Device for optically processing incident light radiation that has propagated through free space

Abstract

The invention relates to an optical processing device (DR) for optically processing incident light radiation (I0) that has propagated through free space and whereof the phase front is liable to be distorted. The optical processing device (DR) comprises a modal conversion device (M) configured to decompose at least one portion of the incident light radiation (I0) into at least two elementary light radiations (R1-RN) and an integrated photonic device (C) configured to coherently recombine the at least two elementary light radiations (R1-RN) and produce, over an output optical port (P2), at least one recombined light radiation (Rc). The elementary light radiations (R1-RN) propagate, respectively, between an input optical port (P1) and an output optical port (P2) along at least two separate optical paths. The optical processing device (DR) is characterised in that at least one of the optical paths comprises at least one static delay element (D1, D2, D3) configured to equalise the propagation time of the elementary light radiations along the optical paths.

Description

device for optical processing of incident light radiation propagated in free space FIELD OF THE INVENTION

[0001] The invention relates to a device for optical processing of incident light radiation propagated in free space and whose phase front is likely to be distorted. This distortion may originate from atmospheric disturbances during optical communication in free space. More generally, this distortion is caused by the propagation of light radiation in its medium. The invention finds an application in the field of optical telecommunications in free space. TECHNOLOGICAL BACKGROUND OF THE INVENTION

[0002] In free-space optical telecommunications, a transmitter modulates light radiation (usually produced by a laser) to transmit information, the light radiation taking the form of a narrow beam that is emitted towards a receiver. After propagation in its medium (air will be taken as an example in the rest of this description, but the medium can be of any nature, such as water in the case of underwater telecommunications), the light radiation is collected and demodulated to recover the transmitted information.

[0003] The propagation of light radiation subjects this radiation to disturbances in the atmosphere. These erratic disturbances, whose variation dynamics are of the order of kHz, lead to its deformation, which affects its wavefront. This deformation materializes in the form of so-called "speckle" figures in the spot formed by the projection of the beam onto the radiation collection device and by a flickering phenomenon. It leads to limiting the information flow of the link between the transmitter and the receiver, to deteriorating the quality of the transmission, or even to making the link unavailable.

[0004] To overcome this limitation, the state of the art proposes to decompose modally, by means of a modal conversion device, the incident light radiation into a plurality of elementary light radiations, and to recombine in a coherent manner, by means of an integrated photonic device, these elementary light radiations, for example to inject this recombined light radiation into a single-mode optical fiber in order to guide it towards a detector / demodulator. In this regard, reference may be made to documents US7974543 or EP3672109A1.

[0005] The coherent combination of light radiation naturally requires that these radiations be coherent with each other, which requires that the length differences between the different optical paths (i.e. in a homogeneous medium, the distance traveled by the radiation multiplied by the refractive index that this radiation encountered during its journey) along which the elementary light radiations propagate are less than the coherence length of these radiations. This coherence length is notably dictated by the spectral width of the incident light radiation (more precisely this coherence length evolves as the square of the wavelength of the carrier divided by the spectral width).

[0006] It is well established that the spectral width of the modulated light radiation increases with the information rate to be transmitted. The spectral efficiency of the modulation technique used is thus measured as the ratio between the transmitted information rate (in bit / s) and the spectral width (or spectral congestion, expressed in Hz) of the modulated light radiation. For example, this spectral efficiency is of the order of ½ in the case of amplitude modulation of the ASK type (Amplitude Shift Keying), which means that the spectral width increases linearly, with a ratio of 2, with the information rate.

[0007] Transmission at a relatively high data rate therefore increases the spectral width of the incident radiation and decreases its coherence length, which requires very high precision control of the length of the optical paths in the incident light radiation processing device. For example, a data rate of 50 Gbauds / s may require a coherence length of around 100 micrometers or a few hundred micrometers. Providing an incident light radiation processing chain in which each element is configured to perfectly control the spread of the lengths of the different optical paths that pass through it is particularly delicate.

[0008] In the solutions proposed by documents EP3672109A1 and the publication by Vincent Billault et al "Free space optical communication receiver based on a spatial demultiplexer and a photonic integrated coherent combining circuit," Opt. Express 29, 33134-33143 (2021), the modal conversion device and the integrated photonic device are optically coupled to each other via a plurality of single-mode fibers. This fiber link makes it possible to move part of the device (the integrated photonic device) away from the telescope or an equivalent means of collecting the incident radiation, which may be important for practical reasons of implementing this solution.

