FR2988542A1 - Merging device for merging optical components for e.g. optical access network, has combination element to obtain blocking signal by combining duplicated optical components other than optical component for merging - Google Patents

Abstract

The device (20) has an optical element (31) arranged to switch to a blocking position for blocking an optical component for merging, as a function of a blocking signal. A duplication element (41) is arranged to duplicate the optical component for merging for sending to a combination element (61), where the combination element is arranged to obtain the blocking signal by combining duplicated optical components other than the optical component for merging. A merging module (70) is arranged to merge the optical components output by the optical element. Independent claims are also included for the following: (1) an optical combiner for optical signals (2) an optical node for aggregating optical signals (3) a method for merging optical components associated with one wavelength into an optical component associated with the wavelength (4) a method for optically combining optical signals.

Description

The invention relates to the field of optical transmission networks and relates to devices for fusing a plurality of optical components associated with a length. waveform into an optical component associated with the same wavelength in these networks. The targeted scope is that of optical switching equipment, in particular for optical access networks and data centers ("Data Center" in English).

There are two types of switching equipment in an optical network: - electronic routers or switches, allowing the waiting of packets in case of congestion; optical switches which do not deal with packet-level traffic but which make it possible to establish permanent optical circuits between two routers or electronic switches.

In order to avoid an electronic processing of each packet, which is expensive in energy, and to increase the switching speeds, the document entitled "Optical burst switching (OBS) - A new paradigm for an optical Internet" by C. Qiao et al, published in the Journal of High Speed Networks, 8: 69-84, 1999, proposes to aggregate the packets in bursts and to switch the bursts optically. However, a signaling packet must be sent beforehand to configure the optical switches before the arrival of the burst. The performance of this method in the use of resources is not optimal, the period between the signaling packet and the burst being generally not usable to switch other gusts. O'Mahony et al's "Application of Optical Packet Switching in Future Communication Networks", published in the IEEE Communications Magazine of March 2001, proposes an optical switching method, in which packets are aggregated into an optical packet according to their destination and quality of service constraints. An optical packet includes a label indicating in particular the destination thereof. An optical switch queues the optical packet to read the tag and configure the switch table and then switches the optical packet. The processing of the label is electronic. This method is not mature for the moment because the performances of the available optical memories are very limited. One of the aims of the invention is to remedy the shortcomings / disadvantages of the state of the art and / or to make improvements thereto. According to a first aspect, the invention relates to a device for fusing a plurality of optical components associated with a wavelength into an optical component associated with said wavelength, the device comprising: for each optical component of said plurality to be fused, - optical means arranged to switch to a blocking position of said optical component to be fused according to a blocking signal; - Duplicating means, arranged to duplicate said optical component to merge to combining means; the combining means, arranged to obtain said blocking signal by combining the duplicate optical components of the plurality, excluding said optical component to be fused; and merging means arranged to merge the optical components at the output of the optical means.

The invention originates from the finding made on existing methods of all-optical switching. In this context, an alternative has been sought and a need for a device capable of fusing optical components associated with a wavelength into an optical component associated with the same wavelength and capable of managing collisions has been identified. . Optical component is here called an optical signal carried by a given wavelength.

The fusion device according to the first aspect is remarkable in that it makes it possible to merge several optical components associated with the same wavelength, while managing the collisions. In the event of simultaneous presence on the input ports of the device of two optical components, at most one of them is present at the output of the device. The collisions at the input of the fusion device are managed without requiring to assign a wavelength to each of the sources of the optical components. In this fusion device, only elementary operations in the optical domain are performed using currently available components. The various modes or features of realization mentioned hereafter can be added independently or in combination with each other, to the fusion device according to the first aspect.

In a first particular embodiment of the device, the optical means are furthermore arranged to amplify the optical component and the duplication means are connected at the output of the optical means. In this case, during an input collision of the device, the optical component, which has been present at the first input of the device, is transmitted. A second optical component later on another input port of the device is blocked, as long as the first optical component is present. The operation of this first embodiment is based on the "first come, first served" principle. In a second particular embodiment of the device, the optical means are further arranged to amplify the optical component and the duplication means are connected at the input of the optical means. In this case, during an input collision of the device, none of the optical components is present at the output of the device.

According to one particular characteristic, the fusion device further comprises, for each optical component of said plurality to be fused, delaying means arranged to apply a predetermined fixed duration delay to said optical component to be fused and connected at input of the associated blocking means said optical component to be merged.

This makes it possible to configure the optical means in the adapted position, before the arrival of the optical component. In a third particular embodiment, in the blocking position, the optical means deflect said optical component to be fused to injection means, arranged to reinject said optical component in an optical fiber to a source node.

Such a device thus makes it possible to signal to a source node that a collision has occurred and that, consequently, the transmitted data have been suppressed, because of the suppression of the corresponding optical component. The source node thus notified can re-issue the data that has been the subject of a collision. The merging device in this third embodiment thus makes it possible to manage the collisions while notifying the sending source nodes of the possible data losses. It is emphasized here that this is done entirely optically and thus without use of signaling message or electronic processing of optical components. According to a particular characteristic, a control unit forms the combination means.

