EP0908029A1

EP0908029A1 - Optical network

Info

Publication number: EP0908029A1
Application number: EP97933954A
Authority: EP
Inventors: Magnus ÖBERG; Bengt Johansson; Lars Erik Egnell
Original assignee: Telefonaktiebolaget LM Ericsson AB
Current assignee: Telefonaktiebolaget LM Ericsson AB
Priority date: 1996-07-18
Filing date: 1997-07-11
Publication date: 1999-04-14
Also published as: SE507415C2; CN1135752C; WO1998004058A1; CN1231087A; JP2000515697A; BR9710364A; SE9602806L; SE9602806D0; AU3712997A; TW387169B

Abstract

The invention relates to a wavelength routed optical communication network in a hubbed configuration. The network comprises a hub node (H) and a number of satellite nodes (A, B, C...) connected to two optical fibre means (F1, F2) in an optical ring architecture. Each satellite node works in an individual wavelength channel. The satellite nodes (A, B, C...) are connected to the fibres (F1, F2) by passive multiplexers and demultiplexers (4' to 7') and all kinds of switching in the network is provided in the hub means (H).

Description

Optical network

This invention relates to an optical hub network device, and particularly for a wavelength routed optical network in a hubbed configuration.

Technical Field of the Invention

Optical systems, circuits and fibre networks have become more and more important for data communication and telecommunication systems. Optical fibres have large transmission capacity without electromagnetic interference and ground loop problems.

Optical multichannel systems have an increasing demand, and will probably change network design strategies during the coming years. By using multichannel techniques, increased transmission capacity and flexibility can be realised on existing fibre cables without increasing modulation speed or adding more complex control functions.

In a star network one of the nodes in the system is a central node, called the hub node, and the rest of the nodes are satellite nodes. Each satellite node is able to communicate with any other satellite node but only via the hub node.

A bus architecture can be used for communication networks of this kind. The bus is formed as a ring with the two end nodes connected to the hub. The different satellite nodes in the system are allotted to individual wavelength channels thereby to provide the star features of the network. Several satellite nodes may then share the same fibre where each fibre can bear up to N wavelength channels. Each satellite node always transmits and listen towards both end nodes. The bus carrying normal satellite to hub traffic towards one end node. In case of fibre failure the traffic is re-routed, by the hub, to the other end node and the traffic is, in this way, restored. In this way, each satellite node in the network can be reached in two separate ways from the hub node with just one optical fibre cable comprising two fibres, so if a fibre break occurs in one direction, the traffic can be re-routed to the other direction.

Thus, in case of a cable failure the hub node can receive the incoming signals from certain satellite nodes from one direction and transmit the outgoing signals in the same direction as the incoming signals.

Description of Related Art

An optical ring network is described in an article "Novel Optically Restorable WDM Ring Network", by B. Glance, C. Doerr, I. P. Kaminow, R. Montagne, AT & Bell Laboratories, Crawford Hill Laboratory, Holmdel, New Jersey 07733, Published in OFC '96 Technical Digest. In this network each satellite node is provided with a switching system in order to transmit and receive signals on the network.

Objects of the invention. Summary

It is an object of the present invention to provide a method and a device for providing an optical telecommunication network comprising a hub node and a plurality of satellite nodes and having a simple and inexpensive structure.

Another object of the invention is to provide a method and a device for connecting many satellite nodes and a hub node in an optical logical star network connected in a ring and having as few critical components, i.e. inclined to failure, as possible.

Another object of the invention is to provide a method and a device for making a transmitting and receiving traffic between the hub and each of the nodes even if a cable failure has occurred in an optical network. Still another object of the invention is that cable failure can be a double, a single fibre break or a damage on one or both of the fibres where still some part of the light flowing through the damaged part, still using the same set of wavelength for both upstream and downstream communications.

Yet another object of the invention is to provide a method and a device for keeping a traffic going in an optical telecommunication network comprising a hub node and a plurality of satellite nodes even when a component of the network is damaged.

