Detection and telemetry via electromagnetic radiation pulses

Abstract

A method and system (10) for detection and telemetry using electromagnetic radiation pulses, allowing the separation distance (d e ) is a function that characterizes the radial velocity distribution (v r The measurement signal (SM) collected for each acquisition sequence performed for effective measurement is decomposed using the impulse response from the system. The result of the decomposition includes an estimate of the radial velocity distribution as a function of the separation distance.

Description

Technical Field

[0001] The present description relates to a method and system for detecting and telemetry by electromagnetic radiation pulses, in particular for wind force determination measurements. Background Art

[0002] Systems for detection and telemetry using electromagnetic radiation pulses are used in many applications, particularly for characterizing fluid flows. Radiation pulses are emitted toward a survey area, which may contain one or more targets to be characterized, and a portion of each pulse is backscattered or reflected back from each target and detected upon return by the system. In the case of fluid flows to be characterized, the targets consist of backscattered particles suspended in the fluid and driven to move according to the velocity of the fluid at each location in the survey area. For atmospheric airflow, the targets consist of aerosol droplets, dust particles, ice particles, and the like suspended in the air.

[0003] Such systems for detection and remote sensing using electromagnetic radiation pulses can be of the LIDAR (Light Detection and Ranging) type or the RADAR (Radiowave Detection and Ranging) type. In the case of LIDAR systems, the emitted radiation is of the laser type, with the frequency of this radiation typically being between about 30 THz (terahertz) and 1200 THz, corresponding to a wavelength between about 250 nm (nanometers) and 10 μm (micrometers). In the case of RADAR systems, the radio wave frequencies used are typically between 3 MHz (megahertz) and 300 GHz (gigahertz). High-frequency RADAR systems typically correspond to radiation wavelengths between 1 mm (millimeter) and 7.5 mm.

[0004] It is well known to use these systems to characterize a single target or several targets, each with a separation distance value and a radial velocity value, when the number of targets simultaneously contributing to the measurement signal is small and / or the targets' respective separation distance and radial velocity values ​​are well separated from one another. However, in the case of characterizing fluid flows, the backscattered targets are continuously distributed throughout the survey area involved in each measurement and have continuously varying radial velocity values. Determining the radial velocity distribution as a function of separation distance is therefore much more difficult.

[0005] A known method for determining the radial velocity distribution as a function of the separation distance using an electromagnetic radiation pulse detection and telemetry system consists in calculating the intensity of the detected radiation for each detection instant and for each emitted radiation pulse. To this end, the weight of the portion of the content from the survey area corresponding to the same detection time value of the backscattered radiation is increased after the radiation has traveled back and forth between the system and each portion of the content. This calculation method uses a "stacked model," in which the survey area is cut into successive spatial slices depending on the separation distance, and a single radial velocity value is associated with each spatial slice. The backscattered radiation intensity is thus calculated for several possible radial velocity distributions as a function of the separation distance, and then the one of these distributions that best corresponds to the actually detected backscattered radiation is accepted as the measurement result. However, this measurement method has the following disadvantages:

[0006] - the proportion of backscattered radiation detected simultaneously that corresponds to the same transmitted pulse but to parts of this pulse emitted at different times and backscattered from different separation distances in the survey area. For this reason, the spatial resolution achieved by this method is approximately C·τ / 2, where C is the propagation speed of radiation external to the detection and telemetry system and τ is the duration of each pulse. In other words, a spatial resolution finer than C·τ / 2 is not possible with the stack model method; and

[0007] Since the radial velocity of the target is inferred from the frequency of the detected backscattered radiation, it can only be estimated based on an analysis of this backscattered radiation within a time window. Since this time window is usually chosen to have a duration equal to the duration of the emitted laser radiation pulse, the spatial resolution is approximately C·τ; and

[0008] To obtain a spatial resolution better than C·τ, it is necessary to consider the weighting of all measurements associated with different detection times from all spatial slices, rather than the weighting of a single spatial slice onto each measurement.

