WO2002075990A2

WO2002075990A2 - Add/drop multiplexer based on a dual core fiber comprising means for optimizing a connection

Info

Publication number: WO2002075990A2
Application number: PCT/FR2002/000931
Authority: WO
Inventors: Philippe Yvernault; Erwan Goyat; Laurent Brilland; David Pureur
Original assignee: Highwave Optical Technologies
Priority date: 2001-03-19
Filing date: 2002-03-15
Publication date: 2002-09-26
Also published as: FR2822314B1; AU2002247820A1; FR2822314A1; WO2002075990A3

Abstract

The invention relates to an add/drop multiplexer comprising an interferometer produced on a section of dual-core fiber (100) connected to mono-core fibers, characterized in that it comprises means (Ζ1, Ζ2) producing a difference in phase between the two arms (10, 20) of the interferometer equal to Π modulo 2Π.

Description

INSERTION-EXTRACTION MULTIPLEXER BASED ON A TWO-CORE FIBER HAVING CONNECTION OPTIMIZATION MEANS

The present invention relates to the field of insertion multiplexers.

Extraction (MIE) carried out using optical fibers.

Such multiplexers allow for example the routing of signals between different loops of telecommunications networks, without external energy supply and without having to resort to electronics. More specifically, the invention relates to an Insertion Multiplexer

Extraction performed from Bragg gratings on a Mach-Zehnder interferometer with bi-core fiber.

The creation of Fiber Insertion-Extraction Multiplexers from Bragg gratings has already been the subject of numerous publications. The function of a conventional MIE extraction insertion multiplexer is illustrated diagrammatically in FIG. 1 attached.

As illustrated in FIG. 1, this MIE multiplexer comprises an input channel 1, an extraction port 2, an insertion port 3 and an output channel 4.

In such a multiplexer the wavelengths λi, λ ₂ ... λ, - ...., λ _n introduced on the input channel 1, exit through the output channel 4, except the wavelength λj extracted on the extraction port 2. Furthermore, the wavelength λj 'injected on the insertion port 3 is found on the output channel 4.

Known Multiplexers can be classified into two categories: those with single-core fibers and those with multi-core fibers (generally two). A diagrammatic illustration has been illustrated in the appended FIG. 2, an insertion-extraction multiplexer produced from a single single-core fiber 5. This fiber 5 comprises at least one Bragg grating 6.

The use of a single fiber and one or more Bragg 6 network (s) requires the use of 2 ', 3' optical circulators to couple the Insertion 2 and Extraction 3 channels to the propagation of signals, (see doct [1]). These devices generally have high losses and are more expensive. The realization of low-cost Insertion-Extraction Multiplexers therefore meant that there was no need for optical circulators.

The cheapest and low-loss way is to use fiber interferometers. We thus obtain realizations from two single-core fibers:

- documents [2] and [3] describe two exemplary embodiments of directional couplers

- documents [4] and [5] describe examples of interferometers.

A diagrammatic illustration has been illustrated in attached FIG. 3 of a multiplexer with insertion and extraction, produced from two single-core fibers 7, 8.

Each of these two fibers 1, 8 has at least one Bragg grating 6. The fibers 7, 8 are respectively coupled on either side of the Bragg grids 6. In FIG. 3, the dividing couplers thus formed are referenced 9 .

The first adjacent ends of the two fibers 7, 8 respectively form the inlet channel 1 and the extraction port 2.

The second adjacent ends of the two fibers 7, 8 respectively form the insertion path 3 and the exit path 4.

Thus the wavelengths λ _l5 λ ₂ .... λ, .... λ _n introduced on the first end 1 of the fiber 7 exit through the second end 4 of the fiber 8, except for the wavelength λβ extracted by the first end 2 of the fiber 8. Furthermore, the wavelength λβ injected on the second end 3 of the fiber 7 is found on the second end 4 of the fiber 8.

The ideal is to produce the dividing couplers 9 using fibers 7, 8 which serve to register the Bragg gratings 6, and which are consequently photosensitive. In document [6] it has been shown that the use of a fiber with a photosensitive cladding makes it possible to considerably reduce the coupling to the cladding modes, irretrievably present in conventional fibers. The photosensitivity of the sheath is obtained by co-doping the latter.

However, the production of couplers based on photo-sensitive fibers poses difficulties. Indeed, the cores of single-core fibers 7, 8 placed side by side, are generally spaced apart by about 125 μm (ie the outside diameter of the fibers). To make a coupler 9 (and there is an exchange of energy between the two cores), it is therefore necessary to bring the cores closer together, generally by fusion-stretching, so that they are typically less than ~ 20 μm.