Still another object of the invention is to provide a method and a device for an optical telecommunication network normally having a network interruption only at the hub node.

At least some of these objects are achieved by a network having the characterizing features in claim 1. Further features and improvements of the invention are given in the dependent claims.

According to the invention satellite nodes are connected to the fibres by passive multiplexers and demultiplexers, for instance comprising fibre couplers. All kinds of switching in the net is provided in the hub means.

Advantages of the invention

When using two fibres for the network communication the theoretical minimum number of wavelength channels could be used, i.e. only one wavelength channel per satellite node connected to the network. No optical switches are used on the fibre ring, they are provided only in the hub node. The satellite nodes could then have an extremely simple design without any required intelligence.

The cable protection could be handled by the hub node.

At least one spare hub node for back-up could be installed at any occasion with no modifications on the already installed hardware.

Brief Description of the Drawings

For a more complete understanding of the present invention and for further objects and advantages thereof, reference is now made to the following description taken in conjunction with the accompanying drawings, in which:

FIG 1 shows schematically a first embodiment of a communication network according to the invention; FIG 2A-2E show schematically different traffic propagations for different kinds of failure of the network; and

FIG 3 shows schematically a second embodiment of a communication network according to the invention.

Detailed Description of Embodiments

Referring to FIG. 1, physically the communication network comprises one central node, called hub node H, and a number of satellite nodes A, B, C... All the nodes are connected by a two fibre ring, having the fibres F 1 and F2, with counter-propagating traffic on the two fibres. The fibres are preferably singlemode fibres. Logically the network is a star network since every satellite node has a unique designated wavelength channel on which it transmits and receives. Each satellite node is able to communicate directly only with the hub node H. Thus traffic between two satellite nodes, such as A and B, always has to go via the hub node H. The hub node transmits and receives on each wavelength channel belonging to each satellite node A, B, C... connected to the network. Thus, each fibre has a number of N wavelength channels in a network having a number of N satellite nodes. The channels in the hub node could then, via the electrical interfaces on its transmitters Tx' and receivers Rx', in a manner known per se also be connected to other communication networks either of the same design as the one described herein or of some other design known in the art.

According to the invention, each satellite node A, B, C... is connected to the fibres by passive multiplexers and demultiplexers in order to avoid switching components. Switching elements need to be controlled individually and are also inclined to failure. As illustrated in the station D in FIG. 1, each satellite node comprises an access point 1 having a transmitter and a receiver (not shown), two l-by-2 fibre couplers 2 and 3, two passive multiplexers 4 and 5, and two passive demultiplexers 6 and 7.

The function of the fibre coupler 3 is to distribute the transmitter signal from the transmitter in the access point 1 to the two multiplexers 4 and 5. The function of the fibre coupler 2 is to bring together the received signals from the two demultiplexers 6 and 7 to the receiver in the access point 1. The transmitter and the received signals lie within the same wavelength channel for each station.

The multiplexers 4 and 5 couple the transmitter signal onto the bus fibres FI and F2. The demultiplexers tap the desired wavelength channel for the satellite node in question out from the bus fibres to the receiver in the work station 1. As apparent from the station C, which illustrates the satellite node design in a simplified form, the multiplexers may consist of simple fibre couplers 4' and 5'. The demultiplexers may be replaced by fibre couplers 6' and 7' and a bandpass filter 8 adapted to the wavelength channel for the station in question.

The hub node means (H) is provided with a fibre break of the fibres FI and F2 and transmitting and receiving means 9, 10, 13 and 11, 12, 14 connected to each end of the fibres FI and F2. For this end the hub node H comprises one light transmitter Tx', such as a wavelength stabilised modulated laser, and one optical receiver per satellite node connected to the network. Each fibre has a deconnection in the hub node. A demultiplexer 10 and a multiplexer 1 1 are connected to each end of fibre FI . In the same way but on opposite sides, a demultiplexer 9 and a multiplexer 12 are connected to fibre F2.

