RU2804262C1 - Method for amplitude, phase and polarization control in phased array of fibre amplifiers and control of distributed state of intensity, wave front and polarization of synthesized beam in the far optical field and device for its implementation - Google Patents

Abstract

FIELD: laser technology and fibre optics.

SUBSTANCE: invention can be used to create systems for transmitting light energy through free space, wireless optical communications and cryptography, and laser processing of materials. The device comprises a coherent source of linearly polarized radiation, which is divided by a radiation divider into N channels of equal power. Each channel is connected in series with an optical phase-shifting element (phase modulator) that regulates the phase of the optical wave, a polarization controller that regulates the polarization state of the optical wave, a power amplifier with controlled gain based on an active fibre that does not maintain the polarization state, and has an optical collimator at the output. All N optical collimators are located at the lattice nodes and are adjusted so that the optical axes of the output beams are parallel to each other, forming a lattice of beams (subbeams), a small part of the radiation of which is removed using a beam splitter to form feedback channels, and the remaining part of the radiation propagates in in a given direction, forming a synthesized beam in the far optical field, resulting from diffraction overlap and interference of subbeams. The part of the radiation that is used to generate feedback is in turn divided in half by a second beam splitter, forming two feedback channels. The amplitude of the subbeams is controlled by power amplifiers with adjustable gain, which are located after the phase modulators and polarization controllers. Thus, the phase, direction of linear polarization and amplitude of the grating subbeams are independently controlled, the joint action of which leads to the formation in the far optical field of a synthesized beam with a given non-uniform distribution of intensity, wave front and polarization in the cross section.

EFFECT: formation of a synthesized beam with a controlled structure of intensity, wavefront and polarization, by independently controlling the amplitude, phase and direction of linear polarization of individual beams in an array of fibre amplifiers.

5 cl, 17 dwg

Description

Field of technology to which the invention relates

The invention relates to laser technology and fiber optics and can be used to create systems for transmitting light energy through free space, wireless optical communications and cryptography, and laser processing of materials. The invention can find application in various fields of technology where laser beams with non-uniform intensity, wavefront and polarization distributions and control of these distributions are required.

The proposed method and device are based on the principles of optical phased arrays, polarization control and adaptive optics.

Optical phased array systems are typically used to coherently sum unidirectionally polarized laser beams to achieve maximum far-field power density and are the subject of a number of Russian and foreign patents (US8548017 B1, US7058098 B1, US7187492, RU2470334, RU2488862). All of these patents differ mainly in the way they organize a feedback circuit to isolate a signal that controls the phasing elements that regulate the current phases of individual elements in the array, with the goal of bringing all elements in the array into a state with the same phases, as well as the presence or absence of a reference channel.

In a number of patents, to achieve co-phase unidirectional polarization in an array of non-polarized amplifiers, the principles of continuous monitoring of the polarization state in the feedback circuit are used, which, along with phasing elements to bring all radiation channels into a co-phase state, uses polarization controllers that regulate the current state of polarization in individual array channels in order to bringing all radiation channels into a state with unidirectional polarization (US6317257 B1, US8922771 B2, US8922772 B2, WO2015023335). In these patents, the presence of a reference linearly polarized beam is required, which sets the direction of polarization, common to all beams of the grating, and there is no possibility of controlling the polarization state of the synthesized beam.

The closest to the claimed invention is RF patent No. 2716887 “Method of forming a laser beam with an arbitrarily specified intensity distribution in the far optical field and a device for its implementation.” This patent proposes a method and device for forming in the far optical field a beam synthesized by an array of fiber amplifiers with a controlled intensity distribution by controlling the phase and amplitude of each individual beam (subbeam) in the array. The disadvantage of this method is the requirement that all subbeams in the initial state have unidirectional linear polarization, which does not allow the formation of beams with a controlled polarization structure.

The problem to be solved by the claimed invention is to create a method for amplitude, phase and polarization control in a phased array of fiber amplifiers and control of the distributed state of intensity, wavefront and polarization of a synthesized beam in the far optical field and a device for its implementation.

The technical result consists in the formation of a synthesized beam with a controlled structure of intensity, wavefront and polarization, by independently controlling the amplitude, phase and direction of linear polarization of individual beams in an array of fiber amplifiers.

The device contains a coherent source of linearly polarized radiation, which is divided by a radiation divider into N channels of equal power. Each channel is connected in series with an optical phase-shifting element (phase modulator) that regulates the phase of the optical wave, a polarization controller that regulates the polarization state of the optical wave, a power amplifier with controlled gain based on an active fiber that does not maintain the polarization state, and has an optical collimator at the output. All N optical collimators are located at the lattice nodes and are adjusted so that the optical axes of the output beams are parallel to each other, forming a lattice of beams (subbeams), a small part of the radiation of which is removed using a beam splitter to form feedback channels, and the remaining part of the radiation propagates in in a given direction, forming a synthesized beam in the far optical field, resulting from diffraction overlap and interference of subbeams. The part of the radiation that is used to generate feedback is in turn divided in half by a second beam splitter, forming two feedback channels.