This control unit makes it possible to improve the management of the blocking signals and to avoid the transmission of incomplete packets. The fusion device according to the first aspect is intended to be integrated in an optical combiner according to a second aspect. The invention thus also relates to an optical optical signal combiner, an optical signal comprising optical components respectively associated with a plurality of wavelengths, said optical combiner comprising: demultiplexing means respectively associated with an optical signal input, arranged to obtain from the input optical signal a plurality of optical components; a plurality of fusion devices according to the first aspect, respectively associated with a wavelength and connected at the output of the demultiplexing means; multiplexing means, arranged to multiplex the optical components at the output of the fusion devices into an optical output signal. This optical combiner thus makes it possible to multiplex in the time domain several optical signals, comprising optical components, into a single optical signal at the output, eliminating on each wavelength the packets causing collisions. The optical combiner is dynamic because it does not require the assignment of a wavelength to each of the sources of the optical signals. The utilization of the capacity of the optical fibers is thus improved.

The optical combiner is also dynamic, in that it is possible to merge only a part of the optical components forming the optical signal. The optical combiner according to the second aspect provides a simple and effective solution to the problem of optical packet switching. Indeed, the combiner simply combines the optical signals: it does not perform a switching operation itself. Switching is done by wavelengths: for example, a single output port is associated with each wavelength on each input port. The switching table is therefore static, but the combination of flows (when the flows of several input interfaces converge to the same output interface) is dynamic, achieved by means of the invention.

In this optical combiner, only elementary operations in the optical domain are performed using currently available components. This also makes it possible to implement it at the packet level on a current volume of traffic. Switching is done at the packet level, without the need for packet aggregation. The use of optical network resources is thus optimized. The optical switching of a packet is performed in a time compatible with the current switching constraints, that is to say of the order of the duration of a packet, ie a duration of the order of one microsecond . In addition, no electronic packet processing is required, which limits the power consumption of the combiner. The optical combiner does not require exchange of signaling traffic, prior to the transmission of a packet. The utilization of the capacity of the optical fibers is thus improved. Furthermore, the optical combiner does not require packet aggregation or queuing of the optical packet. The optical combiner according to the second aspect is intended to be integrated in an optical optical signal aggregation node according to a third aspect.

The invention thus also relates to an optical optical signal aggregation node, an optical signal comprising a plurality of optical components respectively associated with a plurality of wavelengths, said node comprising: an optical signal optical combiner according to the second aspect, arranged to combine optical signals respectively received from source optical nodes into an optical signal; a device for distributing an optical signal, arranged to distribute a received optical signal to said source nodes. According to a fourth aspect, the invention also relates to a method of fusing a plurality of optical components associated with a wavelength into a fused optical component associated with said wavelength, comprising: a step of duplication of the components optical plurality of said plurality to merge, during which said optical components to be merged are duplicated; a step of obtaining blocking signals, in which a blocking signal associated with an optical component to be fused is obtained by combining the duplicate optical components of the plurality, with the exception of said optical component to be fused; a blocking step, during which an optical component to be fused is blocked as a function of the associated blocking signal; a step of obtaining the fused optical component associated with the wavelength from an unblocked optical component. According to a fifth aspect, the invention also relates to a method for optically combining optical signals, an optical signal comprising optical components respectively associated with a plurality of wavelengths, comprising: a step of demultiplexing the input optical signals, to obtain from an input optical signal a plurality of optical components; an implementation of the fusion method according to the fourth aspect, for the optical components associated with a wavelength obtained in the demultiplexing step from the input optical signals; a step of multiplexing the output optical components obtained by implementing the fusion process into an output optical signal. Other features and advantages of the invention will emerge on examining the detailed description below, and the accompanying drawings in which: - Figure la shows schematically an optical combiner according to a particular embodiment of the invention; FIG. 1b schematically represents an optical combiner according to another particular embodiment of the invention; FIG. 2 diagrammatically represents a fusion device according to a particular embodiment of the invention; FIGS. 3a, 3b and 3c show three operating states of the fusion device according to a particular embodiment of the invention; FIG. 4 represents an optical aggregation node in its environment according to a particular embodiment of the invention; FIG. 5 illustrates the steps of a method of fusing a plurality of optical components according to a particular embodiment of the invention; FIG. 6 illustrates the steps of a method for optically combining optical signals according to a particular embodiment of the invention; - Figure 7 schematically shows a fusion device according to a particular embodiment of the invention.