However, the doping which makes the sheath photosensitive also makes it more fusible than the core of the fiber, which makes the fusion-stretching of such a fiber extremely delicate. Indeed, the different dopants contained in the sheath do not diffuse in the same way. In addition, the sheath may be weakened at the coupling when it is subjected to a melting-stretching step.

From there emerges the idea of using a multi-core fiber, generally bi-core. A bi-core fiber indeed has two hearts surrounded by a sheath. The sheath is photosensitive to allow the registration of Bragg gratings. The hearts are initially spaced a few tens of microns: the center distance. The value of the center distance is the minimum allowed by a limited crosstalk on the length of the component. The two cores being closer together (30-100 μm) than when using single-core fibers (125 μm), there is less need to merge the multi-core fiber to perform the coupling. It follows that the diffusion of the dopants of the photosensitive sheath is negligible. An example of the use of such a bi-core fiber is described in document [7]. Other embodiments exist although this is for other reasons than those mentioned. Black documents [8] and [9].

However, if certain optical performances are excellent with such components, the fact remains that the use of a bi-core fiber poses certain specific problems. In particular :

- the hearts are not perfectly identical (in refractive index, in size, etc.);

- an intrinsic phase shift in transmission as in reflection generally appears;

- the connection to single-mode fibers of the network is not identical on the 4 input / output ports and causes insertion losses; - the Bragg gratings photo-inscribed on each arm of the interferometer are not perfectly identical (in particular in wavelength or rate of reflection ...).

Connection solutions have already been proposed. However, none is entirely satisfactory. As a result, so far, when a bi-core fiber is used, only one of the two cores has an optimal coupling efficiency with the connector used.

The object of the present invention is therefore to propose an Insertion-Extraction Multiplexer based on a bi-core fiber whose injected and transmitted signals are carried by the same core in order to minimize the insertion losses for these signals.

This object is achieved in the context of the present invention thanks to an interferometer produced on a section of bi-core fiber, characterized in that it comprises means introducing a phase difference equal to π modulo 2π between the two arms of the interferometer. The interferometer is advantageously of the Mach-Zehnder type.

The realization of an Insertion-Extraction Multiplexer from a Mach-Zehnder on bi-core fiber has the following advantage in particular: the cores being contained in the same sheath, the difference in optical path between the arms of the interferometer essentially depends on the small difference in index between the two cores. The stability of a bi-core fiber interferometer is therefore much higher than that of its single-core fiber counterpart.

The performance on the output and extraction channels can be optimized by judicious balancing of the optical phase shift between the arms of the interferometer both in transmission and in reflection. Other characteristics, aims and advantages of the present invention will appear on reading the detailed description which follows, and with reference to the appended drawings, given by way of nonlimiting examples and in which:

- Figures 1 to 3 previously described schematically illustrate Insertion-Extraction multiplexers, in accordance with the state of the art, - Figures 4 and 5 show schematic views in cross section, of two examples of known bi-core fibers, FIG. 6 represents an Insertion-Extraction Mutliplexer based on a bi-core fiber in accordance with the state of the art,

- Figures 7 and 8 schematically represent two variants of Insertion-Extraction Mutliplexers based on bi-core fiber according to the invention, - Figure 9 shows the phase difference between the signals reflected and transmitted through two Bragg gratings of slightly different spectral characteristics,

- Figure 10 shows the general structure of a multiplexer obtained in the context of the present invention, and - Figure 11 shows the general structure of a multiplexer according to a variant of the present invention comprising three ports.

Bi-core fibers are well known per se to those skilled in the art. They will therefore not be described in detail below.

Nevertheless, we will recall below some basic data relating to bi-core fibers, in relation to FIGS. 4 and 5.

Figure 4 shows a bi-core fiber in its most general form.

The bi-core fiber generally has two cores 10, 20 delimited by the surfaces S4 and S4 ', most often circular but not necessarily, through which the light signals propagate. According to Figure 4, the heart 10 is circular in revolution while the heart 20 is elliptical. These cores can be doped with rare earth ions for applications relating to amplification, d characterizes the distance between the two cores 10, 20.

The cores 10, 20 can be surrounded by a zone 30, 40 of different properties, delimited by the surfaces S3 and S3 '(note: S3 (S3') is not necessarily in contact with S4 (S4 '). zones 30, 40 can be photosensitive for applications requiring photo-inscription of Bragg gratings S3 and S3 'can have any geometric shape and accept any of the usual dopants for applications such as polarization maintenance , amplifier etc ... The area delimited by the surface S2, generally called the optical sheath 50, is often undoped. It can take various and varied geometric shapes.