The first feedback channel contains a phase-forming element, after passing through or reflecting from which each subbeam acquires an individual additional phase shift of such a magnitude that the phase difference between this subbeam and any other subbeam of the grating corresponds to a given phase difference, and a focusing lens that collects all subbeams in the focal plane, where a photodetector is located, the field of view of which is limited by a small aperture and the signal of which is fed to an N -channel optimizing phase control processor, operating in accordance with the global maximum/minimum search algorithm (stochastic optimization) and controlling the phase-shifting elements so that all subbeams in the plane of the photodetector turn out to be in phase, and the phase differences between the subbeams in the far zone take on the required values opposite to the values of the additional phase shift specified on the phase-forming element, forming the intensity distribution and wavefront of the synthesized beam.

The second feedback channel contains a controlled N -channel polarization rotator, as a result of passing through which, each subbeam of the grating rotates the direction of its linear polarization by an angle equal to the specified one with the opposite sign, a linear polarizer that determines the unique one specified for all subbeams in the second feedback channel connection between the direction of linear polarization and an array of focusing lenses corresponding to the array of subbeams so that each subbeam is focused to its focal point, where the photodetector corresponding to this subbeam is located. The radiation power recorded by the photodetector will reach its maximum possible value when the direction of linear polarization of the subbeam passing through the polarization rotator coincides with the direction of the optical axis of the linear polarizer. Photodetector signals are fed to an N -channel optimizing polarization processor, in which each subbeam corresponds to its own channel of the optimizing processor, operating in accordance with the principle of stochastic optimization and generating control voltages for the corresponding polarization controller, which generates the state of the optical polarization of the subbeam corresponding to the maximum possible level of the photodetector signal . In this case, each subbeam of the grating acquires its required direction of linear polarization.

The amplitude of the subbeams is controlled by power amplifiers with adjustable gain, which are located after the phase modulators and polarization controllers.

Thus, the phase, direction of linear polarization and amplitude of the grating subbeams are independently controlled, the joint action of which leads to the formation in the far optical field of a synthesized beam with a given non-uniform distribution of intensity, wave front and polarization in the cross section.

Unlike the prototype:

- between the phase-shifting element (phase modulator) and the fiber amplifier in each channel there is a polarization controller, which brings the random polarization of the subbeam at the output of the fiber amplifier into a controlled state of linear polarization;

- fiber amplifiers contain an active fiber that does not maintain a polarization state;

- part of the radiation that is used to generate feedback is divided in half using the second beam splitter plate, forming a second feedback channel, where an N -channel controlled polarization rotator is located, which rotates the polarization direction of each subbeam independently of other subbeams by an angle equal to the required one, with opposite sign;

- all subbeams after the polarization rotator fall on a linear polarizer, which determines the only direction of linear polarization specified for all subbeams in the second feedback channel and an array of N focusing lenses corresponding to the array of subbeams so that each subbeam is focused on a photodetector corresponding to a given subbeam.

As a result of the introduced differences, the radiation power recorded by a separate photodetector will reach its maximum possible value when the direction of linear polarization of the subbeam passing through the polarization rotator coincides with the direction of the axis of the linear polarizer. Photodetector signals are fed to an N -channel optimizing polarization processor, in which each subbeam corresponds to its own channel of the optimizing processor, operating in accordance with the principle of stochastic optimization and generating control voltages for the corresponding polarization controller, which generates the state of the optical polarization of the subbeam corresponding to the maximum possible level of the photodetector signal . In this case, each subbeam of the grating acquires the required direction of linear polarization.

The relative amplitudes (power values) of each sub-beam are set to the required values by adjusting the power of the corresponding fiber amplifiers.

Thus, independent control of the phase difference, the direction of linear polarization and the amplitude of the grating subbeams is carried out, the joint action of which, as a result of diffraction overlap and beam interference, leads to the formation in the far optical field of a synthesized beam with the required non-uniform distribution of intensity, wave front and polarization in the cross section.

It should be noted that as a result of controlling the phase and polarization in accordance with the stochastic optimization algorithm, all phase fluctuations caused by thermal, mechanical and acoustic fluctuations of the refractive index will be synchronized in time, and all polarization changes caused by thermal, mechanical and acoustic fluctuations of the birefringence inherent fiber systems will be compensated, as happens in coherent beam combining systems of non-polarized amplifiers and continuous polarization control systems.

The technical result achieved by implementing the claimed invention is to obtain an optical beam with a controlled distribution of intensity, wavefront and polarization in the far optical field through the formation of a synthesized beam resulting from diffraction overlap and interference of N Gaussian beams forming a lattice of beams (subbeams), by controlling the amplitude of the subbeams, the phase difference between the subbeams and the direction of linear polarization of the subbeams.

The advantage of the proposed method is that, unlike known methods that form synthesized beams with unidirectional polarization, the proposed method allows the formation of beams with a controlled polarization distribution in the cross section of the beam.

New to the method is:

- control of the direction of linear polarization of individual subbeams that make up the synthesized aperture, as a result of placing an N -channel controlled polarization rotator in the feedback circuit, as a result of which the polarization direction in each channel becomes corresponding to the required polarization direction;

- formation of a synthesized beam with controlled distributions of intensity, wavefront and polarization in the cross section of the beam in the far optical field as a result of the joint action of independently controlled amplitude, phase difference between subbeams and direction of linear polarization of the grating subbeams.

One of the variants of the proposed method is a method for controlling the direction angle of uniform linear polarization of a synthesized beam formed by an array of beams with unidirectional polarization.