[0009] Technical issues

[0010] In view of this situation, an object of the present invention is to improve the efficiency of measurements by using electromagnetic radiation pulses for detection and remote sensing. More precisely, an object of the present invention is to improve the spatial resolution and the accuracy of radial velocity measurements obtained for a continuous medium to be characterized, for example for a fluid flow to be characterized, in particular in a part of the atmosphere.

[0011] Another object of the present invention is to take into account the operation of a detection and telemetry system using electromagnetic radiation pulses in order to reduce the amount of calculations necessary to infer measurements characterizing targets contained in the survey area from measurement signals collected during an acquisition sequence. Summary of the Invention

[0012] To achieve at least one or another of these objectives, a first aspect of the present invention proposes a new detection and telemetry method that uses electromagnetic radiation pulses to characterize radial velocity distribution as a function of separation distance within a survey area. This method of the invention comprises the following steps:

[0013] 1) obtaining a system for detection and telemetry using electromagnetic radiation pulses, adapted to emit during an acquisition sequence at least one electromagnetic radiation pulse, to detect a portion of said at least one radiation pulse backscattered by at least one target present in the survey area, and to generate a measurement signal corresponding to this backscattered and detected portion of the radiation pulse, wherein this measurement signal contains information about the separation distance and the radial velocity of each target, the information about the radial velocity of each target corresponding to a frequency shift due to the Doppler effect occurring when radiation is backscattered by this target; and

[0014] 2) performing an acquisition sequence by controlling the system to transmit the at least one pulse into the survey region.

[0015] The method of the present invention further comprises the following additional steps:

[0016] 3) obtaining, in particular by simulation or measurement, a characterization of an impulse response of the system, wherein this impulse response corresponds to a measurement signal generated by said system during an acquisition sequence and when a single backscatter element is located in the survey region (this single backscatter element corresponding to a single separation distance value and having a known radial velocity value relative to the system, which may be zero) and when the survey region has no backscatter elements other than the so-called single backscatter element; and

[0017] 4) By considering the measurement signal generated in step 2) as the sum of weights corresponding to impulse responses applied to separation distance values ​​and radial velocity values, and multiplying by the backscatter amplitude value, the measurement signal is decomposed into several weights to obtain measurement results in the form of pairs, each pair consisting of a backscatter amplitude value and a radial velocity value, and which are respectively assigned to several separation distance values ​​within the survey area.

[0018] In other words, the results of the method of the present invention include a backscatter amplitude distribution and a radial velocity distribution, wherein both distributions are functions of the separation distance between the system being used and the interior of the survey area. The backscatter amplitude distribution characterizes the concentration or type of targets present at variable locations in the survey area, and the radial velocity distribution characterizes the displacement of these targets at the same location, wherein this displacement can vary between different locations within the survey area. When the medium to be characterized is a flowing fluid, the radial velocity distribution constitutes a representation of the radial velocity field of the flow.

[0019] Unlike prior art methods that sum the contributions of radiation backscattered from spatial slices of the survey region and subsequently detected at a specific time, the method of the present invention is based on summing the contributions of spatial slices of the survey region to the measurement signals collected by the system used during an acquisition sequence. In this way, the method of the present invention simultaneously considers the contributions of the resolved spatial slices of the survey region and the operational effects of the system used. Specifically, it considers the shape of the electromagnetic radiation pulse emitted in each acquisition sequence, as well as the detection mode and signal processing applied within the system's detection path. In other words, the method of the present invention simultaneously considers the characteristics of the radiation emitted by the system used, the effects of radiation propagation in both directions between the system and the survey region and within the survey region, the locally effective backscattered power in the survey region, and the characteristics of the signal detection and processing within the detection path of the system used. For this reason, the method of the present invention is particularly effective in reducing the amount of computation required. This efficiency is achieved by using an impulse response that incorporates the characteristics of the radiation emitted by the system used, as well as the signal detection and processing characteristics of that system. The measurement signal is then interpreted according to the invention as the result of the convolution of this impulse response with the backscatter amplitude distribution, which is a function of the separation distance values ​​and radial velocity values ​​corresponding to the contents of the survey area. Step 4) involves performing an operation by a calculation that is the inverse of this convolution and is sometimes called deconvolution.