In Figure la is schematically shown an optical combiner 110 according to a particular embodiment of the invention. More precisely, the optical combiner 110 makes it possible to combine N optical signals at the input into an optical signal at the output. In the particular example of FIG. 1a, three optical signals are combined. This example can easily be transposed to other configurations as a function of the number N of optical signals to be combined. The first optical signal is received via a first optical fiber F01, connected to the optical combiner 110 on a first input port. The second optical signal is received via a second optical fiber F02, connected to the optical combiner 110 on a second input port. The third optical signal is received via a third optical fiber F03, connected to the optical combiner 110 on a third input port. The optical signal at the output of the optical combiner 110 on an output port is transmitted through a fourth optical fiber F04. Each of the input and output optical signals comprises a plurality W of optical components, each optical component being associated with a given wavelength. In the particular example of FIG. 1a, four optical components form an optical signal. No limitation is attached to the number of optical components forming the optical signal. In addition, only a subset of optical components forming the optical signal can be combined. On the other hand, the numbers N of optical signals and W of optical components can be chosen independently of one another.

The first optical signal at the input is demultiplexed into four wavelengths X1, X2, X3, X4 by a first demultiplexer 11. At the output of this first demultiplexer 11, four optical components X1,1, X2,1, X3,1, X4,1 of the first optical signal are obtained. The second input optical signal is demultiplexed into four wavelengths X1, X2, X3, X4 by a second demultiplexer 12. At the output of this second demultiplexer 12, four optical components X1,2, X2,2, X3,2, X4,2 of the second optical signal are obtained. The third input optical signal is demultiplexed into four wavelengths X1, X2, X3, X4 by a third demultiplexer 13. At the output of this third demultiplexer 13, four optical components X1,3, X2,3, X3,3, 24.3 of the third optical signal are obtained. The optical components of the first, second and third optical signals are then grouped by wavelength X1, X2, X3, X4 and transmitted at the input of fusion devices 20f, 202, 203, 204. The fusion devices according to the invention allow a plurality of optical components to be combined into a resulting optical component such that collisions between optical components at the input of the fusion device are avoided. At most one of the optical components generating a collision is selected.

More specifically, the optical components X1,1, X1,2, X1,3 are transmitted at the input of the first fuser 201; the optical components X2,1, X2,2, X2,3 are transmitted at the input of the second fusion device 202, the optical components X3,1, X3,2, X3,3 are transmitted at the input of the third fusion device 203; the optical components X4,1, X4,2, X4,3 are transmitted at the input of the fourth fusion device 204. At the output of the first fusion device 201, a first optical component X1,4 associated with the wavelength Xi is obtained. At the output of the second fusion device 202, a second optical component X2,4 associated with the wavelength X2 is obtained. At the output of the third fusion device 203, a third optical component X3,4 associated with the wavelength λ3 is obtained. At the output of the fourth fusion device 204, a fourth optical component X4,4 associated with the wavelength kt is obtained.

The four optical components X1,4, X2,4, X3,4, X4,4 respectively at the output of the fusion devices 20f, 202, 203, 204 are then injected into a multiplexer 14 in order to form the optical signal at the output. FIG. 1b illustrates another embodiment of the optical combiner 110. More specifically, the optical combiner 110, as shown in FIG. 1b, makes it possible to combine three input optical signals into two optical signals at the output. The optical signals at the output of the optical combiner 110 are respectively transmitted via fourth F04 and fifth F05 optical fibers. The optical combiner comprises wavelength demultiplexing means 11, 12, 13 and fusion devices 20f, 202, 203, 204 similar to those previously described in connection with FIG. Two of the four optical components X1,4, X3,4, respectively at the output of the fusion devices 201, 203, are then injected into a first multiplexer 141 in order to form a first optical signal at the output. The two other optical components X2,4, X3,4 respectively at the output of the fusion devices 202, 204 are then injected into a second multiplexer 142 in order to form a second optical signal at the output. It is thus possible in an optical combiner to combine N optical signals present at the input into M optical signals at the output. No limitation is attached to the number of optical signals input or that of output signals.

It is also possible to provide in such an optical combiner to combine the optical components associated with a given wavelength present on a first set of input ports of the combiner to a first output port and those present on a second set of combiner input ports to a second output port. No limitation is attached to these examples. It is thus possible to switch a wavelength on an input port to an output port.

The optical combiner has been described in a three-way configuration. The number of channels is to be adapted according to the number of wavelengths to be processed. The number of input ports is to define the number of sources to combine. A device 20 for fusing a plurality of optical components associated with a wavelength into an optical component associated with this wavelength according to a particular embodiment of the invention will now be described with reference to FIG. For a first optical component associated with the wavelength Xi, the fusion device 20 comprises the following means, forming a first channel: first optical means 31, arranged to switch to a blocking position of this first optical component according to a first blocking signal; first duplication means 41, connected at the output of the first optical means 31, arranged to duplicate the first optical component to two output ports. The first output port of the first duplication means 41 is connected to the input of a merger module 70, comprising three input ports and one output port. The fusion module 70 is arranged to merge input optical components into an output optical component. The first duplicated optical component is thus injected at the input of the fusion module 70. The second output port of the first duplication means 41 is connected at the input of a first optoelectronic converter 51. The first duplicated optical component is thus converted into a first electrical signal.