An individual switching unit 13 including a PIN diode and a 2-by-2 optical cross-bar switch is connected to each light transmitter Tx' and the two demultiplexers 9 and 10. The function of the 2-by-2 cross-bar switches on the receiver side is to choose which of the two demultiplexers 9 or 10 the receiver in question should listen to, i.e. from which fibre end its belonging satellite node signal should be provided. The switching unit could of course be replaced by some other switching device having the same function, but cross-bar switch structure with the PIN diode provides a cheap application.

One of the demultiplexers, 9 in FIG 1, is the one in operation when there is no fibre break in the fibre ring. The PIN diode automatically listens to the other demultiplexer 10 on the same wavelength channel, and provides a change-over for providing signal information from the demultiplexer 10 when there is an incoming signal information only on the demultiplexer 10. Thus, the two demultiplexers are connected to the cross-bar switches in the switching units 13 so that a certain switch and thereby a certain receiver/PIN-diode pair always listen to the same wavelength channel, no matter the position of the switch. The satellite nodes always transmits the transmitter signals on both fibres FI and F2 but in different directions, as apparent from FIG 1. The satellite receivers are coupled to listen in both directions even if the same channel never will come simultaneously on the both fibres FI and F2.

In each channel in the hub node, an optical l-by-2 space switch 14 is connected between the transmitter, only indicated by the reference Tx', and the two multiplexers 11 and 12 for transmitting either on the fibre FI or on the fibre F2.

Transmitter/receiver-pairs in the hub node operating on the same wavelength channel and thereby handling the bi-directional communication with a certain satellite node always transmit to and listen from the same direction. In other words, if the receiver switch 13 for a certain channel changes from a first position, below called left, to a second, below called right, the transmitter switch 14 for the same wavelength channel should do the same. Thus, each transmitter switch unit 14 is controlled by the position of the corresponding receiver switch 13.

If a fibre failure occurs, for instance at 15, on at least one of the fibres FI and/or F2 the hub node will lose the incoming signals from at least certain channels either on the receivers or on the PIN diodes. The action from the hub node will then be to reconfigure the transmitter and the receiver switches according to the following two rules.

1) Change the positions of the switches of those channels that have been lost on the receivers.

2) If channels have been lost on the PIN diodes, change the positions of all switches except those that have been lost.

The rules above presume that all switches have the same position, for example to the left. The hub can of course mix left and right communication even if the network is in working order. However, the simple rules for protection described above can not be used in that case. Referring to FIG 2A, if there is no cable failure for any of the channels in the ring architecture the hub node H will only transmit through its left side multiplexer 11 by having all transmitter switches to the left. All the receiver switches 13 are in the bar position so that all communications goes out from and into the hub node H on its left side. Each satellite node is normally designed to only receive and transmit on an individual channel. However, nothing prevents having satellite nodes with several channels communicating with the hub node. Each satellite node will receive one channel from the hub node and transmit the same (or other) wavelength channel in both directions through its multiplexers 4 and 5 (4' and 5'). The signal propagation of the different channels of the satellite nodes A, B, C... has been illustrated by the same node reference but in small letters. Only those signals multiplexed onto the clockwise propagating bus fibre will reach the proper receiver in the hub node, whereas those multiplexed on the other fibre will reach the corresponding monitor PIN diode in the switching unit 13 in the hub node.

On the counter clockwise propagating bus fibre there will accumulate more and more "unnecessary channel signalling". These are however separated from the useful one by wavelength filtering in the bandpass filter 8 in the satellite nodes.

Also, the same information from the same satellite nodes could be provided twice in the propagation direction on the fibres FI and F2 when it reaches the hub node on its right side. However, this will not cause any problems since the hub node is adjusted to receive only information signals coming from the left side.

Referring to FIG 2B illustrating that a double cable break has occurred between the satellite nodes C and D. In that case the hub node H loses the incoming channels of the satellite nodes A, B and C. The action of the hub node is to change the receiver switches for those channels from bar to cross state and the transmitter switches of the same channels from the left to the right position. Full communication is restored, which according to FIG 2B means that the hub node communicates with nodes E and D on the left incoming and outgoing bus fibres as before and with the satellite nodes A, B and C on the right bus fibres, as illustrated by the node references in small letters. (Even without a cable break, this could of course be the normal way of operating the network.)