The optical sheath 50 is in turn surrounded by a protective coating 60 delimited by the surface SI, generally circular.

FIG. 5 corresponds to the simplified case where the surfaces S3 and S3 ', S4 and S4' are identical. SI is circular and has a diameter of several hundred microns. Let d be the distance between the hearts and d2 the width of the rectangle circumscribing the surface S2, the length of this rectangle is d2 + d. FIG. 6 illustrates a conventional Insertion-Extraction Multiplexer (MIE) comprising:

a section of bi-core fiber 100 having on each of its two elementary cores 10, 20, a Bragg grating 102, 102 ′ tuned on a wavelength

- two fibers 130, 140 respectively serving as inlet and extraction port,

a first connection device 120 ensuring the connection of the fibers 130, 140, with a first end of the cores 10, 20 of the fiber 100,

- two fibers 130 ′, 140 ′ respectively serving as insertion and outlet ports, and

- a second connection device 120 'which ensures the connection of the fibers 130', 140 'with the second end of the cores 10, 20 of the fiber 100.

Couplers 110, 110 'are defined by drawing fusion of the bi-core fiber 100, respectively on either side of the Bragg gratings 102, 102'.

In this classic design, there are crossing signals between the input core and the output core. Thus the signals λl to λn entering at 130 (first heart) exit at 140 '(second heart), except for the Bragg wavelength λβ which is extracted at 140. Likewise the wavelength λβ inserted at 130' in 140 '.

In practice, the hearts 10, 20 are never strictly identical. The difference in opto-geometric properties of the cores is either determined on purpose for better adaptation of the modes, or is detrimental and inherent in the manufacturing process. In both cases, it is necessary to favor the transmission of the signals (those which do not "see" the wavelength of Bragg). To take these constraints into account, as illustrated in FIGS. 7 et seq., Means are provided according to the present invention capable of adding a phase difference between the arms 10 and 20 of the interferometer, additional to the phase shift inherent in the structure , such that the total phase shift is equal to π modulo 2 π.

This phase difference can result both from a difference in positioning of the networks 102, 102 ′, as well as from a difference in index between the two cores 10, 20 (either natural, or induced by sunshine UN ... ), or even a slight difference in the characteristic of the Bragg gratings 102, 102 ′ (or even of any combination of such means). In other words, in the context of the present invention, the phase correction means suitable for generating a phase shift of π modulo 2π between the arms 10 and 20 of the interferometer, can be formed by any of the means mentioned above or any combination of such means (difference in positioning of the networks 102, 102 ', difference in index between the two cores 10, 20, difference in characteristic of the networks 102, 102'). Furthermore preferably within the framework of the present invention, the phase correction means must take into account any phase shift inherent in the production of the interferometer, whatever the origin of this phase shift inherent in the production (difference in positioning networks 102, 102 ', difference in index between the two cores 10, 20, difference in characteristic of networks 102, 102').

Other means can also be used to define the desired phase shift, and in particular: local heating (which modifies the optical index of the fiber locally), mechanical stress, for example by traction or torsion, or even an application of an electric field. These alternative means can be used separately or in combination with those mentioned above.

Thus, according to the present invention, the signals entering at 130, on the first end of a first heart 10, exit on the second end 130 'of this same heart 10, with the exception of the Bragg wavelength λβ which is extracted at 140 on the first end of the second heart 20. Furthermore, the wavelength λ _B injected at the second end 140 'of the second heart 20 also leaves at 130', that is to say on the second end of the first heart 10. Figures 7 and 8 show an Insertion-Extraction Multiplexer (MIE) comprising:

a section of bi-core fiber 100 having on each of its two elementary cores 10, 20, a Bragg grating 102, 102 ′ photo-registered and tuned on a wavelength λ _B ,

- two fibers 130, 140 respectively serving as inlet and extraction port,

a first connection device 120 ensuring the connection of the fibers 130, 140 with a first end of the cores 10, 20 of the fiber 100,

- two fibers 130 ', 140' serving respectively as an insertion and output port, and - a second connection device 120 'which ensures the connection of the fibers 130', 140 'with the second end of the cores 10, 20 of the fiber 100.

