A teletransmission network, network elements therefor and a method of identifying the synchronization of a network element The invention relates to a teletransmission network com- prising a plurality of network elements, each of which is adapted to emit and/or receive data signals, and of which one or more may serve as a synchronization source for other network elements, the data rate of the signals emitted from the synchronization source being decisive for the data rate of the signals emitted from these other network elements. The invention moreover relates to net- work elements for such a teletransmission network and to a method of identifying whether a network element in a teletransmission network is synchronized to a given syn- chronization source and use of the method in an SDH sys- tem.

In synchronous data communication networks, e. g. SDH net- works (Synchronous Digital Hierarchy), advanced control of the synchronization mechanisms takes place. This con- trol is based on a synchronization hierarchy so that one of the nodes (network elements) in the network is se- lected as the primary reference for other nodes, which may in turn serve as a reference for other nodes. Nor- mally, no clock signals proper are transmitted in the network. When a node is synchronized to a reference node, this therefore takes place by extracting a clock signal from the data signals received from the reference node, said clock signal being then used in the node concerned partly for internal control and partly for synchroniza- tion of the data signals emitted from the node.

Where several networks are interconnected, e. g. by inter- connection of a plurality of national networks to estab-

lish a common international network, there will typically be several primary references. Several parallel synchro- nization hierarchies will thus be provided. In SDH termi- nology, this normally occurring interconnection is re- ferred to as a pseudo-synchronous mode of the network.

Although the individual primary references are specified within quite narrow tolerances, the tolerances may cause significant differences in transmission rates, when data are to be transmitted between two networks synchronized to various primary references. This and other reasons may have as a result that nodes in a single or several net- works lose their synchronization with the upper layers in the synchronization hierarchy. In SDH terminology this state of lost synchronization is referred to as a plesio- chronous mode. In the plesiochronous mode, a node has selected its own internally generated synchronization to which network elements on a lower layer in the synchroni- zation hierarchy may synchronize. This internally gener- ated synchronization may be based on a free running clock signal, and the node is said to be in a free running mode. If the clock signal is based on the clock signal received last by the network element, the mode is called holdover.

Information on the set-up of the synchronization hierar- chy, i. e. how the individual network elements are syn- chronized, will normally be known in a form of monitoring system (management system). If this information is not easily accessible, is lost or changes occur, e. g. if the whole or part of the network is in a plesiochronous mode, it will be expedient to be able to provide this information by performing measurements on the system. For example, it is conceivable that it is desired to check at one end of an SDH network that the synchronization comes from the right source at the other end of the country,

and that a holdover or free-running mode has not been as- sumed somewhere en route.

It is known to check the synchronization state of a net- work element by using a so-called test set, which is a device that can be connected to a synchronization source, and which causes the data signals from the synchroniza- tion source to be transmitted in the network with a fixed offset value in frequency, and then it may be checked by means of another device, e. g. a frequency counter which is coupled to a given other network element, whether the signals received in this network element also have this offset value in frequency. If this is the case, the net- work element has been synchronized to the synchronization source concerned.

This, however, has the drawback that staff and equipment have to be sent out to the synchronization source as well as to the network element whose synchronization is to be measured, and this is therefore a very costly solution.

In particular, it is inexpedient that the measurement cannot be activated or read from an overall management or monitoring system, but just locally by means of said equipment.

Accordingly, an object of the invention is to provide a teletransmission network of the type mentioned in the opening paragraph which can be activated directly from an overall monitoring system.

This is achieved according to the invention in that one or more of the network elements capable of serving as a synchronization source are adapted to vary the data rate of the emitted signals.

When it is the synchronization itself that can vary the data rate, a much simpler solution is achieved, capable of being controlled directly from a monitoring system and capable of being implemented whenever it is desired, without first having to send staff and equipment out to the individual network elements.

When, as defined in claim 2, at least one of the other network elements comprises means for detecting whether the data rate of the received signals varies in the same manner as said variation of the signals emitted from the synchronization source, it is ensured that the entire measurement can be controlled from the monitoring system, as this can vary the data rate from the synchronization source and also detect in the other network elements whether this variation can be found there.