New to the device are:

- presence of a second feedback channel;

- the presence of an N -channel controlled polarization rotator located in the feedback circuit, using a small fraction of the power generated by the synthesized aperture, which allows scaling the power of the generated laser beam.

In one of the variants of the proposed device, a system consisting of two quarter wave plates, the main axes of which are orthogonal to each other, and a phase spatial light modulator located between them, operating on transmission, oriented at an angle of 45 degrees relative to the main ones, is used as an N- channel controlled polarization rotator axes of a quarter of the wave plates, and the action of which is organized in such a way that each subbeam in the array corresponds to its own area of pixels on the display of the spatial light modulator.

In one of the variants of the proposed device, instead of a spatial light modulator, a sectioned or stacked liquid crystal retarder (retarder) is installed.

The invention is illustrated by graphic materials.

A description of the operation of the device is given using the example of a synthesized aperture consisting of a hexagonal lattice of 6 subapertures (fiber collimators) located close to each other, however, the proposed method has no fundamental restrictions on the number of subapertures, which can only be limited by technical capabilities. Examples of synthesized apertures consisting of 6 (a), 7 (b), 19 (c), 37 (d) subapertures are presented in Fig. 1. Examples of possible distributions of linear polarization directions for the number of subapertures N =6 are presented in Fig. 2: a) unidirectional linear polarization; b) unidirectional linear polarization, rotated by 45°; c) radial distribution; d) azimuthal distribution; e) radial distribution with a rotation of directions by 20°; f) azimuthal distribution with a rotation of directions by 20°; g) hybrid radial-azimuthal distribution; h) hybrid azimuthal-radial distribution.

Carrying out the invention

The operating principle of the method and device is as follows:

The field of the synthesized aperture can be represented as follows: In the initial plane on a circle of radius R , N sources of field strength with a certain direction of linear polarization are located at equal distances, as shown in Fig. 3. This corresponds to a grating, which is a set of subapertures (subbeams) of fiber lasers emitting Gaussian beams with a radius a .

The centers of the subapertures are separated by an angle

relative to the center of the synthesized aperture, where N is the number of apertures, j is the number of subapertures. The beams have different linear polarization, which is determined by the unit vector

where α _j is the angle of inclination of the polarization vector relative to the OX axis. The subaperture position vector is given by

Where x ₀,y ₀ - basis vectors of the coordinate system. The coordinate of an arbitrary point relative to the center of the subaperture is related to the coordinate relative to the common centerO ratio

We assume that all beams propagate through a common focusing lens with a center at point O and focal length F. To determine the total field, we use the known solution for the propagation of a focused Gaussian beam along the OZ axis:

For a beam whose center is shifted relative to the center and propagates at an angle to the OZ axis, this solution in the coordinate system associated with the beam center takes the form:

where k is the wave number, A _{0 j} is the initial amplitude of the field, z is the propagation distance,

Then for the total (vector) field we can write the expression:

In the focal plane ( z = F ), the condition ρ _j ( z = F ) = ρ is satisfied, and expression (8) takes the form:

Where $$a_{0j}=A_{0j}/A$$ , φ _j is the value of the initial phase in the channel of the j -th subbeam.

Expression (9) allows, using numerical modeling methods, to determine the required set of parameters $$a_{0j}$$ , φ _j , $$r_j$$ to achieve the required field characteristics of the synthesized beam in the focal plane.

Relative amplitudes (peak power values) of subbeams $$a_{0j}$$ are controlled by changing the pump power of the fiber amplifiers and can be adjusted to the required values using an amplitude controller.

In a fiber, the phase magnitude of an electromagnetic wave changes randomly over time as a result of uncontrolled fluctuations in the fiber's refractive index caused by acoustic and thermal vibrations. In addition, the use of power amplifiers with controlled gain based on an active fiber that does not maintain a polarization state causes fluctuations in the direction of polarization of the radiation in the channel. To compensate for these fluctuations, we separate a small part of the power generated by the system from the total flow with a beam splitter plate in a ratio of 99:1 to provide feedback. Thus, the radiation flux in the far optical field (at the lens focus) will contain almost all of the generated power. In turn, we will divide this small part of the radiation in half using the second beam splitter plate and form two independent feedback channels.

In the first feedback channel, which serves to control the phase of subbeams φ _j , we introduce a phase-forming element in the form of a reflective liquid crystal spatial light modulator (SLM) or a digital micromirror device (DMD). Let us take into account that this element is exclusively phase, and accordingly does not affect the radiation intensity, and allows individual phase shifts to be independently introduced into each subbeam. Let's introduce the required phase shifts with the opposite sign in each subbundle. After SLM or DMD, we will collect all subbeams in the focal plane of the lens, where we will place a small diaphragm (pinhole), limiting the field of view of the photodetector installed behind the diaphragm. The radiation power recorded by a photodetector, the receiving area of which is limited by a small aperture, serves as a control signal (metric) of the stochastic optimization algorithm, on the basis of which an N -channel phase optimizing processor operates, generating voltages to control the amount of phase delay introduced by phase modulators for each subbeam of the grating. Due to the fact that the phase modulator introduces a phase shift of the same magnitude into both field components, and the SLM is capable of introducing a phase shift into only one field component directed along the long side of the SLM display, while the orthogonal field component is reflected from the SLM without a phase shift, in To avoid ambiguity, we will install a linear polarizer in front of the photodetector, blocking the orthogonal component of the field.