For a second optical component Xi, 2 associated with the wavelength λ 1, the fusion device 20 comprises the following means, forming a second channel: second optical means 32, arranged to switch to a blocking position of this second optical component as a function of a second blocking signal; second duplicating means 42, connected at the output of the second optical means 32, arranged to duplicate the second optical component to two output ports. The first output port of the second duplication means 42 is connected to the input of the fusion module 70. The second duplicated optical component is thus injected at the input of the fusion module 70. The second output port of the second duplication means 42 is connected. at the input of a second optoelectronic converter 52. The second duplicated optical component is thus converted into a second electrical signal. For a third optical component 1,1,3 associated with the wavelength λ1, the fusion device 20 comprises the following means forming a third channel: third optical means 33, arranged to switch to a blocking position this third optical component as a function of a third blocking signal; third duplicating means 43, connected at the output of the third optical means 33, arranged to duplicate the third optical component to two output ports.

The first output port of the third duplication means 43 is connected to the input of the fusion module 70. The third duplicated optical component is thus injected at the input of the fusion module 70. The second output port of the third duplication means 43 is connected. at the input of a third optoelectronic converter 53. The third duplicated optical component is thus converted into a third electrical signal. The first 31, second 32, third 33 optical means are further provided to amplify the optical component received at the input, in the absence of a blocking signal. This position is called transfer. This is for example a SOA semiconductor optical amplifier. No limitation is attached to this type of device. It will be recalled here that when the blocking signal is present, these optical means 31, 32, 33 switch to a blocking position of the optical signals at the input of these means. The first 41, second 42, third 43 duplicating means are for example 80-20 couplers.

The second and third electrical signals, respectively from the second and third channels, are then combined by a first coupler 61 to form the first blocking signal of the first optical means 31. The first and third electrical signals, respectively from the first and third channels , are then combined by a second coupler 62 to form the second blocking signal of the second optical means 32. The first and second electrical signals, respectively from the first and third channels, are then combined by a third coupler 63 to form the third signal blocking the third optical means 33. Thus, for a given optical component, the fusion device 20 comprises combining means, arranged to obtain a blocking signal for optical means associated with the given optical component by combining the optical components to merge, excluding the opti than given. At the output of the fusion module 70, an optical component Xi, 4 is obtained. This optical component at the output of the fusion device 20 corresponds to the fusion of the three optical components at the input of the fusion device 20. Thus, in this first embodiment, when a data collision between several optical components occurs as input, the data of a single optical component are present at the output. Indeed, when an optical component is selected, the other optical components are blocked. Thus, the optical component becoming active first is selected to form the output optical component. When the selected optical component becomes inactive, another optical component that has become active can be selected in turn. A single optical component is selected from the input active components of a fusion device to form the output optical component, based on a "first come, first served" mechanism. Operating states of the fusion device 20 will now be detailed in connection with FIGS. 3a, 3b and 3c.

FIG. 3a corresponds to a state in which no optical component is present on one of the input ports of the fusion device 20. The first 31, second 32, third 33 optical means are in their respective transfer positions. FIG. 3b corresponds to a state in which a first optical component X'i associated with the wavelength λ is present on the first input of the fusion device 20. The first duplication means 41 duplicate the first optical component and enable to form the first electrical signal, then the second and third blocking signals. The second 32 and third 33 blocking means then switch to their locking positions under control respectively of the second and third blocking signals.

The optical component at the output of the fusion device 20 is then formed of the first optical component. When the first optical component becomes inactive, the first electrical signal is no longer present and the second 32 and third 33 locking means return to the transfer position, illustrated in Figure 3a. FIG. 3c corresponds to a state in which a second optical component X'2 associated with the wavelength λ is present on the second input of the fusion device 20. The second duplication means 42 duplicate the second optical component and enable forming the second electrical signal, then the first and third blocking signals. The first 31 and third 33 locking means then switch to their locking positions under control respectively of the first and third blocking signals.

The optical component at the output of the fusion device 20 is then formed of the second optical component. When the second optical component becomes inactive, the second electrical signal is no longer present and the first 31 and third 33 locking means return to the transfer position, illustrated in Figure 3a. In the state illustrated in FIG. 3c, when a third optical component Xi, 3 associated with the wavelength λ is present on the third input of the fusion device 20, this third component will be blocked as long as the second component component is present at the input of the device. A residual optical signal is then likely to be transmitted when the blocking means 33 pass to the transfer position. In order to avoid the transmission of this residual optical signal, it is possible to provide, as an option, elementary electronic gates, making it possible to hold the locking means 31, 32, 33 in the locked position as long as a residual optical signal is present at the input . This improves the performance of the device by blocking the transmission of possibly incomplete packets. In the first particular embodiment described with reference to FIGS. 2, 3a, 3b, 3c, the duplicating means 41-43 are connected at the output of optical means 31-33.