[0009] The static and dynamic delays introduced by the differences in length of these fibers, by the chromatic dispersion of the incident radiation through the different media it passes through (atmosphere, fibers, etc.) are compensated by multiple variable delay lines combined with multiple adjustable phase shifts arranged in the integrated photonic device. These variable delay lines and adjustable phase shifts must be controlled dynamically, in real time, which can be difficult to achieve. SUBJECT OF THE INVENTION

[0010] An aim of the invention is to propose an optical communication system addressing, at least in part, this problem. More particularly, an aim of the invention is to propose an optical processing device for a free-space telecommunications system allowing a high transmission rate without, however, requiring the dynamic control of a large number of variable delay lines. BRIEF DESCRIPTION OF THE INVENTION

[0011] In order to achieve this aim, the subject of the invention proposes a device for optical processing of incident light radiation having propagated in free space and the phase front of which is likely to be distorted, the optical processing device comprising: an optical input port for receiving the incident light radiation; a modal conversion device optically coupled to the optical input port by a shift waveguide formed of at least one multimode fiber, the modal conversion device being configured to decompose at least part of the incident light radiation into at least two elementary light radiations;an integrated photonic device optically coupled, in free space and without fiber intermediary, to the modal decomposition device, the integrated photonic device being configured to coherently recombine the at least two elementary light rays and produce, on an optical output port, at least one recombined light ray.;

[0012] The elementary light rays propagate respectively, between the input optical port and the output optical port of the device, along at least two distinct optical paths.

[0013] According to the invention, at least one of the optical paths comprises at least one static delay element configured to equalize the propagation time of the elementary light rays along the optical paths and the device is devoid of any adjustable delay compensation element.

[0014] According to other advantageous and non-limiting characteristics of the invention, taken alone or in any technically feasible combination: the modal conversion device consists of at least one multi-plane conversion device comprising a plurality of optical parts each having a reflective face for guiding the propagation of the incident light radiation, at least one of the reflective faces being microstructured for decomposing the incident light radiation into at least two elementary light radiations during a plurality of reflections; the static delay element is integrated into the integrated photonic device; the integrated photonic device comprises a plurality of waveguides for propagating the elementary light radiations towards an optical combiner, two waveguides of the plurality of waveguides having different lengths, the difference in length constituting the static delay element;the optical processing device comprises a shift waveguide, formed of at least one multimode fiber, having at least two guided modes coupled to the input optical port to guide the incident light radiation towards the modal decomposition device; the shift waveguide and the modal decomposition device are configured to introduce a time spread less than a threshold value; the threshold value is less than 0.5 ps or 0.2 ps; the modal decomposition device is configured to decompose the incident radiation according to the guided modes of the shift waveguide; the shift waveguide is formed of a multimode fiber and the multiplane conversion device has a plurality of input modes, the multiplane conversion device being configured so that each input mode is formed of a linear combination of modes of the multimode fiber, these modes having similar group velocities;the multi-plane conversion device is configured so that each input mode is formed from a linear combination of the modes of the same group of modes of the multi-mode fiber; the modal decomposition device is directly assembled to the integrated photonic device.;

[0015] According to another aspect, the subject of the invention proposes an optical telecommunication system for processing incident light radiation produced by a transmitter and carrying, by modulation, information to be transmitted, the optical telecommunication system comprising an optical processing device as described above optically coupled to an optical receiver.

[0016] The optical receiver may be configured to demodulate the recombined light radiation when the incident light radiation has an information rate to be transmitted greater than 10 Gbit / s or, preferably, greater than 50 Gbit / s.

[0017] The modulation produced by the transmitter can be carried out at a working wavelength. BRIEF DESCRIPTION OF THE FIGURES

[0018] Other characteristics and advantages of the invention will emerge from the detailed description of the invention which follows with reference to the appended figures in which:

[0019]

[0020]

[0021]

[0022] Figures 1a, 1b, 1c respectively represent optical processing devices according to the invention.

[0023]

[0024] The represents a photonic device of an optical processing device according to the invention;

[0025]

[0026]

[0027] Figures 3a, 3b represent two examples of implementation of the invention;

[0028]

[0029] The figure represents an example of application of the optical processing device according to the invention;

[0030] Illustrates an optical combiner based on a Mac Zehnder architecture. DETAILED DESCRIPTION OF THE INVENTION

[0031] Figures 1a, 1b, 1c represent modes of implementation of a DR optical processing device.