When the maximum power value at the detector is reached, the phase difference between subbeams introduced by the SLM (a given phase difference) is equal to the phase difference between subbeams of the grating in the plane of the pupil of the system, correspondingly equal to the phase difference between subbeams in the far optical field (the required phase difference) and determines the field distribution synthesized beam, which can be calculated based on expression (3).

Thus, setting the corresponding phase shifts using a phase-forming element in the first feedback channel makes it possible to obtain the required (predetermined), controlled relationships between the phases of subbeams in the subaperture plane at the required values of the directions of linear polarization of the subbeams and, accordingly, control the type of intensity distribution, wave front and polarization of the synthesized beam in the far optical field. In particular, in the case of unidirectional polarization of the initial subbeams, it is possible to form beams that have a phase singularity (a region on the beam axis in which the phase value is not defined), so-called scalar vortex beams (see Fig. 8, Fig. 9). Note that the operating principles of phase modulators based on changing the refractive index in a crystal or the length of an optical fiber under the influence of an applied voltage are well known to specialists. These devices are presented on the market, in particular, the products of the company EoSpace (www.eospace.com) can be noted.

The operation of the first feedback channel, which controls the phase distribution in the channels, is discussed above under the assumption that each subbeam of the grating is characterized by its required linear polarization slope. However, the gain medium of the fiber amplifiers used in our design is an active fiber that does not maintain a given linear polarization state, and the polarization state of each grating subbeam is characterized by its linear polarization direction randomly varying with time. To control the polarization state of a light beam, polarization controllers are widely used, the operating principle of which is based on a change in birefringence in an optical system equivalent to a set of rotating wave plates and is well known to specialists (F. Heismann and MSWhalen, “Fast automatic polarization control system”, IEEE Photonics Technology Lett ., vol.4, No.5, 503-505 (1992); R. Noé, H. Heidrich, and D. Hoffman, “Endless polarization control systems for coherent optics,” IEEE J. Lightwave Techn. Vol. 6, No.7, 1199-1207 (1988)). The polarization controller is located in a closed feedback loop at the input to the fiber amplifier, which does not maintain a fixed state of polarization, and under the influence of control voltages generated by a special optimizing processor, based on a metric, which is the signal of the photodetector, which registers the light power in the required direction of linear polarization at the output from the amplifier, changes the state of polarization of the light beam at the input to the amplifier so as to obtain the maximum possible value of this power and, thus, achieve the required state of polarization at the output of the amplifier. To control polarization controllers, appropriate optimization algorithms have been developed, described in detail in the literature and well known to specialists in this field (L. Xi, X. Zyang, F. Tian, X. Tang, X. Weng, G. Zhang, X. Li, and Q.Xiong, “Optimizing the operation of LiNbO ₃ -based multistage polarization controllers through an adaptive algorithm” IEEE Photonics J., vol.2, No.2, 195-202 (2010); R.Su, Y.Liu, B .Yang, P.Ma, X.Wang, P.Zhou, and X.Xu, “Active polarization control of a 1.43 kW narrow linewidth fiber amplifier based on SPGD algorithm,” J.Opt., vol.19, 045802 (2017 )).

In order to obtain the required distribution of the slopes of the linear polarization of the subbeams r _j in the grating, in the second feedback channel we install sequentially an N -channel polarization rotator, which rotates the direction of the linear polarization of each subbeam to its specified angle relative to the OX axis, a linear polarizer with an axis in the OX direction and a matrix of N focusing lenses, with the help of which each sub-beam is assembled in the plane of its small aperture, which limits the field of view of the corresponding photodetector.

The operating principle of the polarization rotator is known and is as follows:

Let us consider an optical system that receives radiation from an array of N collimated parallel beams of linearly polarized radiation, each of which has an individual direction of linear polarization. We need to rotate the polarization direction of each beam so that at the exit from the system all beams have a given direction of linear polarization. There is a known scheme of an optical rotator (Chun Ye “Construction of an optical rotator using quarter-wave plates and an optical retarder”, Opt.Engineering, 34(10), 3031-3035 (1995)), consisting of two crossed quarter-wave plates, between which an optical retarder or modulator with a variable retardation coefficient is located. The specified modulator has adjustable birefringence and its retardation is Δ φ = (2π /λ )( n ₀ - n _e ) d , where n ₀ is the refractive index of an ordinary wave, n _e is the refractive index of an extraordinary wave, d is the thickness of the retarder, λ - wavelength. If the fast axis of the first quarter-wave plate is directed along the x- axis, the axis of the second quarter-wave plate is perpendicular, i.e. along the y axis, and the modulator axis is at an angle of 45° to it and the direction of linear polarization of the incident beam coincides with the direction of the fast axis of the first plate (see Fig. 6 (a, b)), then the Jones matrix of such a system can be represented as :

Where - Jones matrix of a quarter-wave plate, the fast axis of which is perpendicular to the x- axis, - Jones matrix of a quarter-wave plate, the fast axis of which is directed along the x axis and the middle matrix describes the procedure for rotating the main axis of the polarization ellipse by an angle ψ = Δ φ /2.