This embodiment thus has the advantage of managing the data collisions at the input of the merging device 20 on the basis of a "first come-first served" principle. In a second particular embodiment, not shown, the duplicating means 41-43 are placed at the input of the merging device 20. In this case, the data collisions at the input of the merging device 20 lead to a lack of data. This second embodiment thus makes it possible to manage the collisions but is however less advantageous than the previous one. In a first variant of the first embodiment of the fusion device 20, the combining means 61, 62, 63 combine optical signals. The second output port of the first duplication means 41 is connected at the input of second 62 and third 63 combining means, after duplication. The second output port of the second duplication means 42 is connected at the input of the first 61 and third 63 combination means, after duplication. The second output port of the third duplication means 43 is connected at the input of the first 61 and second 62 combining means, after duplication.

The optical signal at the output of the combining means 61 is then converted into an electrical signal by a first optoelectronic converter to form the first blocking signal. The optical signal at the output of the combining means 62 is then converted into an electrical signal by a second optoelectronic converter to form the second blocking signal.

The optical signal at the output of the combining means 63 is then converted into an electrical signal by a third optoelectronic converter to form the third blocking signal. The optical combining means have the advantage of being passive and thus of not requiring power supply.

In the second variant of the first embodiment of the fusion device 20, first delay means are connected at the input of the first optical means 31. These first delay means are arranged to apply a predetermined fixed duration delay D to the first input optical component. of the fusion device 20. This is for example a fixed optical delay line or "fiber delay line" in English. The fixed duration is determined in advance as a function of the time required for the propagation of the optical signal and the configuration of the first optical means 31. Likewise, second delaying means are connected at the input of the second optical means 32, and third delaying means are connected at the input of the third optical means 33. The second and third delay means are similar to the first delay means. A device 20 for fusing a plurality of optical components associated with a wavelength into an optical component associated with this wavelength according to a third particular embodiment of the invention will now be described in connection with FIG. 7. For a first optical component associated with the wavelength λ 1, the fusion device 20 comprises the following means, forming a first channel: first optical means 101, arranged to switch to a blocking position of this first component optical according to a first blocking signal; first duplication means 41, connected to a first output port of the first optical means 101, arranged to duplicate the first optical component to two output ports. The first output port of the first duplication means 41 is connected to a merger module 70, comprising three input ports and one output port. The fusion module 70 is arranged to merge input optical components into an output optical component. The first duplicated optical component is thus injected at the input of the fusion module 70. The second output port of the first duplication means 41 is connected to the input of a first optoelectronic converter 51. The first optical component is thus converted into a first signal electric.

An output of first delay means 91 is connected at the input of the first optical means 101. The first delay means 91 are arranged to apply a predetermined fixed duration delay D to the first input optical component. This is for example a fixed optical delay line. The fixed duration is determined in advance as a function of the time required for the propagation of the optical signal and the configuration of the first optical means 101. In a first position, called the transfer position, the first optical means 101 are arranged to switch the first component optical Xo to the first output port, connected to the first duplication means 41. This transfer position corresponds to that taken in the absence of the first blocking signal.

In a second position, corresponding to the blocking position, the first optical means 101 are arranged to transmit the first optical component to a second output port. This second output port of the first optical means 101 is connected to a first circulator 71 which reinjects into the optical fiber the first optical component Xi, i which has suffered a collision, in the direction opposite to the direction of propagation. The first circulator 71 is thus placed on the first input port of the fusion device 20 and is arranged to pass the first optical component in the direction of propagation and to reinject a signal back in the opposite direction of the direction of propagation. To summarize, when the first blocking signal is absent, the first optical means 101 are positioned in the transfer position of the first optical component to the first output port; when the first blocking signal is present, the first optical means 101 are positioned in the blocking position of the first optical component and deviate it to the second output port. This is for example a "Microring resonators" type switch. No limitation is attached to this type of device. For a second optical component Xi, 2 associated with the wavelength λ 1, the fusion device 20 comprises the following means, forming a second channel: second optical means 102, arranged to switch to a blocking position of this second optical component as a function of a second blocking signal; second duplication means 42, connected on a first output port of the second optical means 102, arranged to duplicate the second optical component to two output ports. The first output port of the second duplication means 42 is connected to the second input of the fusion module 70. The second duplicated optical component is thus injected at the input of the fusion module 70. The second output port of the second duplication means 42 is connected at the input of a second optoelectronic converter 52. The second duplicated optical component is thus converted into a second electrical signal. An output of second delay means 92 is connected at the input of the second optical means 102. These second delay means 92 are arranged to apply a predetermined fixed duration delay D to the second input optical component X1,2 and are similar to the first delay means 91. The second optical means 102 are similar to the first optical means 101. When the second blocking signal is absent, the second optical means 102 are positioned in the transfer position of the second optical component X1,2 to the first output port. connected to the second duplicating means 42; when the second blocking signal is present, the second optical means 102 are positioned in the blocking position of the second optical component X1,2 and deviate it to a second output port. The second output port of the second optical means 102 is connected to a second circulator 72 which reinjects into the optical fiber the second optical component X1,2 which has suffered a collision. The second circulator 72 is placed on the second input port of the fusion device and is similar to the first circulator 71.