[0032] This device is designed to process incident light radiation I0 that has propagated in free space and whose phase front is likely to be distorted. This incident radiation I0 is collected on an optical input port P1 of the optical processing device DR.

[0033] With reference to figures 1a, 1b, 1c the optical processing device DR comprises a modal decomposition device M, optically connected to the input optical port P1. The modal decomposition device is configured to decompose at least part of the incident light radiation I0 into at least two elementary light radiations R1-R N Advantageously, these elementary light rays are each composed of a single optical mode.

[0034] For simplicity of representation and expression, only two elementary light rays are described and represented, but an optical processing device DR in accordance with the invention makes it possible to decompose the incident light radiation I0, at least in part, using any number of elementary light rays R1-R N , for example a dozen, several dozen or even several hundred elementary light rays R1-R N .

[0035] The modal decomposition device M can be implemented by a multiplane optical conversion device, referred to as an “MPLC device” in the remainder of this description. It is recalled that in such an MPLC device, incident light radiation undergoes a succession of reflections and / or transmissions, each reflection and / or transmission being followed by propagation of the radiation in free space. At least some of the optical parts on which the reflections and / or transmissions take place, and which guide the propagation of the incident radiation, have microstructured zones which modify the incident light radiation.

[0036] By "microstructured zone" we mean that the surface of the optical part has a relief on this zone, which can for example be broken down into the form of "pixels" whose dimensions can be between a few microns and a few hundred microns. These can be metasurfaces. The relief or each pixel of this relief has a variable elevation relative to a mean plane defining the surface in question, of at most a few microns or at most a few hundred microns. Whatever the nature of the microstructuring of the zones, an optical part having such zones forms a phase mask introducing a local phase shift within the transverse section of the radiation which is reflected or transmitted there.

[0037] Thus, light radiation propagating within an MPLC device undergoes a succession of local phase shifts separated by propagations. The succession of these elementary transformations (for example at least four successive transformations such as 8, 9, 10, 12, 14, or even at least 20 transformations) establishes a global transformation of the spatial profile of the incident radiation. It is thus possible to configure the microstructured reflection or transmission zones to transform a first light radiation, which in particular has a specific shape, into a second radiation whose shape is different.

[0038] Theoretical foundations and practical implementation examples of an MPLC device can be found in the documents “Programmable unitary spatial mode manipulation”, Morizuret Al, J.Opt. Soc. Am. A / Vol. 27, No. 11 / November2010; N. Fontaine et Al, (ECOC, 2017), “Design of High Order Mode-Multiplexers using Multiplane Light Conversion”; US9250454 and US2017010463.

[0039] As presented in detail in the aforementioned documents, the microstructured zones carried by the optical part(s) forming the MPLC device are designed and configured to carry out a modal conversion aimed at decomposing a first light radiation received at the input port according to a family of modes called "input". The energies present in the modes of the input family are transported and respectively conformed to the modes of a family of "output" modes at the output port of the MPLC device. The MPLC device is configured to respectively match the modes of the input base and the modes of the output base. It is a passive device whose transfer function is particularly stable and robust and which, moreover, does not affect or very little affects the state of polarization of the light radiation passing through it.

[0040] In the context of the present description, and by way of example, the family of input modes may comprise a Hermite-Gauss basis formed of N Hermite-Gauss modes, arranged spatially opposite the incident radiation I0 when the latter is collected on the input port P1. The family of output modes may be formed of N Gaussian modes, spatially separated, these modes defining the elementary radiations R1-R N . The MPLC device is configured to associate a Hermite-Gauss mode of the input basis with a Gaussian mode of the output basis. The energy of the incident radiation I0 received at the input port P1 is decomposed according to the modes of the input basis and transported into the MPLC device to distribute and conform to the output Gaussian modes with which the modes of the input basis are associated.

[0041] Of course, the Hermite-Gauss and Gaussian modes taken as examples are given only for illustration purposes and other modes could be chosen to carry out the decomposition.

[0042] The modal decomposition device M may comprise other elements than the MPLC device detailed above. In particular, it may comprise at least one optical element in transmission or reflection allowing the shaping of the incident radiation I0 or the elementary light radiation R1-R N , for example one or a plurality of free-form optics (“free form optic” according to the accepted Anglo-Saxon expression).

[0043] Regardless of how the modal decomposition device M is implemented, the elementary light radiations R1-R Npropagate along separate optical paths, which may have different lengths. This phenomenon introduces differences in the propagation time of the incident light radiation I0 in the optical processing device DR, which imposes a significant coherence length and / or limits the transmission rate as explained in the introduction to this application.