For a linearly polarized beam, the Jones vector can be written as follows

where α is the angle of inclination of linear polarization relative to the OX axis.

If the direction of linear polarization of the incident beam makes an angle α with the direction of the fast axis of the first quarter-wave plate, then the Jones vector of the beam at the exit from system (10) corresponds to the following expression:

Expression (13) describes the procedure for rotating the direction of linear polarization of the incident beam by the angle Δ α = Δ ϕ /2.

In the work (Chun Ye “Construction of an optical rotator using quarter-wave plates and an optical retarder”, Opt.Engineering, 34(10), 3031-3035 (1995),) a commercial single-channel liquid crystal rotator with retarder was used as a retarder , controlled by the magnitude of the applied voltage from 0 to 10 volts (2 kHz) (Meadowlark Optics, www.meadowlark.com). If it is necessary to control multiple radiation channels in parallel, it is necessary to arrange an array of such rotators (see Fig. 5b), carefully calibrate the rotators, optically match the array of rotators with the array of beams, so that each beam has its own rotator, and synchronize the action of these rotators in time.

A liquid crystal spatial light modulator operating in transmission (see Fig. 5a) containing parallel aligned nematic crystals (Jeffrey A. Davis, Dylan E. McNamara, Don M. Cottrell, and Tomio Sonehara “Two-”) can also be used as a rotator. dimensional polarization encoding with a phase-only liquid-crystal spatial light modulator,” Appl. Opt. vol. 39 (10), 1549 - 1554 (2000)). Such spatial light modulators are available on the market and are manufactured by a number of companies (for example, Meadowlark Optics, www.meadowlark.com, HOLOEYE Photonics Ag, www.holoeye.com, etc.). Unlike single-channel rotators, spatial light modulators consist of multiple independently controlled birefringent liquid crystal cells, which allows simultaneous control of the direction of linear polarization of multiple light beams. The response time of rotators and spatial light modulators is limited by the electro-optical response of nematic crystalline materials and averages 10 ms, which is sufficient for a wide range of applications. When a multichannel polarization rotator is present in the feedback circuit, each j -th subbeam, in accordance with (14), rotates its current direction of linear polarization in the plane of the photodetector at a given angle ψ _j .

The maximum value of the radiation power recorded by the photodetector on the axis in the plane of the j -th subbeam is observed when the direction of polarization of the subbeam coincides with the axis of the polarizer. The fulfillment of this condition is achieved as a result of the operation of one of the stochastic optimization algorithms well known to specialists in this field and underlying the N -channel polarization optimizing processor that controls the polarization controllers. In particular, the stochastic parallel gradient descent algorithm SPGD can be used (MA Vorontsov, GW Carhart, M. Cohen, and G. Cauwenberghs, “Adaptive optics based on analog parallel stochastic optimization: analysis and experimental demonstration,” J. Opt. Soc. Am A 17(8), 1440-1453 (2000)). The principle of operation of electrically controlled polarization controllers, based on changes in birefringence in an optical system equivalent to a set of wave plates, is also well known to specialists. These devices are presented on the market, in particular, the products of the companies EoSpace (www.eospace.com), General Photonics (www.generalphotonics.com), etc. can be noted. As a result of the operation of the feedback circuit based on the stochastic optimization algorithm, current random changes in the state of optical polarization in the channels are compensated by the operation of polarization controllers, which maintain the polarization direction coinciding with the polarizer axis.

Thus, a controlled change in the specified retardations in a multichannel polarization rotator leads to a controlled change in the polarization state of the beam synthesized in the far optical field.

In addition to the formation of synthesized beams with the required polarization distribution, the proposed method allows you to rotate the required polarization distribution around the beam propagation axis at a given angle. To do this, in each channel of the optical rotator it is necessary to add a retardation of the same value Δψ _j = Δψ _s to the retardations corresponding to the given polarization distribution, then the entire picture of the polarization distribution, determined by the value β _j in each channel, will rotate as a whole by an angle Δψ _j . The sign of the addition Δψ _j will determine the direction of rotation. Separately, it should be noted that it is possible to control the direction angle of uniform linear polarization of a synthesized beam formed by an array of beams with unidirectional polarization. To do this, in each channel of the optical rotator it is necessary to introduce retardations equal to ψ _j = ψ _s , then, according to condition (10), the phase difference between the subbeams will become equal to zero, and the angle of inclination of the linear polarization of each subbeam will become equal to ψ, which corresponds to the coherent addition of beams with unidirectional polarization, the direction angle of which is equal to ψ in the selected coordinate system.

As a result, the combined action of two feedback loops makes it possible to establish the required values of the phase differences between the subbeams and the required distribution of the directions of linear polarization of the subbeams, which, together with the required values of the amplitudes of the subbeams, leads to the formation in the far optical field of a synthesized beam with the required distribution of amplitude, wavefront and polarization in its cross section. In particular, this method allows the formation of beams that have a polarization singularity on the axis (the area where the polarization azimuth is not defined), the so-called cylindrical vector beams (see Figs. 10-14, 16,17). It should be noted that, based on Kotelnikov’s theorem, such a two-loop feedback system can operate effectively provided that the control time constant of one of the feedback loops is at least two times less than the control time constant of the second feedback loop, which is easily achieved by choosing corresponding sampling frequencies of control optimizing processors. In fig. 6 - fig. Figure 16 shows examples of intensity and polarization distributions of synthesized beams in the far optical field (on the right) and the corresponding distributions of linear polarization tilt angles (in parentheses) and phase shifts of subbeams (on the left). The intensity distribution is shown in grayscale, and the polarization distribution is shown by line segments in linear polarization regions and elliptical curves in elliptical polarization regions with corresponding polarization azimuth slopes.