[0020] In various embodiments of the invention, the impulse response may be a function of both the separation distance or the round trip time of radiation from the optical exit of the system, and also a function of one of:

[0021] - the frequencies of the spectral components of the portion of the radiation pulse that is backscattered and then detected by the system;

[0022] - the frequency shift between the spectral components of the portion of the radiation pulse that is backscattered and then detected by the system and the radiation of each pulse emitted by the system, or the radial velocity value associated with this frequency shift according to the Doppler effect; and

[0023] - The frequencies of the spectrum of the measurement signal used in step 3).

[0024] It is possible that the characterization of the impulse response can be obtained in step 3) by performing at least one acquisition sequence using a system for detection and telemetry using electromagnetic radiation pulses and using a single backscatter element positioned at a determined separation distance in the survey area. In this case, step 3) comprises an experimental calibration of the system used. Alternatively, the characterization of the impulse response can be obtained by performing a numerical simulation of the operation of the system for detection and telemetry using electromagnetic radiation pulses when a single backscatter element is present at a determined separation distance in the survey area. In both cases, the characterization of the impulse response obtained may be imperfect. Specifically, in the first case, in which an acquisition sequence is used to calibrate the system, the characterization of the impulse response obtained may be affected by detection noise, effects of sampling and digitization of the applicable values, etc. In the second case, in which the characterization of the impulse response of the system is a numerical simulation, secondary physical phenomena may not be taken into account.

[0025] In general, the method of the present invention is compatible with monostatic or bistatic systems that perform detection and telemetry via electromagnetic radiation pulses. In the context of the present invention, monostatic is understood to mean a system whose output and detection optical ports coincide, are adjacent, or are close to each other relative to the separation distance between the system and the survey area. On the other hand, bistatic is understood to mean a system whose output and detection optical ports are spatially separated.

[0026] In general, the system for performing detection and telemetry using electromagnetic radiation pulses, as implemented in the present invention, can be adapted to perform measurements according to any detection mode. Specifically, a detection mode can be used in which the frequency shift of the backscattered radiation due to the Doppler effect occurring during the reflection or backscattering of the transmitted pulses on the moving target is measured by causing the detected backscattered radiation to interfere with itself. This detection mode is referred to as "direct detection." However, the system can preferably be adapted to implement a heterodyne detection mode in which the detected backscattered radiation is mixed with a reference wave and the beat frequency resulting from this mixing is analyzed. In this case, the measurement signal generated at each acquisition sequence and used in step 4) is a heterodyne measurement signal.

[0027] Again generally, the system for performing detection and telemetry using electromagnetic radiation pulses for implementing the present invention may be of the LIDAR type. In this case, the pulsed radiation emitted by the system is laser radiation, and each radiation pulse may have one or more of the following characteristics:

[0028] -The wavelength of radiation within the pulse can be between 250nm and 10μm;

[0029] - the duration of each pulse may be between 50 ns (nanoseconds) and 1 μs (microseconds); and

[0030] - The pulses may have a half-peak frequency width of less than 1 GHz, preferably less than 20 MHz.

[0031] Alternatively, the system for implementing the present invention by detecting and telemetry using electromagnetic radiation pulses may be of the RADAR type. In this alternative, the pulsed radiation emitted by the system has a vacuum wavelength between 1 mm and 7.5 mm.

[0032] In a first preferred embodiment of the invention, the measurement results can be inferred from the measurement signals in step 4) by applying a two-dimensional decomposition algorithm using a posteriori maximum method, or a maximum likelihood method (which can be regularized, in which case also called penalized, or not regularized), or even a stochastic method such as a Markov chain Monte Carlo method. Possibly, the method can then include additional steps, which are performed on the basis of the measurement results provided by the two-dimensional decomposition algorithm, in order to reduce the width of the radial velocity distributions individually assigned to several separated distance values ​​in the survey area.