For a third optical component Xi, 3 associated with the wavelength λ 1, the fusion device 20 comprises the following means, forming a third channel: third optical means 103, arranged to switch to a blocking position of this third optical component according to a third blocking signal; third duplication means 43, connected to a first output port of the third optical means 103, arranged to duplicate the third optical component to two output ports. The first output port of the third duplication means 43 is connected to the third input of the fusion module 70. The third duplicated optical component is thus injected at the input of the fusion module 70. The second output port of the third duplication means 43 is connected at the input of a third optoelectronic converter 53. The third optical component is thus converted into a third electrical signal. An output of third delaying means 93 is connected at the input of the third optical means 103. These third delaying means 93 are arranged to apply a predetermined fixed duration delay D to the third input optical component 1,1,3 and are similar to the first means of retardation 91. The third optical means 103 are similar to the first optical means 101. When the third blocking signal is absent, the third optical means 103 are positioned in the transfer position of the third optical component 1,1,3 to the first output port connected to the third duplicating means 43; when the third blocking signal is present, the third optical means 103 are positioned in the blocking position of the third optical component 1,1,3 and deviate it to a second output port. The second output port of the third optical means 103 is connected to a third circulator 73 which reinjects into the optical fiber the third optical component 1,1,3 which has suffered a collision. The third circulator 73 is placed on the third input port of the melter and is similar to the first circulator 71. The first 41, second 42, third 43 duplicating means are for example 80-20 couplers.

The second and third electrical signals, respectively from the second and third channels, are then combined by a first coupler 61 to form the first blocking signal of the first optical means 101. The first and third electrical signals, respectively from the first and third channels , are then combined by a second coupler 62 to form the second blocking signal of the second optical means 102.

The first and second electrical signals, respectively from the first and third channels, are then combined by a third coupler 63 to form the third blocking signal of the third optical means 103. Thus, for a given optical component, the fusion device 20 comprises combining means arranged to obtain a blocking signal for optical means associated with the given optical component by combining the optical components to be fused, excluding the given optical component. At the output of the fusion module 70, an optical component Xi, 4 is obtained. This optical component at the output of the fusion device 20 corresponds to the fusion of the three optical components at the input of the fusion device 20. By way of illustration, examples similar to those described with reference to FIGS. 3a, 3b and 3c are now described. In a first state, no optical component is present at the input of the fusion device 20.

The first 101, second 102, third 103 optical means are in their respective transfer positions. In a second state, a first optical component X1,1 associated with the wavelength λ is present on the first input of the fusion device 20. The first duplication means 41 duplicate the first optical component and make it possible to form the second optical component. and third blocking signals. The second 102 and third 103 optical means then switch to their locking positions under control respectively of the second and third blocking signals. When a second optical component is present at the input of the device, it is then re-injected into the optical fiber at the second circulator 72 to the corresponding source node as long as the first component is present at the input of the device.

When a third optical component is present at the input of the device, it is then re-injected into the optical fiber at the third circulator 73 to the corresponding source node as the first component is present at the input of the device. The optical component at the output of the fusion device 20 is then formed of the first optical component. When the first optical component becomes inactive, the second and third blocking signals are no longer present and the second 102 and third optical means 103 return to the transfer position, corresponding to the first state described above. In a third state, a second optical component X'2 associated with the wavelength λ is present on the second input of the fusion device 20. The second duplication means 42 duplicate the second optical component and make it possible to form the first and third blocking signals. The first 101 and third 103 optical means then switch to their locking positions under control respectively of the first and third blocking signals. When a first optical component X1, associated with the wavelength X, is present at the input of the fusion device 20, this first component will be blocked and reinjected into the optical fiber at the level of the first circulator 71 towards the source node. corresponding as long as the second component is present at the input of the device. When a third optical component X, 3 associated with the wavelength λ, is present on the third input of the fusion device 20, this third component will be blocked and reinjected into the optical fiber at the third circulator 73 towards the third corresponding source node as long as the second component is present at the input of the device. The optical component at the output of the fusion device 20 is then formed of the second optical component. When the second optical component becomes inactive, the first and third blocking signals are no longer present and the first 101 and third 103 optical means return to the transfer position, corresponding to the first state described above. Thus, in this third embodiment, similarly to the first embodiment, when a data collision between several optical components occurs as input, the data of a single optical component is present at the output. Indeed, when an optical component is selected, the other optical components are blocked and are returned by the fusion device 20 to the source nodes having transmitted them. Thus, the optical component becoming active first is selected to form the output optical component. When the selected optical component becomes inactive, another optical component that has become active can be selected in turn. A single optical component is selected from the input active components of a fusion device to form the output optical component, based on a "first come, first served" mechanism. The first variant described in relation to the first embodiment is also provided for this third embodiment.