Device Fig. 5 includes a coherent source of linearly polarized radiation 1 with a fiber output and a fiber radiation divider into N channels 2 , each connected to one of N fiber-integrated optical phase modulators 3 , regulating the phase of the optical wave within ±2mπ (where m is the number more than 1). The radiation from the output of each phase modulator goes to the polarization controller 4 , which controls the state of the optical polarization of the corresponding beam. Next, each of the N channels is amplified by a corresponding fiber amplifier 5 containing an unpolarized active fiber and having at the output a lens collimator 7 that forms a collimated (parallel) beam of rays (subbeam). The power value of each amplifier 5 is set by the amplitude controller 6 . All N collimators are positioned in such a way as to form a synthesized aperture (composite beam) at the output, which is an array of closely spaced subapertures. As a result of diffraction, all subbeams overlap and interfere with each other on target 9 located in the far optical field, forming a synthesized beam characterized by random distributions of intensity and polarization state. The random nature of the intensity distribution arises as a result of uncontrolled fluctuations in the phases of subbeams, due to changes in the lengths of optical paths in the fiber under the influence of external factors. In turn, the nature of the distributed polarization state is associated with different states of optical polarization of subbeams that change over time due to the uncontrolled distribution of birefringence in the fiber, thermal fluctuations of birefringence associated with the generation of radiation, acoustic vibrations and mechanical stresses in the fiber. In order to synchronize the phases in the channels and stabilize the state of optical polarization, two feedback loops are used, for the operation of which a small fraction of the power of the entire composite (consisting of a set of subbeams) beam is separated from the whole composite beam by the first beam splitter 8 before the subbeams overlap. The 2-lens telescope 10 converts the dimensions of the composite beam to match other parts of the device. The radiation reflected from the beam splitter plate 8 is used to form the first feedback channel, ensuring phase synchronization of the subbeams, and establishing the required phase difference between the subbeams. To do this, the radiation reflected from the beam splitter 8 is supplied to the N -channel phase-forming element 12 (SLM or DMD), which introduces independent adjustable phase shifts into each subbeam of the grating. After acquiring the specified phase shifts, the radiation of all subbeams is focused using a lens 13 into the plane of a small aperture 14 , where all the beams overlap and interfere with each other, forming interference maxima and minima. Photodetector 15 , installed behind a small aperture 14 , registers the radiation power at the interference maximum and generates a feedback signal, which is processed by an N -channel optimizing processor 16 , operating in accordance with the stochastic optimization algorithm and generating control voltages for phase modulators 3 , which introduce the appropriate shifts phases into each subbeam, so that the radiation power recorded by the photodetector reaches and maintains the maximum possible value. This ensures phase synchronization and maintenance of the required phase difference between subbeams. The radiation passed through the beam splitter plate 11 is used to form a second feedback channel, which ensures the formation of the required directions of linear polarization of the subbeams. To do this, the radiation is supplied to a controlled N -channel polarization rotator 17 , which in turn consists of two quarter-wave plates 18 and 20 with an orthogonal arrangement of the main axes and a phase modulator 19 located between them, the main axis of which is 45 degrees with the main axes of the quarter-wave plates 18 and 20 . The N -channel polarization rotator independently rotates the direction of linear polarization of each subbeam by an angle equal to the required one with the opposite sign. Then the radiation of all subbeams passes through a linear polarizer 21 and an array of focusing lenses 22 , in which each subbeam has its own focusing lens. In the focal plane of the lens array 22 there is a corresponding array of optical fibers 23 , optically coupled with an array of photodetectors 24 , so that each subbeam corresponds to its own photodetector, which records the power of the polarization component of the subbeam radiation passed through the linear polarizer 21 . The signal from the photodetector 24 is fed to the corresponding channel of the N -channel optimizing polarization processor 25 , operating on the basis of a stochastic optimization algorithm and generating voltage to control the corresponding polarization controller 4 so that the value of the recorded radiation power reaches the maximum possible value. As a result, the direction of linear polarization of the subbeam incident on the photodetector 24 will coincide with the direction of the axis of the linear polarizer 21 , and the direction of linear polarization of the corresponding subbeam in the plane of the synthesized aperture and in the far optical field will coincide with the required direction.

The required values of the relative amplitudes of the subbeams, the phase differences between the subbeams, and the directions of linear polarization of the subbeams are set programmatically using a control computer 26 . As a result of the combined action of controlled amplitudes of subbeams, phase differences between them and the corresponding distribution of linear polarization directions due to diffraction overlap and interference of subbeams in the far optical field, a synthesized beam is formed with the required distribution of amplitude, wavefront and polarization in its cross section.

In fig. Figure 1 shows examples of synthesized apertures consisting of N = 6, N = 7, N = 19 and N = 37 subapertures.