[0033] In a second preferred embodiment of the present invention, step 4) includes assigning a single radial velocity value and a single backscatter amplitude value to each separation distance value for the sampling of separation distance values ​​within the survey area. A reconstruction of the measurement signal is then calculated for all separation distance values ​​in the sampling as the sum of the weights of the impulse responses, each equal to the separation distance value and the radial velocity value assigned to that separation distance value, multiplied by the backscatter amplitude value also assigned to the same separation distance value. An iterative adjustment of the assigned radial velocity and backscatter amplitude values ​​is then performed to reduce the deviation between the measurement signal generated in step 2) and the reconstruction of the measurement signal generated by the values ​​assigned to the radial velocity and backscatter amplitude. A measurement result is then formed from the radial velocity and backscatter amplitude values ​​assigned to the separation distance value that provide the minimum deviation between the measurement signal and the reconstruction of the measurement signal.

[0034] Finally, the method according to the first aspect of the invention can be advantageously used in at least one of the following applications:

[0035] - Meteorological measurements, such as measurements of atmospheric turbulence;

[0036] -Measurement of the dispersion of atmospheric pollutants;

[0037] - measurement of the local concentration of backscattered particles suspended in the environment or chemical compounds that absorb and re-emit radiation from the pulse;

[0038] - For example, measurement of shear force of airflow at airports;

[0039] - measurement of the position and / or lifetime of at least one vortex present in the fluid flow;

[0040] - measurements of wind speed carried out from an aircraft in flight, in particular from a spacecraft or drone;

[0041] - measurements of the measured wind force performed to optimize the operation of the wind turbine; and

[0042] -Measurements of wind speeds performed to adjust aircraft flight formations or to adjust the flight of drones.

[0043] Furthermore, a second aspect of the present invention provides a system for detection and telemetry using electromagnetic radiation pulses, comprising:

[0044] - a transmission path adapted to transmit at least one electromagnetic radiation pulse into the survey area when performing an acquisition sequence;

[0045] a detection path adapted to detect, during an acquisition sequence, a portion of at least one emitted radiation pulse after said portion has been backscattered by at least one target present in the survey area, wherein this detection path is further adapted to generate a measurement signal containing information about the separation distance and the radial velocity of each target present in the survey area, the information about the radial velocity of each target corresponding to a frequency shift due to the Doppler effect occurring when radiation is backscattered by this target;

[0046] - a controller arranged to activate the transmit and detect paths according to an acquisition sequence;

[0047] - means for storing a representation of an impulse response of the system, wherein this impulse response corresponds to a measurement signal generated by said system during an acquisition sequence and when a single backscatter element is located in the survey region, this single backscatter element corresponding to a single separation distance value and having a radial velocity value relative to the system, which may be zero, and when the survey region has no backscatter elements other than the so-called single backscatter element; and

[0048] - calculation means suitable for decomposing the measurement signal into the sum of proportions corresponding to impulse responses applied to separation distance values ​​and radial velocity values, and multiplying by a backscatter amplitude value, so as to output measurement results in the form of pairs, each pair consisting of a backscatter amplitude value and a radial velocity value, respectively assigned to several separation distance values ​​within the survey area.

[0049] This system is particularly suitable for carrying out the method according to the first aspect of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The characteristics and advantages of the present invention will become more apparent in the following detailed description of non-limiting embodiments, with reference to the accompanying drawings, in which:

[0051] Figure 1 is a weather map of a system for detection and telemetry using electromagnetic radiation pulses to which the present invention is applicable;

[0052] Figure 2 Can be Figure 1 A timing diagram of radiation pulses emitted by a system for implementing the present invention;

[0053] Figure 3 It is a three-dimensional diagram, which is displayed according to Figure 1 The system is targeted based on Figure 2 the impulse response of the emitted radiation pulse; and

[0054] Figures 4a to 4c is a three-dimensional diagram illustrating the application of the present invention to the characterization of Von Karman type turbulence. DETAILED DESCRIPTION

[0055] For clarity of the following description, components that are not directly involved in the present invention or that can be spontaneously adapted by those skilled in the art to implement the present invention are not shown or described.