[0044] Returning to the description of figures 1a, 1b, 1c, the optical processing device DR shown also comprises an integrated photonic device C. This is optically coupled to the modal decomposition device M in order to receive the (at least) two elementary light rays R1-R N The integrated photonic device C is configured to coherently recombine these two elementary light rays R1-R N and produce, on an optical output port P2, a recombined light radiation R cThis recombinant light radiation Rc is advantageously single-mode or has a reduced number of modes, much lower than the number N of elementary light radiations R1-R N . This reduced number of modes is for example less than 3 or 5. It can be supplied to an optical receiver or to any other instrument arranged downstream, for example by simple propagation in free space or, preferably, by means of a fiber Fs as shown in figures 1a, 1b, 1c. The fiber can be single-mode Fs when the recombined light radiation Rc is itself single-mode. It can be a fiber with few modes ("few mode fibers" according to the established Anglo-Saxon expression) when this combined radiation Rc has more than one mode.

[0045] The photonic device C can be in the form of an integrated photonic chip (or a plurality of integrated photonic chips). The chip is then formed of waveguides WG making it possible to guide two elementary radiations R1-R N which are presented at its input port to an optical combiner Co allowing their coherent recombination to produce the recombined radiation Rc. The optical combiner Co can be implemented by any suitable technique. It can in particular be an optical combiner Co having a Mach-Zehnder architecture as shown in the.

[0046] The optical combiner Co shown in this figure is formed by two arms connecting separators BS arranged in series. Each arm is equipped with a controllable phase shifter PS. At the input, the optical combiner Co receives, respectively on the two arms, the two elementary radiations R1-R Nto be combined. At the output, the optical combiner provides, on one of the two arms, the recombined radiation Rc. The other arm is connected to a PD photodetector allowing the controllable phase shifters PS to be controlled and to form the recombined radiation Rc at the output of the optical combiner Co. This regulation of the PS phase shifters aims to minimize, or more generally to optimize, the signal measured on the PD photodetector. It can be implemented by an integrated M computer or connected to the photonic device.

[0047] When the photonic device C aims to recombine more than two elementary radiations R1-R Nit is planned to equip this device with several cascaded stages of optical combiners Co, each stage being composed of optical combiners each recombining two radiations from the previous stage. Such an example of a photonic device is shown schematically on the. A computer, not shown, integrated or connected to the photonic device C, makes it possible to operate the optical combiners and form the coherent recombination of all the elementary radiations R1-R N supplied as input to device C.

[0048] The invention is of course in no way limited to the example of implementation of the integrated photonic device C which has just been presented by way of illustration. Generally speaking, this integrated photonic device can be implemented by any technique making it possible to coherently recombine elementary light radiations R1-R N and produce, on an optical output port P2, a light radiation Rc recombining at least part of these elementary radiations.

[0049] In the embodiment shown in the, the modal decomposition device M is optically coupled to the photonic device C by a plurality of waveguides, each guide making it possible to guide one of the elementary radiations R1-R N This type of coupling can be achieved, for example, by a bundle of single-mode optical fibers. In some cases, these fibers can be polarization-maintaining.

[0050] But these waveguides are disadvantageous because, as noted in the prior art documents, they lead to the introduction of optical path length deviations that are not controlled. These length deviations are linked to the variability of the waveguide lengths, which cannot in practice be chosen or formed to be exactly identical. The uncertainty concerning the length deviations of a given DR optical processing device therefore does not allow the spread of the lengths of the different optical paths that pass through it to be known in advance. This requires, as proposed in the prior art documents, the provision of adjustable delay lines.

[0051] Also, very advantageously and preferably, the optical processing device DR is devoid of any waveguide between the modal decomposition device M and the photonic device C. This coupling can be achieved by propagation in free space. Illustrates such an implementation mode, directly assembling the modal decomposition device M and the photonic device C to each other, without a fiber intermediary. The modal decomposition device M and the photonic device C can also be precisely positioned relatively to each other, to allow their coupling by propagation in free space, by assembling them to a common support.

[0052] In the embodiment shown in the, in which the modal decomposition device M and the photonic device C are optically coupled by propagation in free space, provision has been made to integrate an offset waveguide Fd between the input optical port P1 of the optical processing device DR and the modal decomposition device. This offset guide Fd in some way compensates for the absence of a fiber connection between the modal decomposition device M and the photonic device C to allow the modal decomposition device M and the photonic device C to be offset from the location of collection of the incident radiation I0. The offset waveguide Fd may be composed of a multimode optical fiber or a bundle of multimode optical fibers.