In case of a single fibre break on the clockwise fibre between the satellite nodes C and D the hub node looses the incoming channels a, b, and c. The action of the hub node will be the same as for the double cable break, as illustrated in FIG 2C.

In case of a single fibre break on the counter clockwise fibre between the satellite nodes C and D the hub node will not lose any incoming traffic channels, as apparent from FIG 2D showing the state in the network just when the failure has occurred.

However, it will lose the channels d and e on the PIN-diodes in the receiver switching devices 13 monitoring the clockwise propagating fibre. It will therefore change all switches except those that not have been lost on the PIN-diodes. The action will be the same as in FIG 2B, i.e. change of the receiver switches 13 of channels a, b, and c from bar to cross state and the transmitter switches 14 of the same channels from the left to the right position. The channel signal propagation after this switching is shown in FIG 2E.

Since all traffic passes through the hub node H, this is a very sensitive part of the network. Therefore, in accordance with a second embodiment shown in FIG 3, in which the elements in FIG 1 having the same function have got the same references, a spare hub node SH could be placed anywhere along the bus fibre ring and is shown inserted between the satellite nodes D and E. This spare hub SH is illustrated as a simple kind of hub node having the same kind of connection to the two fibres FI and F2 as the satellite nodes A, B, C..., i.e. the fibre chain should not be broken by the spare hub node SH, only by the hub node H. As in the ordinary hub node H the spare hub node SH has as many transmitters and receivers as there are satellite transceivers (normally equal to the number of satellite nodes) connected to the fibre ring. However, it has only one multiplexer 17 and one demultiplexer 18, each connected to both fibres FI and F2 by fibre couplers 4" and 5" in the same way as for each satellite node.

In the normal case, when the ordinary hub node H is operational, the transmitters Tx" in the spare hub node SH are off. The spare hub node is then totally transparent and does not affect the traffic passing on the network except for some power losses caused by the fibre couplers 4", 5", 6" and 7". Optical amplifiers could of course be installed anywhere in the network (not shown).

The spare hub node SH is activated as soon as a failure of the hub node H occurs. When it has been activated, the signals from the transmitters Tx" are sent out in both directions through the fibre couplers 4" and 5". The receivers Rx" listen simultaneously on both bus fibres F I and F2. If no other failure in the network has occurred at the same time as the failure of the hub node H, which should be rare, all satellite nodes have full communication with the spare hub node SH. The signals from the satellite nodes arrives to the spare hub node SH either from the left or from the right. All signals from the spare hub node SH are sent out in both directions.

Some of those sent to the right are received by the satellite nodes and the rest are received from the signals sent out to the left. The signals are prevented from circulation all around the fibre ring due to the design of the ordinary hub node H including a fibre break.

If a fibre break occurs when the spare hub node SH is operational then it can communicate with all the satellite nodes except for the ones placed between the break and the ordinary hub node H.

Even though it is described above that each satellite node uses the same wavelength channel for both its transmitted and received traffic it is within the scope of invention to have separate wavelength channels for both without changing the network. It is then possible to change the network so that only one fibre for both transmitted and received traffic is used. However, using the same fibre for both downstream and upstream traffic implies that implemented optical amplifiers need to operate in both directions. This is rather cumbersome, and therefore an application of this kind will be used only in very rare occasions.

While the invention has been described with reference to specific embodiments, it will be understood by those skilled in the art that various changes may be made and equivalents may be substituted for elements thereof without departing from the true spirit and scope of the invention as it is defined in the accompanying claims. In addition, modifications may be made without departing from the essential teachings of the invention as defined in the claims.