When, as stated in claim 3, the network element or ele- ments capable of varying the data rate are adapted to produce said variation by modulating a frequency corre- sponding to said data rate with a signal of low frequency with respect thereto, it is additionally ensured that it is possible to vary the frequency in a manner which does not build up a constantly increasing phase difference relative to other frequencies which are not subjected to a corresponding variation. That the variation is of low frequency moreover ensures that it may be transferred un- obstructedly through the network to the network element whose synchronization conditions it is desired to check.

A variation of higher frequency might be attenuated by low-pass filters in the network elements which are passed en route.

In an expedient embodiment, which is defined in claim 4, the network element or elements capable of varying the data rate are adapted to perform said variation under re-

mote control from a monitoring system. Correspondingly, it is expedient that said means for detecting whether the data rate of the received signals varies in the same man- ner as said variation of the signals emitted from the synchronization source, are adapted to be remote-con- trolled and monitored from a monitoring system. This em- bodiment is defined in claim 5.

When, as stated in claim 6, the low-frequency signal is a pseudo random sequence, a variation is obtained, which ensures that no phase difference is built up on average, and which is easily recognized in the network element or elements in which the variation is measured, as these may be adapted to recognize precisely such a sequence.

The invention also relates to a network element for use in such a teletransmission network and adapted to operate as a synchronization source. As stated in claim 7, the network element is moreover adapted to vary the data rate of the emitted signals.

When it is the synchronization source itself that can vary the data rate, a simpler solution is achieved, as mentioned above, which can be controlled directly from a monitoring system, and which can be implemented whenever it is desired, without first having to send staff and equipment out to the individual network elements.

Claims 8-10 define expedient embodiments of this network element with the advantages described above.

The invention moreover relates to a network element for use in such a teletransmission network and adapted to be synchronized to signals which are received from a syn- chronization source via the network. As stated in claim 11, the network element moreover comprises means for de-

tecting whether the data rate of the received signals varies in a manner given beforehand.

When the individual network elements themselves can de- tect a variation in the data rate, it is correspondingly ensured that the described measurements, with respect to detecting the synchronization source of a network ele- ment, can be performed directly, without having to mount special equipment in or at the network element.

Claims 12-14 define expedient embodiments of this network element with the advantages described above.

As mentioned, the invention also relates to a method of identifying whether a network element in a teletransmis- sion network of the described type is synchronized to a given synchronization source. The method is described in claim 15. When the data rate of said signals from the synchronization source is varied, and it is checked in said network element whether the data rate of the re- ceived signals varies in the same manner and then, if this is the case, the network element is considered to be synchronized to said synchronization source, a much sim- pler solution is achieved, as described above, which can be controlled directly from a monitoring system, and which can be implemented whenever it is desired, without first having to send staff and equipment out to the indi- vidual network elements.

Claims 16-18 define expedient embodiments of the method with the advantages described above.

Finally, as stated in claims 19-20, the invention relates to a use of said method in an SDH system and a SONET sys- tem, respectively. The SDH system is extremely wide- spread, and the mentioned advantages will be particularly

distinct in precisely an SDH system. SONET is a North American version of SDH.

The invention will now be described more fully below with reference to the drawing, in which fig. 1 shows a synchronization hierarchy for a synchro- nous data communications network, fig. 2 shows an example of how a network element in the synchronization hierarchy of fig. 1 can lose the synchro- nization, fig. 3 shows two parallel synchronization hierarchies which are synchronized from their respective synchroniza- tion sources, fig. 4 shows a transmitter which is adapted to transmit communications signals and modulate the data rate accord- ing to the invention, fig. 5 shows an example ouf hou a simple modulation signal may be designed, and fig. 6 shows an example of a network element which is adapted to detect a modulation of the data rate.

Fig. 1 shows an example of the structure of a synchroni- zation hierarchy of a synchronous data communications network. The example involves an SDH network with a plu- rality of network elements, of which the seven elements 1-7 are shown in the figure. One of the network elements 1 is here selected as a primary reference, so that the clock signal (KPRC) from there is distributed down through the hierarchy to the other network elements 2-7.

The network elements are illustrated as circles in which

the synchronization source of the network element is in- dicated. Although the data flow in the communications network does not necessarily follow the same structure, it should be stressed that, in many situations, the clock signals are not distributed as clock signals proper, but can be extracted from the received data signals in the individual network elements. For example, the network element 1 transmits data with the clock frequency KPRC to the network element 2. This element can then regenerate KPRC on the basis of the received data and use it for the data which are transmitted to the network elements 4 and 5. It will thus be seen that in the example shown all the network elements are synchronized to the frequency KPRC which originates from the element 1. The arrows show how the synchronization is transferred between the network elements.