[0053] The fiber or plurality of fibers that make up the Fd offset waveguide may be of any suitable type. It may thus be a circular multimode graded-index fiber (for example, a 50 micrometer or 62.5 micrometer core and a 125 micrometer cladding, of the OM1, OM2, OM3, OM4, OM5 type), a circular multimode step-index fiber, a gradient or step-index elliptical core fiber, a spun fiber, a gradient or step-index fiber with stress bars (of the panda or bow tie type), a photonic crystal fiber or a hollow core to benefit from their low dispersion, a multicore fiber (gradient or step-index per core, each core being able to be elliptical or circular, with or without stress bars.In the case of a multi-core fiber, each core can be optically associated with a mode of the family of input modes of the MPLC device forming the modal conversion device M.

[0054] This fiber or this plurality of fibers may be polarization-maintaining, if it is important to control this polarization during the optical processing carried out by the DR device. In this case, it may also be possible to provide the input optical port P1 with a device for conditioning the polarization of the incident light radiation. This device aims to conform the polarization of the incident light radiation to the maintained polarization states of the offset waveguide Fd. It is particularly useful when the polarization of the incident radiation I0 is not perfectly controlled and it is therefore not possible to directly inject the incident radiation into the offset waveguide Fd without risking affecting the polarization of the radiation propagating therein.

[0055] It is noted that the offset waveguide Fd affects the propagation of the incident radiation I0 by modal dispersion. Consequently, the optical path traveled by certain modes is different from the optical path traveled by other modes, these optical paths being able to have different lengths. This phenomenon affects the necessary coherence length of the radiation propagating there. The propagation deviations which affect the modes propagating in the waveguide Fd are added to those potentially introduced by the modal decomposition device M.However, and unlike the case of a waveguide connection between the modal decomposition device M and the photonic device C, the spreading of the lengths of the different optical paths caused by this dispersion is perfectly known in advance and therefore controlled, because it depends on the intrinsic characteristics of the offset waveguide Fd, in particular its index profile, its length and the nature of the materials which compose it.

[0056] The modal dispersion induced by the offset waveguide Fd tends to distort, by temporal spreading, a symbol carried by the incident radiation which propagates through the offset waveguide Fd and the modal conversion device M. This spreading can be characterized by measuring, on each of the outputs of the modal conversion device M, the duration of the optical pulses having propagated to these outputs, after having injected an optical excitation pulse at the free end of the offset waveguide Fd. For each of these outputs, a spreading value can be established, this value corresponding to the difference between the measured duration of the pulse on the output considered and the duration of the optical excitation pulse. The temporal spreading to be considered is then the maximum spreading value on all the outputs.

[0057] Also, and in general, we seek to ensure that this time spread is less than a threshold value. As an illustration, we will seek to limit the time spread below a threshold value of 0.5 ps, to allow wavelength multiplexed transmission, the transmission wavelengths being chosen over a wavelength range of 20 nm centered at 1550 nm. We will seek to limit the time spread below a threshold value of 0.2 ps for a wavelength range of 40 nm centered at 1550 nm (C band).

[0058] Such a result can be achieved when the offset waveguide Fd is relatively short, typically less than 5 m, or exhibits little modal dispersion. It can therefore be a graded-index multimode optical fiber of the OMx type. This fiber can also have an elliptical core.

[0059] Alternatively, or in addition, and when the modal conversion device M is formed of an MPLC device, the latter can be configured so that each mode of the input mode family is chosen, during the design of this device, as a linear combination of modes having similar group speeds, in such a way that this mode is only slightly deformed during its propagation in the offset waveguide Fd. By similar, we mean that these speeds are identical to within 10%, advantageously to within 5%.

[0060] For example, each mode of the input mode family can be chosen as a linear combination of the modes of the same group of modes of the multimode fiber. It is recalled that the modes of the same group of modes of a fiber propagate with identical speeds. With this configuration, one avoids constituting an elementary radiation produced by the MPLC device from modes of the multimode fiber having propagated with different speeds. As a further illustration of such a configuration, the modal decomposition device M can be configured to decompose the incident radiation I0 according to guided modes of the offset waveguide Fd.

[0061] As seen in the introduction to this application, the spread of the lengths of the different optical paths which pass through the optical processing device DR, between the input optical port P1 and the output optical port P2, leads to imposing a relatively large coherence length. Consequently, and when the optical processing device DR is used in a free-space optical telecommunications application, the transmission rate is limited.