Two couplers (at 3dB) 110 and 110 ′ are produced by fusion drawing of the bi-core fiber on either side of the Bragg gratings 102, 102 ′ to form the Mach-Zehnder interferometer. The parameters of the fusion are adjusted so as to obtain couplers

50/50 at the same wavelength as that reflected by the photo-registered networks while having the most achromatic response possible.

The insertion and extraction functions are not affected by the phase shift π thus introduced since the cumulative phase shift before recombination in the couplers is equal to 0 [2π]. In this case, the connection to the connection device is optimal.

As indicated previously, the phase shift introduced according to the present invention (using at least any of the previously defined means), in addition to the initial phase shift, is suitable for obtaining a total phase shift of π. In practice, the phase can be tuned to take into account the phase shift caused by a slight detuning of the networks in wavelength and spectral width, as well as by a slight difference in index between the two cores. In practice, the Bragg gratings are never identical, which causes a phase shift in reflection and in transmission as illustrated in FIG. 9. In FIGS. 7 and 8, φi, φ ₂ , φ ₃ , φ represent the phase shifts directly linked to the optical path (which takes into account the length separating the Bragg gratings 102, 102 'from the couplers 110, 110', as well as the difference in index). φ _rl , φ _r2 are the phase shifts undergone by the Bragg length signals reflected by the networks 102 and 102 '. φ _tl , φ _t2 are the phase shifts undergone by the transmitted signals (outside the Bragg wavelength) after crossing the Bragg gratings 102, 102 '.

In this case if Φ = φi - φ ₂ + (φ _rl - φ _r2 ) / 2 is positive, the invention consists in producing a photoinduced phase shift such that Φi = Φ + 2kπ in the bottom heart (case of Figure 7 ) or Φi = 2π-Φ + 2kπ in the top heart (not shown) (the weaker of the two). If Φ is negative, then we produce a phase shift Φi = 2π + φ + 2kπ on the bottom heart (case of Figure 7) or Φi = Φ + 2kπ on the top heart (not shown) (the lower of the two) ).

Once Φi, determined, it remains to freeze Φ ₂ = π - (φι-φ2 + φ3-φ4 + φti-φt2) + Φ ι + 2kπ. According to the sign of (φ ₁ -φ ₂ + φ3-φ ₄ + φtι-φt2-Φι) ₅ the phase shift will be applied to the heart at the top (figure 8) or at the bottom (figure 7).

Note that

φtι-φt2— φt (defined in figure 9).

Those skilled in the art will understand that the present invention makes it possible to transmit the signals transmitted outside the Mach-Zehnder through a single heart, the one whose mode size is most suited to the guides of the connection device.

In the above description, symmetrical connection devices are assumed, with in particular similar optical guides. In the case of different optical guides, the difference of which takes into account that between the cores of the bi-core fiber, it is possible to use a pair "guide-of-the-connecting-device / one-of-the-cores -fiber-bi-core "whose coupling will be optimized. The above description remains valid.

Is illustrated in Figure 10, a multiplexer with insertion extraction in accordance with the present invention. This FIG. 10 shows a section of bi-core fibers 100. Each of the two cores 10, 20 (asymmetrical diagram) has a Bragg grating 102, 102 'between two coupling zones 110, 110' formed by fusion-stretching , to form an interferometer-type multiplexer Mach-Zehnder interferometer. The bi-core fiber 100 is connected at the input and output to standard connection means (possibly based on planar guide) or standard fibers 130, 140, 130 ', 140'. The fibers 130, 140 serve as an entry and extraction route. The fibers 130 ′, 140 ′ serve as exit and insertion path. The connection of the input channel 130 and the output channel 130 'is optimized on the same core 10.

An alternative embodiment according to the present invention is illustrated in FIG. 11, comprising 3 ports.

The elements illustrated in FIG. 11, comparable to elements represented in previous and previously described figures bear identical references to the latter.

The three ports can have input, output and insertion, or input, output and extraction functions.

When there are only three ports, there is no need to optimize the connection on both ends of the dual-core fiber. It is enough to optimize the connection on the end of the bi-core fiber for which the two hearts are used. We then optimize the connection of the input channel in one case and the output channel in the other. The connection of the third port is carried out without difficulty: at the end of the dual-core fiber where there is only one port, it is possible to directly connect the core of the dual-core fiber to a single-core fiber without intermediate connection device.

Of course the present invention is not limited to the particular embodiment which has just been described, but extends to any variant in accordance with its spirit.