Fig. 2 shows the same network as in fig. 1, but in a situation where one of the network elements, in this case the element 3, has lost the synchronization to the pri- mary reference KPRC, e. g. because of a drop-out of the connection between the elements 1 and 3, as is shown in dashed line. Then, the network element 3 instead uses a self-generated clock signal KNE3 for the data which it passes further on in the system, and the network elements 6 and 7 further down in the hierarchy will therefore be synchronized to this clock signal instead of to the sig- nal KPRC like before. In SDH terminology, the lost syn- chronization means that the network element 3 assumes a free running or holdover mode in dependence on whether its internal clock signal is entirely free running, or whether it was locked to the primary clock signal immedi- ately before the drop-out of the latter. The mode in fig.

2 is also called a plesiochronous mode. If a network ele- ment loses the synchronization to a primary synchroniza- tion source KPRC, then it will not be readily possible in

the network elements synchronized to this network element to detect the source to which the synchronization was ac- tually made. Thus, there is a need for a facility allowing it to be detected in a simple manner, e. g. from a monitoring system, whether a given network element is synchronized to KPRC or to another clock signal, such as e. g. KNE3. This is made possible with the invention.

Fig. 3 shows an example where several networks are inter- connected so that one or more network elements are con- nected to both networks. It may e. g. be a plurality of national networks which are interconnected to establish a common international network, and which will typically have a primary reference each. In the example, the net- work elements 8 and 14 are primary references and dis- tribute the clock signals KPRC1 and KPRC2 in the two net- works. It appears that the network elements 9-13 are syn- chronized to KPRC1, while the network elements 15-19 are synchronized to KPRC2. The network element 20, however, is connected to both networks, and it is thus uncertain whether it is synchronized to KPRC1 or KPRC2, as it re- ceives both signals. Thus, there is a need for a facility allowing it to be detected in a simple manner, e. g. from a monitoring system, whether the network element 20 is synchronized to KPRC1 or KPRC2. This need, too, may be satisfied by the invention.

Fig. 4 shows a network element 21 which is adapted to transmit ordinary communications signals with a frequency which is superposed by an easily recognizable signal ac- cording to the invention. This network element therefore lends itself for use as a synchronization source accord- ing to the invention. The idea is that this superposed signal is applied when it is desired to detect whether a given other network element is synchronized to this syn- chronization source. When the frequency is modulated, it

is simultaneously checked in the given other network ele- ment whether the clock signal received or regenerated there is modulated with the same signal. If this is the case, it is known that synchronization has been estab- lished to this synchronization source.

Fig. 4 shows how more or less advanced processing of data is performed in a known manner in such a network element 21, which causes the data having been subjected to proc- essing to arrive in an irregular stream. It is thus well- known to have a buffer 23 interposed between a processing unit 22 and an output module 25, which may e. g. be in the form of an electrical/optical converter. Outputting of data from the buffer 23 is controlled by phase-locked loop. This phase-locked loop consists of a phase/frequency detector 26, which measures the differ- ence between a local synchronization clock and an output clock signal to the buffer 23. The output signal from the phase/frequency detector 26 is filtered in the filter 27, and the filtered signal controls-in the form of a con- trol signal-the frequency of the oscillator 28. This frequency controls the transmitted data in the multi- plexer 24. This results in a jitter-free data rate out of the network element 21. This is prior art.

The above-mentioned modulation of the clock frequency may be performed e. g. by means of a modulator 29 which is in- serted in the connection between a local synchronization clock LS and the phase/frequency detector 26. The modula- tor is controlled by the signal MODULATION, which is se- lected so that it is easily recognizable in other network elements, and so that it does not cause an increasing phase difference relative to the unmodulated clock sig- nal. The signal MODULATION may e. g. be a pseudo random sequence, which is also called a PRBS (Pseudo Random Bit Sequence) signal, since this frequency is simple to de-

tect and is frequently implemented beforehand in the transmission equipment for use for other test purposes.

The activation of the signal MODULATION may be controlled from an overall control and monitoring system, which can thereby connect the signal in a situation where it is de- sired to check the synchronization state of a network element further down in the synchronization hierarchy. It will thus not be connected in normal operation.