[0062] Also, in a device in accordance with the present description, provision is made to place, in at least one of the optical paths traveled by the two elementary light rays R1-R N, a static delay element. For the sake of clarity, it is specified that these optical paths extend from the input port P1 of the optical processing device DR to its output port P2. The static delay introduced by this element is chosen to equalize the propagation time of the elementary light rays along the optical paths, i.e. to reduce the deviations in propagation time of the elementary light rays R1-R Npropagating along these paths at the working wavelength, i.e. a wavelength contained in a wavelength range of interest. In this way, the need for coherence length is limited, which makes it possible to preserve the optical transmission rate. By "static delay" we mean a fixed and unmodifiable delay. To avoid any doubt, it is specified that the static delay introduced by the static delay element is linked to an optical path deviation by the propagation speed of the radiation in the optical processing device DR. In practice, we can seek to equalize the length of the different optical paths so that they differ by a maximum of less than 3000 microns at most, in order to allow high-speed transmission, greater than 10 Gbit / s at a single working wavelength.

[0063] When the transmission uses several working wavelengths, as is the case in the context of WDM transmission, the aim will be to more strictly reduce the difference between the different optical paths. Thus, to allow transmission at two wavelengths, for example 1553.3 nm and 1536.6 nm, the aim will be to equalize the length of the different optical paths so that they differ by 120 microns or less. More generally, a transmission operated by a device according to the invention can use a single wavelength, two wavelengths or more.

[0064] It is noted that the difference in the lengths of the plurality of optical paths in the offset waveguide Fd or in the modal conversion device M are particularly stable over time. They can be characterized and measured simply, for example by measuring the propagation deviation of the radiation propagating on these different paths, and be compensated by the static delay element, without needing to adjust this delay over time. The characterization techniques can implement white light interferometry or with a wavelength-tunable laser or a "time-of-flight" measurement via the emission of very short pulses or even via sinusoidal modulated sources.As already noted, the difference in the lengths of the plurality of optical paths is also stable from one optical processing device to another, in particular because any fiber connection between the modal conversion device M and the integrated photonic device has been eliminated. Also, it is perfectly possible to compensate for the difference in length by a static, fixed and non-modifiable delay element.

[0065] According to a first approach, the static delay element D1 is integrated into the integrated photonic device C. Thus, the two waveguides WG of the integrated photonic device C which respectively propagate the two elementary light rays R1-R N supplied by the modal decomposition device M to an optical combiner Co are designed to have different lengths, the difference in length constituting the static delay element D1.

[0066] In practice, this involves differentiating the lengths of the waveguides WG in order to try to equalize the lengths of the optical paths which extend between the input optical port P1 and the output optical port P2 of the optical processing device DR.

[0067] When the integrated photonic device C recombines more than two elementary light rays R1-R N , it was previously indicated that this device could then be provided with a plurality of stages of optical combiners Co, the stages being arranged in cascade. It is of course necessary in this case to equalize the optical paths extending between the optical input port P1 and each optical combiner Co, in order to allow this coherent recombination.

[0068] According to the first solution, shown in the figure, the length of all the waveguides (except, possibly, the one identified as being the longest path) of a first stage (the so-called "input" stage) of optical combiners Co is adjusted. This adjustment makes it possible to equalize all the propagation times of the elementary light rays R1-R N along the optical paths extending from the input optical port P1 to the first stage of optical combiners Co. The static delay elements D1 are therefore placed on each of the waveguides (except, possibly, on the one forming part of the longest optical path) leading to the optical combiners Co of the input stage. The waveguides connecting the stages together in this case all have the same length, which there is no need to adjust.

[0069] According to an alternative solution, shown in the figure, the length of a waveguide of each optical combiner Co is adjusted. This adjustment makes it possible to equalize the propagation times of the elementary light rays R1-R N along the optical paths extending from the input optical port P1 to each optical combiner Co.

[0070] In a second approach, the static delay element D2 is integrated into the modal decomposition device M. This can be obtained by forming an elevation gap (with respect to a principal plane) of one of the reflecting faces, a gap on which one of the two elementary light rays R1-R is reflected. N. This elevation difference leads to modifying the length of the optical path of the elementary light radiation which is reflected there. More generally, the modal decomposition device M may comprise a plurality of elevation differences (relative to the main plane forming the reflecting face) on which at least part of the elementary light radiation R1-R is reflected. N These elevation differences are configured to equalize the propagation time of the elementary light rays R1-R N along the optical paths extending between the input optical port P1 and the output optical port P2.