Claims

1. A wavelength routed optical communication network in a hubbed configuration comprising a hub node means (H) and a number of satellite node means (A,B,C.) connected to optical fibre means (FI, F2) in an optical ring architecture, wherein the satellite node means (A, B, C. ) are connected to the fibre means (FI F2) by passive multiplexers and demultiplexers (4' to 7') and all kinds of cable protection switching in the network is provided in the hub node means (H).

2. A network according to claim 1, wherein, in the satellite node means, the multiplexers comprise two fibre couplers (4' and 5') by which it is connected to the fibre means (FI, F2).

3. A network according to claim 1 or 2, wherein in the satellite node means the demultiplexers comprise two fibre couplers (6' and 7') by which it is connected to the fibre means (FI, F2) and a bandpass filter (8) adapted to the wavelength channel for the satellite node means in question.

4. A network according to any one of the preceding claims, wherein each satellite node comprises an access point (1) having a satellite node transmitter and a satellite node receiver, the two passive multiplexers (4 and 5), the two passive demultiplexers (6 and 7), one coupler (3) distributing the transmitter signal from the satellite node transmitter to said two multiplexers, and one coupler (2) bringing together the received signals from the two demultiplexers (4 and 5) to the satellite node receiver in the work station ( 1 ).

5. A network according to any one of the preceding claims, wherein the hub node means (H) is provided with a fibre break and with transmitting and receiving means (9, 10, 13 and 11, 12, 14) connected to each end of the fibre means.

6. A network according to any one of the preceding claims, wherein the hub node means (H) comprises at least one optical transmitter (Tx'), for instance a wavelength stabilised laser, and at least one optical receiver (Rx') per satellite node means (A, B, C.) connected to the network.

7. A network according to claim 6, wherein the fibre ring comprises two fibres and each fibre has a deconnection in the hub node means, the hub node means having a demultiplexer (9 or 10) connected to one end of each fibre (FI or F2) and a multiplexer (11 or 12) connected to the other end, the demultiplexer of one of the fibres of the two being connected to the same side of the fibre as the multiplexer of the other fibre of the two.

8. A network according to claim 6 or 7, wherein, for each light receiver (Rx') in the hub means (H), an individual switching unit (13) is connected to it and controllably connectable to one or the other of the two demultiplexers (9 or 10) and normally connected to a predetermined one of the demultiplexers (9 or 10) when the fibre ring is in working order and changed over to the other one of the demultiplexers (10 or 9) at occurrence of a failure of the fibre ring influencing the signalling to the light receiver (Rx') in question.

9. A network according to claim 6 or 7, wherein for each light transmitter (Tx') in the hub node means (H), an individual switching unit (14) is coupled to the light transmitter and controllably connected to one or the other of the two multiplexers

(11 or 12) and normally connected to a predetermined one of the multiplexers (1 1 or 12) when the fibre ring is in working order and connected to the other one of the multiplexers (12 or 11) at occurrence of a failure of the fibre ring influencing the signalling of the light transmitter (Tx') in question.

10. A network according to claim 6, and 8 or 9, wherein the switching units of the light transmitter and the light receiver in the hub node means related to the signalling of the same satellite node are connected in the same sense and have their states changed at the same occasion.

11. A network according to any of the claims 8 to 10, wherein the receiver switching units in the hub node means is automatically listening to a second of the hub node means demultiplexers (10), and in case of a fibre failure, the hub node means reconfigures the transmitters and the receiver means according to the following rules^*.

1) change the positions of the switching means in the hub node means for those channels that have been lost on the hub node receivers.

2) if channels have been lost on the channels to which the switching units are automatically monitoring, change the positions of all the hub node switching means except for those that have been lost.

12. A network according to any one of the preceding claims, wherein the transmitter and the received signals lie within the same wavelength channel for each satellite node means.

13. A network according to any one of the preceding claims, wherein a spare hub node (SH) can be connected anywhere along the bus fibre ring and controllable into action by a failure of the hub node (H).

14. A network according to claim 13, wherein the spare hub node (SH) comprises only one multiplexer (17) and one demultiplexer ( 18), each connected to the fibre means (FI and F2) by fibre couplers (4" to 7").