The ability to suppress fluctuations in the data rate caused by the internal processing functions of the net- work element depends primarily on the size of the buffer 23 and the band width of the loop. The band width of the phase-locked loop is determined so that it suppresses rapid variations in the data flow, and so that it allows relatively small and slow variations in the data flow to pass out of the transmitter. This means that precisely the small and slow variations caused by the PRBS sequence will escape through the filter 27, thereby producing a corresponding variation in the frequency of the oscilla- tor 28. The frequency of the transmitted data, i. e. the data rate, will therefore have a corresponding variation which can be detected in another network element receiv- ing these data. Owing to the corresponding low-pass fil- ters in the other network elements which the signal has to pass en route in the network, the frequency of the modulation will normally be below 5 Hz. It should be noted that frequency deviations caused by the modulation are so small that the specifications of the SDH system are still observed.

Fig. 5 shows an example of how the signal MODULATION may modulate the local synchronization signal LS around its nominal frequency N. The modulation depth depends on the requirements made with respect to the data rate stability

of the signal which is transmitted by the network ele- ment. The length of the period during which the modulated local synchronization signal is either greater or smaller than the nominal frequency of LS may be adjusted so that the equipment to detect the control and monitoring sig- nals can detect them correctly. Furthermore, the size of the buffer 23 must be adjusted to the deviation from the nominal frequency as well as the mentioned period during which the frequency of the modulated local synchroniza- tion signal is either greater or smaller than the nominal frequency of LS. The signal shown in the figure may e. g. be the beginning of a PRBS sequence. This sequence is ar- ranged such that the frequency of the modulated signal will on average be equally much above and below the nomi- nal frequency so that the average data rate is not changed.

Fig. 6 shows a portion of the receiver part of a network element 31 which is capable of receiving data signals and of detecting whether a variation occurs, as described above, in the data rate of the signals received. Most of the figure shows a known receiver circuit in the same manner as fig. 4. The input module 32 receives optical signals from a fibre and converts them into corresponding electrical signals which are fed further on to the clock regeneration circuit 33. Here, a clock signal is ex- tracted from the received data signals. The extracted clock signal is transmitted to a selector circuit 35, which will be mentioned more fully below, while the data signals are transmitted to the demultiplexer 34. The de- multiplexed data are placed in the buffer 36, from which they may be transferred to the processing unit 37 at the rate at which this unit can receive them.

Two additional buffers are shown behind the buffer 36.

The additional buffers symbolize that the network element

can receive data signals from several lines which have an input circuit each. Thus, a clock signal may be extracted from each line, as these are not necessarily synchronized to the same synchronization source, and each of these clock signals is therefore fed to the selector 35 which selects the input to which the network element is to be synchronized. If no incoming data signals are received, an internally generated clock signal INT may be selected instead, which means that the network element is in free running or holdover mode. The selected clock signal as the signal CLK is then fed to the buffer 36 and the proc- essing unit 37, and it is this clock signal which con- trols the data rate of the data which are transmitted further out in the network from this network element.

The clock signal CLK, which is fed from the selector 35 to the buffer 36 and the processing unit 37, is moreover fed to a demodulator circuit 38 which is adapted to de- modulate the low-frequency variation in the clock signal CLK described above. The actual demodulation may take place in a known manner, e. g. by means of inter alia a low-pass filter. If the network element is synchronized to a synchronization source where the data rate is momen- tarily modulated as shown in fig. 4, a signal correspond- ing completely to the signal MODULATION in fig. 4 will occur on the output of the demodulator circuit 38. This signal is fed further on to a likewise well-known detec- tor circuit 39 for the mentioned PRBS sequence. A logical signal indicating whether the PRBS sequence is present or not will occur on the output of the detector circuit.

This signal may be used locally in the network element, or it may fed further on to the overall monitoring sys- tem. In the latter case, the synchronization state of the network element may thus be monitored directly from the monitoring system.

If the modulation of the data rate is both activated and detected from a control and monitoring system, the syn- chronization hierarchy of a network may be determined from a central position in spite of the fact that certain network elements are in a mode which is not immediately known by the control and monitoring system, e. g. holdover or free running mode of an SDH system.

Although a preferred embodiment of the present invention has been described and shown, the invention is not re- stricted to it, but may also be embodied in other ways within the scope of the subject-matter defined in the following claims.