[0071] Whatever the solution chosen, the static delay element or the plurality of static delay elements introduced into the optical processing device DR makes it possible to preserve the compatible coherence length of the incident radiation I0 which propagates therein, which facilitates the coherent recombination of the elementary light radiations. The optical paths extending in the optical processing device DR have equalized lengths, that is to say that the differences in lengths which may be present are less significant than in the absence of the static delay element or the plurality of these elements.

[0072] The represents an optical telecommunications system 1 implementing an optical processing device DR according to the invention.

[0073] The optical telecommunications system 1 is intended to process incident light radiation I0 produced by a transmitter and carrying, by modulation, information to be transmitted. The incident light radiation I0 may have a working wavelength or several wavelengths, as is usually the case for WDM type transmissions, and / or exploit several polarizations. The recombined light radiation Rc is supplied to an optical receiver OR, capable of extracting the information from the received radiation. This optical receiver OR is in particular configured to demodulate the recombined light radiation Rc when the transmission (the modulation) is at very high speed, for example a speed of 10 Gbit / s or 50 Gbit / s, or even several hundred Gbit / s.

[0074] In the example shown in this figure 1, the transmitter is arranged in a satellite SAT, but in general, the transmitter can be arranged indifferently on land, in the sea or in space, and propagate in any free space, the atmosphere in the case of terrestrial communication, water in the case of marine communication. The transmitter and the telecommunication system 1 can both be stationary, or move relative to each other.

[0075] The incident light radiation I0 emitted takes the form of a narrow beam directed towards the telecommunications system 1. During its propagation in free space, the emitted radiation is subject to atmospheric disturbances in the atmosphere, so that the incident light radiation I0 arriving at the base station exhibits spatial and temporal fluctuations in amplitude and phase. This phenomenon affects the shape of this radiation, which takes a form that varies over time, in an erratic and irregular manner. The optical telecommunications system 1 aims to compensate, at least in part, for this distortion to allow the exploitation of the radiation by the optical receiver OR and the decoding of the transmitted message, by direct or coherent detection. For this purpose, it may be provided that the receiver OR integrates amplification and / or spectral demultiplexing functions, in particular in the context of WDM transmission.

[0076] The optical telecommunication system 1 comprises a telescope T, this telescope T having an objective O for collecting the incident light radiation I0 and producing it at the optical input port P1 of the optical processing device DR. As is well known per se, this objective may comprise a concave mirror focusing the received radiation into an image focus. This convergent radiation may be returned to the optical input port P1 using a second mirror of the objective O, this second mirror being able to be planar or convex. The telescope T may be orientable in order to point and track the transmitter, here arranged in the satellite SAT.The telescope T may also comprise a TTM guidance device for the incident radiation (such as a tilting mirror or "tip tilt mirror" according to the English expression usually used in the field) in order to best guide the light radiation towards the optical port P1, for example to center this radiation in the port P1 and, more generally, to correct any deviations in the pointing of the telescope T.

[0077] Finally, the optical telecommunications system 1 also comprises, optically downstream of the telescope T, the optical processing device DR in accordance with what has just been described. This device aims to compensate, at least in part, the distortions of the incident light radiation I0.

[0078] The optical processing device DR allows, by modal decomposition of the incident radiation I0 and the coherent recombination of the elementary radiations R1-R Nproducts, to exploit a maximum of the energy of the incident light radiation I0 collected and to compensate, at least in part, the distortion undergone by this radiation during its propagation in free space. The elementary radiations R1-R N propagating along optical paths of substantially the same lengths, the coherence distance of the incident radiation I0 is little affected. Consequently, the elementary radiations R1-R N can be coherently recombined with each other, i.e. form a single-mode recombined light radiation Rc presenting a maximum energy, even when the incident radiation I0 results from a high-rate modulation.

[0079] Of course, the invention is not limited to the embodiments described and variant embodiments can be made without departing from the scope of the invention as defined by the claims.

[0080] Finally, it is noted that a processing device in accordance with the present invention results from an approach completely opposite to that presented in the state of the art. Indeed, according to the invention, it is sought to form a processing device whose disparity in optical path length is perfectly controlled. This disparity being controlled, in particular from one device to another in the context of volume production, it is possible to incorporate at least one static delay element in the device aimed at equalizing the length of the optical paths and limiting their spread. It is then not necessary to provide dynamic compensation or by calibration of the path lengths using an adjustable delay compensation element, as is required in the devices of the state of the art. Also, and very advantageously, a processing device in accordance with the invention is devoid of any adjustable delay compensation element.