In fig. Figure 2 shows examples of possible distributions of linear polarization directions for the number of subbeams N = 6: a) unidirectional linear polarization; b) unidirectional linear polarization, rotated by 45°; c) radial distribution; d) azimuthal distribution; e) radial distribution with a rotation of directions by 20°; f) azimuthal distribution with a rotation of directions by 20°; and)

hybrid radial-azimuth distribution; h) hybrid azimuthal-radial distribution.

In fig. Figure 3 shows a graphical representation of the field of the synthesized aperture using the following notation: a - radius of the Gaussian beam (subbeam); R is the radius of the circle on which the centers of the subbeams are located; θ _j = 2 π( j - 1 ) /N is the angle between the centers of the subbeams relative to the center of the synthesized aperture; N is the number of subbeams; j is the subbundle number; r _j = ( x ₀ cos( α _j ) + y ₀ sin( α _j )) - unit vector of linear polarization of the j - subbeam; α _j is the angle of inclination of the polarization vector of the j -subbeam relative to the OX axis; s _j = R ( x ₀ cos( θ _j ) + y ₀ sin( θ _j )) - vector of the position of the subbeam center on the plane; x ₀ , y ₀ - basis vectors of the coordinate system; ρ ₌ ρ _j - s _j is the coordinate of an arbitrary point in the plane of the synthesized aperture.

In fig. Figure 4 shows the intensity distribution in the plane of a small aperture ( 14 ) (a) and its cross section (b) for N = 6 with a uniform (azimuthal, radial and azimuthal and radial with additional tilt) distribution of linear polarization directions when using DMD as a phase-forming element (top row) and SLM (bottom row).

In fig. Figure 5 shows a diagram of the device.

In the device shown in FIG. 5 a) a system consisting of a first quarter-wave plate 10 , a phase transmission spatial light modulator 11 and a second quarter-wave plate 12 is used as a polarization rotator 9 . The fast axes of the quarter-wave plates are orthogonal to each other, and the direction of orientation of the liquid crystals in the spatial light modulator makes an angle of 45° with the axes of the quarter-wave plates.

In the device shown in FIG. 5 b) a system consisting of a first quarter-wave plate 10 , an array of single-channel rotators 11 and a second quarter-wave plate 12 is used as a polarization rotator 9 . The fast axes of the quarter-wave plates are orthogonal to each other, and the axes of the rotators make an angle of 45° with the axes of the quarter-wave plates.

In fig. 6 - fig. Figure 17 shows examples of intensity and polarization distributions of synthesized beams in the far optical field (on the right) and the corresponding distributions of linear polarization tilt angles (in parentheses) and phase shifts of subbeams (on the left). The intensity distribution is shown in grayscale, and the polarization distribution is shown by line segments in linear polarization regions and elliptical curves in elliptical polarization regions with corresponding polarization azimuth slopes.

Fig. 6. Distribution of intensity and polarization in the case of unidirectional linear polarization of subbeams.

Fig. 7. Distribution of intensity and polarization in the case of unidirectional linear polarization of subbeams with an additional slope π/4 and phase shift φ ^T _j = [0, 0, 0, 0, 0, 0].

Fig. 8. Distribution of intensity and polarization in the case of unidirectional linear polarization of subbeams with a phase shift φ ^T _j = [0, π/3, 2π/3, π, 4π/3, 5π/3].

Fig. 9. Distribution of intensity and polarization in the case of unidirectional linear polarization of subbeams with an additional slope π/4 and phase shift φ ^T _j = [0, π/3, 2π/3, π, 4π/3, 5π/3].

Fig. 10. Distribution of intensity and polarization in the case of a radial distribution of slopes of linear polarization of subbeams (the magnitude of the slope is indicated in parentheses) and phase shift φ ^T _j = [0, π/3, 2π/3, π, 4π/3, 5π/3].

Fig. 11. Distribution of intensity and polarization in the case of a radial distribution of slopes of linear polarization of subbeams with an additional slope (-π/4) (the magnitude of the slope is indicated in parentheses) and a phase shift φ ^T _j =[0, π/3, 2π/3, π, 4π/3, 5π/3].

Fig. 12. Distribution of intensity and polarization in the case of azimuthal distribution of slopes of linear polarization of subbeams (the magnitude of the slope is indicated in parentheses) and phase shift φ ^T _j = [0, π/3, 2π/3, π, 4π/3, 5π/3].

Fig. 13. Distribution of intensity and polarization in the case of an azimuthal distribution of slopes of linear polarization of subbeams with an additional slope (+π/4) (the magnitude of the slope is indicated in parentheses) and a phase shift φ ^T _j = [0, π/3, 2π/3, π, 4π/3, 5π/3].

Fig. 14. Distribution of intensity and polarization in the case of an azimuthal distribution of slopes of linear polarization of subbeams with an additional slope (-π/4) (the magnitude of the slope is indicated in parentheses) and a phase shift φ ^T _j = [0, π/3, 2π/3, π, 4π/3, 5π/3].

Fig. 15. Distribution of intensity and polarization in the case of a hybrid radial-azimuthal distribution of the slopes of the linear polarization of subbeams (the magnitude of the slope is indicated in parentheses) and the phase shift φ ^T _j = [0, 0, 0, 0, 0, 0].