In a third variant of these embodiments, elementary electronic gates are provided, in order to keep the locking means 31-33, 101-103 in the locking position as long as a blocked optical component is present at the input of the fusion device. 20. Indeed, if the blocking means 31-33, 101-103 switch to the transfer position as soon as the selected optical component is no longer present, a residual signal associated with a blocked optical component is likely to be always present. at the input of the fusion device. Elementary electronic gates thus make it possible to avoid the transmission of this residual signal, corresponding to an incomplete packet. In a fourth variant to these embodiments, a control unit is arranged to form the blocking signals, replacing the combination means 61-63.

This control unit is arranged to form and transmit: the second and third blocking signals respectively to the second and third optical means in the presence of the first electrical signal; the first and third blocking signals respectively to the first and third optical means in the presence of the second electrical signal; the first and second blocking signals respectively to the first and second optical means in the presence of the third electrical signal.

The transmitted blocking signals are held for a predetermined duration, equal to the duration of a packet, and as long as a residual signal corresponding to a blocked optical component is present at the input of the fusion device. Thus, the optical means associated with blocked optical components remain in the locked position as long as a residual signal is present at the input. As a result, only complete packets are present at the output of the fusion device 20. This variant also makes it possible to combine a larger number of optical components. In addition, this also improves the performance of the merging device for the second embodiment by avoiding a lack of output data in the event of an input collision of the merging device.

An optical aggregation node 100 of optical signals will now be described in relation to FIG. 4. Such an aggregation node makes it possible to aggregate the optical signals received from optical nodes 131-135 into an optical signal and to transmit this signal. optically all optical to a communication network 1, for example the Internet, and also to distribute to the optical nodes 131-135 an optical signal received from the communication network 1.

These optical nodes belong for example to a local area network LAN, for "Local Area Network". An optical signal is transmitted by the first optical node 131 to the aggregation node 100 via a first optical fiber F01.1. An optical signal is transmitted from the aggregation node 100 to the first optical node 131 via a second optical fiber F01,2. Similarly, the second optical node 132 is connected to the aggregation node 100 via two optical fibers F02,1, F02,2; the third node 133 by two optical fibers F03,1, F03,2; the fourth node 134 by two optical fibers F04,1, F04,2; the fifth node 135 by two optical fibers F05,1, F05,2. An optical signal is transmitted by the aggregation node 100 to the communication network 1 via a first optical fiber F06,1. An optical signal to the aggregation node 100 is transmitted from the communication network 1 via a second optical fiber F06,2. The aggregation node 100 comprises an optical combiner 110 of optical signals as described above, arranged to combine optical signals respectively received from the optical nodes 131-135 into an optical signal and an optical signal distribution device 120, arranged for distributing a received optical signal to the optical nodes 131-135. The optical combiner 110 includes in this case four optical signal input ports.

It is emphasized here that data generating collisions are suppressed by the fusion device as described above. In the first and second embodiments described with reference to FIG. 2, these deleted data can be re-transmitted by the source node when a timer expires. It is also possible to provide that the aggregation node notifies by an explicit signaling message the sending source node of the deleted data. In the third embodiment described with reference to FIG. 7, the source node receives back the optical signal that has not been transmitted due to a collision. He can then re-issue the data. It is therefore not necessary to wait for the expiration of a timer, or to implement specific signaling to notify the source node of the deletion of the data. Upon receiving the optical feedback signal, the source node can re-transmit the deleted data. Such a source node, arranged to cooperate with the fusion device according to the third embodiment, comprises in particular: an adjustable transmitter; an optical circulator; - a photodiode. The optical circulator transmits the optical signal emitted by the adjustable transmitter to its destination and enables the received optical signal to be deflected on the optical fiber to the photodiode. The photodiode converts an optical feedback signal received from a fusion device into an electrical signal. This electrical signal thus makes it possible to notify the source node of the occurrence of a collision. It is emphasized here that the source node does not need to process data carried by the optical signal back. When the propagation time between the source node and the first device comprising a fusion device is less than the duration of a packet, at the occurrence of a collision, the source node must only retransmit the current packet. transmission. This retransmission can be immediate or after the expiration of a predetermined duration. In the opposite case, that is to say when the propagation time between the source node and the first equipment comprising a fusion device is greater than the duration of a packet, the source node then determines as a function of this propagation time which packet was the subject of a collision. The propagation time may have been previously determined according to known methods of the state of the art. In order to allow the optical nodes 131-135 to communicate with each other, particular wavelengths are dedicated to the local traffic. In this case, a first optical combiner 110 may be in charge of merging the wavelengths allocated to the traffic destined for the communication network 1 and a second optical combiner in charge of merging the wavelengths allocated to the local traffic. It is also conceivable to provide a single optical combiner, such as that described with reference to FIG. The distribution device 120 is then arranged to: - merge the optical signal at the output of the second optical combiner and the optical signal from the communication network 1, and - distribute the optical signal resulting from the fusion. In a more general context of the interconnection of several optical access networks comprising several aggregation levels, a set of wavelengths can be assigned to each optical access network. The data traffic is then combined, routed and broadcast at each aggregation node according to the set of wavelengths affected. Referring now to FIG. 5 on which are illustrated the steps of a method of fusing a plurality of optical components associated with a wavelength into a fused optical component associated with said wavelength, these steps being able to advantageously carried out by means of the melting device 20 of FIG. 2. The melting method comprises a step E1 for duplicating the optical components to be fused, during which said optical components to be fused are duplicated. In a step E2 of the melting process, blocking signals are obtained. A blocking signal is associated with an optical component to be fused and is intended to control optical means 31, 32, 33, as described above. A blocking signal associated with an optical component to be fused is obtained by combining the duplicate optical components of the plurality, with the exception of said optical component to be fused. Then, in a step E3, the optical component to be merged is blocked or not, depending on the associated blocking signal obtained.