BIBLIOGRAPHY [1] EP 0730172

[2] Electron Lett. Flight. 33, n ° 9 (1997)

[3] US 9927932

[4] Electron Lett, Vol. 23, n ° 13 (1987)

[5] Photonics Technology Letters, Nol 5, n ° 2 (1993) [6] OFC95, San Diego, PostDealine papers, page 343-346 (1995)

[7] Electron. Letter, Nol 34, n ° 12, 1250-1251 (1998) [8] J. Ligth Wave Tech., Nol 18, n5, May 2000

[9] J. Electrical Engineering and information science, vol4, n ° 4, 1999

Claims

1. Insertion-Extraction multiplexer comprising an interferometer produced on a section of bi-core fiber (100) connected to single-core fibers, characterized in that it comprises means (Φ], Φ ₂ ) producing a difference phase between the two arms (10, 20) of the interferometer equal to π modulo 2π.

2. Multiplexer according to claim 1, characterized in that the interferometer is of the Mach-Zehnder type.

3. A multiplexer according to one of Claims 1 or 2, characterized in that the _phase correction means for adjusting the phase difference between both arms of the interferometer to π, taking into account the phase shifts inherent to the production of the interferometer chosen from the following group: difference in positioning of the networks (102, 102 '), difference in index between the two cores (10, 20), difference in characteristic of the networks (102, 102') . .

4. Multiplexer according to one of claims 1 to 3, characterized in that it comprises:

- a section of bi-core fiber (100) having on each of its two elementary cores (10, 20), at least one Bragg grating (102, 102 ') photo-registered on a wavelength (λβ),

- two single-core fibers (130, 140) serving respectively as an inlet and as an extraction port,

a first connection device (120) ensuring the coupling of the single-core fibers (130, 140) with a first end of the cores (10, 20) of the bi-core fiber (100),

- two single-core fibers (130 ′, 140 ′) respectively serving as insertion and output ports,

a second connection device (120 ′) which ensures the coupling of the single-core fibers (130 ′, 140 ′) with the second end of the cores (10, 20) of the bi-core fiber (100) and - coupling means (110, 110 ′) produced between the cores (10, 20) of the bi-core fiber (100).

5. Multiplexer according to one of claims 1 to 3, characterized in that it comprises: - a section of bi-core fiber (100) having on each of its two elementary cores (10, 20), at least one Bragg grating (102, 102 ') photo-inscribed on a wavelength (λβ),

- a single core fiber (130) serving as an input,

- a single-core fiber (130 ') serving as an outlet port, - an additional single-core fiber (140, 140) serving alternately as an insertion port or an extraction port,

means which ensure the connection of the single-core fibers with the ends of the cores (10, 20) of the bi-core fiber (100) and

- coupling means (110, 110 ′) produced between the cores (10, 20) of the bi-core fiber (100).

6. Multiplexer according to one of claims 4 or 5, characterized in that the coupling means (110, 100 ') are produced by fusion-drawing of the bi-core fiber.

7. Multiplexer according to one of claims 1 to 6, characterized in that the entry path and the exit path pass through the ends of a single core of the bi-core fiber (100).

8. Multiplexer according to one of claims 1 to 7, characterized in that the interferometer comprises two networks (102, 102 ') produced respectively on each of the cores (10, 20) of the bi-core fiber (100) .

9. Multiplexer according to one of claims 1 to 7 taken in combination with claim 8, characterized in that the means producing a phase difference comprise two phase correction modules (12, 22) disposed respectively on opposite ends fiber (100), compared to networks.

10. Multiplexer according to claim 9, characterized in that the two phase correction modules are formed on the same core of the bi-core fiber (100).

11. Multiplexer according to claim 9, characterized in that the two phase correction modules are formed respectively on the two cores of the bi-core fiber (100).

12. Multiplexer according to one of claims 1 to 11, characterized in that it comprises a phase correction module located on one side of a network and which defines a phase correction equal to Φi = Φ + 2kπ or Φι = 2π - / + φ + 2kπ, with

φ _t and Φ ₂ representing the phase shifts directly linked to the optical path depending on the positioning offset of the networks and the difference in index of the cores which carry them, _-; φ _r ι and φ _r2 representing the phase shifts undergone in reflection by the signals at the specific wavelength of the networks.

13. Multiplexer according to claim 12, characterized in that it further comprises a phase correction module located on the other side of a network and which defines a phase correction equal to Φ ₂ = π - (φ ₁ -φ ₂ + φ ₃ -φ + φ _t φ _t2 ) -φ ₁ + 2kπ with φ ₃ and φ _4; counterpart of φ! and φ ₂ , while φti and φt2 represent the phase shift undergone by the transmitted signals.