[0081] When the spread is reduced, to less than 3000 micrometers, transmission with a rate exceeding 10 Gbits can be implemented, as previously presented. The smaller this spread, the higher the transmission rate can be and / or the more it becomes possible to extend the wavelength range used in WDM transmission.

Claims

Optical processing device (DR) for incident light radiation (I0) having propagated in free space and whose phase front is likely to be distorted, the optical processing device (DR) comprising: an input optical port (P1) for receiving the incident light radiation (I0); a modal conversion device (M) optically coupled to the input optical port (P1) by a shift waveguide (Fd) formed of at least one multimode fiber, the modal conversion device (M) being configured to decompose at least part of the incident light radiation (I0) into at least two elementary light radiations (R1-R N ); an integrated photonic device (C) optically coupled, in free space and without fiber intermediary, to the modal decomposition device (M), the integrated photonic device (C) being configured to coherently recombine the at least two elementary light rays (R1-R N) and produce, on an optical output port (P2), at least one recombined light radiation (R c ); elementary light radiation (R1-R N ) propagating respectively, between the input optical port (P1) and the output optical port (P2), along at least two separate optical paths, the optical processing device (DR) being characterized in that at least one of the optical paths comprises at least one static delay element (D1, D2, D3) configured to equalize the propagation time of the elementary light radiation along the optical paths and in that it is devoid of any adjustable delay compensation element. Optical processing device (DR) according to the preceding claim in which the modal conversion device (M) consists of at least one multiplane conversion device comprising a plurality of optical parts each having a reflective face for guiding the propagation of the incident light radiation (I0), at least one of the reflective faces being microstructured to decompose the incident light radiation (I0) into at least two elementary light radiations (R1-R N ) during a plurality of reflections. Optical processing device (DR) according to one of the preceding claims in which the static delay element (D2) is integrated into the modal decomposition device (M). Optical processing device (DR) according to claims 2 and 3 in which the static delay element (D2) is an elevation difference relative to the main plane forming one of the reflecting faces on which one of the elementary light rays (R1-R N ). Optical processing device (DR) according to claim 1 or 2 wherein the static delay element (D1) is integrated into the integrated photonic device (C). Optical processing device (DR) according to the preceding claim in which the integrated photonic device (C) comprises a plurality of waveguides for propagating the elementary light radiation (R1-R N ) to an optical combiner (C0), two waveguides of the plurality of waveguides having different lengths, the difference in length constituting the static delay element (D1). Optical processing device (DR) according to one of the preceding claims comprising a displacement waveguide (Fd), formed of at least one multimode fiber, having at least two guided modes coupled to the input optical port (P1) to guide the incident light radiation (I0) towards the modal decomposition device (M). Optical processing device (DR) according to the preceding claim in which the offset waveguide (Fd) and the modal decomposition device (M) are configured to introduce a temporal spread less than a threshold value. Optical processing device (DR) according to the preceding claim wherein the threshold value is less than 0.5 ps or 0.2 ps. Optical processing device (DR) according to one of claims 7 to 9 in which the modal decomposition device (M) is configured to decompose the incident radiation (I0) according to the guided modes of the offset waveguide (Fd). Optical processing device (DR) according to one of claims 7 to 9 combined with claim 2 in which the offset waveguide (Fd) is formed from a multimode fiber and the multiplane conversion device has a plurality of input modes, the multiplane conversion device being configured so that each input mode is formed from a linear combination of modes of the multimode fiber, these modes having similar group velocities. Optical processing device (DR) according to the preceding claim in which the multiplane conversion device is configured so that each input mode is formed from a linear combination of the modes of the same group of modes of the multimode fiber. Optical processing device (DR) according to one of the preceding claims in which the modal decomposition device (M) is directly assembled to the integrated photonic device (C). Optical telecommunication system for processing incident light radiation (I0) produced by a transmitter and carrying, by modulation, information to be transmitted, the optical telecommunication system comprising an optical processing device (DR) according to one of the preceding claims optically coupled to an optical receiver OR. Optical telecommunication system according to the preceding claim in which the optical receiver (OR) is configured to demodulate the recombined light radiation (Rc) when the incident light radiation (I0) has an information rate to be transmitted greater than 10 Gbit / s or, preferably, greater than 50 Gbit / s.

Images