified order that described thus comparison means (20) can use all parallel links is as by the standard of comparison in the comparison procedure of all sync messages that all parallel links received.

6. as the defined node device of claim 5, it is characterized in that described device (17 ₁, 17 ₂) be arranged to the distributed way of a kind of combination corresponding to all interfaces of all parallel links.

7. as the defined node device of claim 5, it is characterized in that described device (23) is arranged to a kind of centralized system of this control unit in conjunction with this node device.

Images