The fused optical component associated with the wavelength is then obtained in a step E4 from the unblocked optical component. In a particular embodiment, in step E3, the optical component to be merged blocked further forms an optical component in return, when the blocking signal is present. Reference is now made to FIG. 6, in which are illustrated the steps of a method for optically combining optical signals, an optical signal comprising optical components respectively associated with a plurality of wavelengths, these steps being advantageously carried out at the same time. means of the optical combiner 110 of FIG. The method for optically combining comprises a step F1 of demultiplexing the input optical signals, to obtain from an optical input signal a plurality of optical components.

The fusion method as described above is then implemented for the optical components associated with a wavelength obtained in the demultiplexing step from the input optical signals. Each wavelength is thus treated. The method for optically combining then comprises a step F2 for multiplexing the optical output components obtained by implementing the merging method into an output optical signal.

Claims (10)

REVENDICATIONS1. A device for fusing (20) a plurality of optical components associated with a wavelength into an optical component associated with said wavelength, the device comprising: for each optical component of said plurality to be fused, - optical means (31-33; 101-103), arranged to switch to a blocking position of said optical component to be fused according to a blocking signal; duplication means (41-43), arranged to duplicate said optical component to be fused to combining means (61-63); the combining means, arranged to obtain said blocking signal by combining the duplicate optical components of the plurality, excluding said optical component to be fused; and merging means (70) arranged to merge the optical components at the output of the optical means. 2. Fusion device according to claim 1, wherein the optical means are further arranged to amplify the optical component and the duplicating means are connected at the output of the optical means. 3. Fusion device according to claim 1, wherein the optical means are further arranged to amplify the optical component and the duplicating means are connected at the input of the optical means. 4. Fusion device according to claim 1, further comprising for each optical component of said plurality to be fused, delaying means arranged to apply a predetermined fixed time delay to said optical component to be merged and connected to input associated blocking means said optical component to be merged. 5. Fusion device according to claim 1 or 4, wherein in the locking position, the optical means deflect said optical component to be fused to injection means (71-73), arranged to reinject said optical component in a fiber optical to a source node. 6. Fusion device according to any one of the preceding claims, wherein a control unit forms the combining means. An optical signal optical combiner, an optical signal comprising optical components respectively associated with a plurality of wavelengths, said optical combiner comprising: demultiplexing means (11-13) respectively associated with an input optical signal, arranged to obtain from the input optical signal a plurality of optical components; a plurality of fusion devices (201-204) according to claim 1, respectively associated with a wavelength and connected at the output of the demultiplexing means; multiplexing means (14), arranged to multiplex the optical components at the output of the fusion devices into an optical output signal. An optical signal aggregation node (100), an optical signal comprising a plurality of optical components respectively associated with a plurality of wavelengths, said node comprising: an optical signal optical combiner according to claim 7, arranged to combine optical signals respectively received from source optical nodes into an optical signal; a device for distributing an optical signal, arranged to distribute a received optical signal to said source nodes. A method of fusing a plurality of optical components associated with a wavelength into a fused optical component associated with said wavelength, comprising: a step (E1) of duplicating the optical components of said plurality to be merged during which said optical components to be fused are duplicated; a step (E2) for obtaining blocking signals, in which a blocking signal associated with an optical component to be fused is obtained by combining the duplicate optical components of the plurality, with the exception of said optical component to be fused; a blocking step (E3), during which an optical component to be fused is blocked as a function of the associated blocking signal; a step (E4) for obtaining the fused optical component associated with the wavelength from an unblocked optical component. A method for optically combining optical signals, an optical signal comprising optical components respectively associated with a plurality of wavelengths, comprising: a step of demultiplexing the input optical signals to obtain from an optical signal at the input a plurality of optical components; - an implementation of the fusion method according to claim 6, for optical components associated with a wavelength obtained in the demultiplexing step from the input optical signals; a step of multiplexing the output optical components obtained by implementing the fusion process into an output optical signal.