Fig. 16. Distribution of intensity and polarization in the case of a hybrid radial-azimuthal distribution of the slopes of the linear polarization of subbeams (the magnitude of the slope is indicated in parentheses) and the phase shift φ ^T _j =[0, π/3, 2π/3, π, 4π/3, 5π/ 3].

Fig. 17. Distribution of intensity and polarization in the case of a hybrid azimuth-radial distribution of the slopes of the linear polarization of subbeams (the magnitude of the slope is indicated in parentheses) and the phase shift ϕ ^T _j = [0, π/3, 2π/3, π, 4π/3, 5π/ 3].

Claims (5)

1. The method of amplitude, phase and polarization control in a phased array of fiber amplifiers and control of the distributed state of intensity, wavefront and polarization of a synthesized beam in the far optical field consists in obtaining a phased array of N collimated Gaussian beams by controlling N phase-shifting elements, each of which controls the phase the corresponding beam in the array and is included in a feedback circuit operating in accordance with the stochastic optimization algorithm and using a small fraction of the radiation power, while each beam is given a relative amplitude value calculated in advance in accordance with the required field distribution by controlling N fiber power amplifiers, and the value of the phase difference between the beams, calculated in advance in accordance with the required field distribution, which is controlled using a single phase-forming element included in this feedback circuit and arranged in such a way that each individual beam in the array corresponds to a separate region of the phase-forming element, which sets the phase shift by a value that provides the required phase difference between the beams, characterized in that to control the direction of linear polarization of each beam, a polarization controller is placed in each channel between the phase-shifting element and the fiber amplifier, which controls the polarization state of the corresponding beam; the fiber amplifiers contain an active fiber that does not support the polarization state , and the part of the radiation that is used to generate feedback is, in turn, divided in half, forming a second feedback circuit that operates in accordance with the stochastic optimization algorithm and brings the random polarization of the beams at the output of the fiber amplifiers to a state of linear polarization, with each beam the linear polarization tilt direction, calculated in advance in accordance with the required polarization distribution, is set, which is controlled using an N -channel polarization rotator included in this feedback circuit and arranged so that each individual beam in the array corresponds to a separate channel of the polarization rotator, which rotates the direction linear polarization of each beam, independently of each other, at a given angle, forming a beam lattice with the required individual direction of linear polarization, while in the far optical field, as a result of diffraction overlap and interference of the beams, a synthesized beam is formed. 2. The method according to claim 1, characterized in that to form unidirectional linear polarization of the synthesized beam, rotated by a given angle in a given coordinate system, each channel of the N -channel polarization rotator rotates the direction of linear polarization of the corresponding beam by the same given angle . 3. The device for controlling the distributed state of intensity, wavefront and polarization of the synthesized beam in the far optical field includes a coherent source of linearly polarized radiation, a radiation divider into N channels of equal power, each of which is connected in series with an optical phase-shifting element that regulates the phase of the optical wave, and then with a fiber power amplifier that generates power of the required amplitude, controlled by an N -channel amplitude controller, and has a collimator at the output, all N channels are located at lattice nodes, adjusted so that the optical axes of the output beams are parallel to each other, and form a synthesized beam, small a fraction of the radiation of which is deflected using a beam splitter to provide a feedback signal, which contains a phase-forming element, after passing through or reflecting from which each beam acquires an additional phase shift equal to a given one, and a focusing lens that collects all beams in the focal plane where it is located a photodetector, the field of view of which is limited by a small aperture, and the signal of which is supplied to an N -channel optimizing processor operating in accordance with a stochastic optimization algorithm and controlling the phase-shifting elements so that the phase difference between the individual beams of the grating corresponds to the required value, characterized in that between the optical a phase-shifting element and a power amplifier, each channel contains a polarization controller that regulates the state of optical polarization of the radiation at the output of the power amplifier; fiber amplifiers contain an active fiber that does not maintain the polarization state, and part of the radiation, which is used to generate feedback, is in turn divided in half with using a second beam splitter, forming a second feedback channel in which an N -channel controlled polarization rotator is located, setting the polarization direction of each subbeam so that each subbeam, independently of other subbeams, rotates the direction of its linear polarization by an angle equal to a given deviation from the reference direction, with the opposite sign, a linear polarizer placed after the N -channel polarization rotator and setting the reference direction of linear polarization, a matrix of focusing lenses, in which each beam corresponds to its own focusing lens, in the focal plane of which the receiving ends of optical fibers are located, optically coupled with the receiving the platform of the corresponding photodetectors forming a matrix of photodetectors, the signals of which are supplied to the N -channel optimizing polarization processor, operating in accordance with the stochastic optimization algorithm and controlling the polarization controllers so that the radiation at the output of the power amplifier in each channel acquires the required direction of linear polarization. 4. The device according to claim 3, characterized in that as an N -channel polarization rotator, a system is used consisting of two quarter-wave plates, the main axes of which are orthogonal to each other, and a phase spatial light modulator located between them, operating on transmission, oriented under an angle of 45 degrees relative to the main axes of the quarter-wave plates, and the action of which is organized in such a way that each subbeam in the array corresponds to its own area of pixels on the display of the spatial light modulator. 5. The device according to claim 3, characterized in that the N-channel polarization rotator is a system consisting of two quarter-wave plates, the main axes of which are orthogonal to each other, and a sectioned or stacked liquid crystal retarder (retarder) located between them.