[0056] The present invention will now be described for example with respect to a LIDAR pulse system with heterodyne detection. However, the present invention can be readily transferred to detection modes other than heterodyne detection and / or to a RADAR system based on the following description. In general, the present invention relates to converting a measurement signal into a measurement result comprising a radial velocity distribution as a function of the separation distance. Equivalently, each radial velocity value in this representation can be replaced by a frequency shift value resulting from a target having a radial velocity value through the Doppler effect. The formula for expressing this equivalence, known to those skilled in the art, is Δf = -2·v r / λ, where Δf is the frequency shift, v r is the radial velocity of the target, and λ is the wavelength of the radiation. Similarly, each separation distance value can be replaced by the duration between the emission time of the radiation and the subsequent detection time of the backscattered portion of this radiation, by means of formula d e =C·(t d -t0) / 2, where t d is the detection time, and t0 is the emission time, d e is the separation distance. Furthermore, as an example, the LIDAR system described below is suitable for performing wind force measurements by using the backscattering of laser radiation generated by particles suspended in the air. These backscattered particles, which form the target to be characterized and can be aerosol droplets, dust or ice particles, are Figure 1 Designated by reference numeral 100 .

[0057] according to Figure 1Lidar system 10 includes a transmission path 10E and a detection path 10D. Transmission path 10E includes a laser source 1, which generates monochromatic radiation having a wavelength of 1545 nm when propagating in air; at least one acousto-optic modulator 2, labeled MAO and controlled to form continuous pulses from the laser radiation generated by source 1; and an optical amplifier 3, labeled AMPL. AOM 2 further produces a frequency shift in the laser radiation emitted by Lidar system 10 toward a target relative to the laser radiation generated by source 1. In this manner, when the radiation is back-reflected by a stationary target, the heterodyne detection signal generated in detection path 10D has a non-zero beat frequency. Detection path 10D includes a detector 7, labeled DETECT, and an acquisition card 8, labeled ACQ. Acquisition card 8 is connected to receive the analog electrical heterodyne detection signal, labeled SD, generated by detector 7. It is designed to apply processing to this detection signal SD and output a measurement signal, labeled SM, resulting from this processing. Specifically, acquisition card 8 performs sampling and digitization of the detection signal, accumulation of digitized signals for several identical radiation pulses emitted in succession, filtering, and the like. Possibly, an electrical signal amplifier (not shown) may be used in the transmission link of the detection signal SD between detector 7 and acquisition card 8. In heterodyne detection mode, detector 7 receives at its input a mixture of a portion of each pulse received by detection path 10D and a portion of the emitted laser radiation extracted from source 1 and AOM 2. As is known, the portion of the emitted laser radiation transmitted from emission path 10E to detector 7 may undergo various intermediate transformations between its extraction from emission path 10E and detector 7. AOM 2 and acquisition card 8 are commanded by controller 9, labeled CTRL, which provides the system with its LIDAR operations at each acquisition sequence.

[0058] When the LIDAR system 10 is of the monostatic type, the transmission path 10E and the detection path 10D can be coupled to the optical port 5 through the optical circulator 4, where this port serves as both an optical output port for the transmission path 10D and an optical receiving port for the detection path 10D. To this end, the transmission path 10E is optically coupled to the optical input port 41 of the optical circulator 4, the detection path 10D is optically coupled to the optical output port 42 of the optical circulator 4, and the optical port 5 of the LIDAR system 10 is optically coupled to the hybrid input-output optical port 43 of the optical circulator 4.

[0059] This LIDAR system 10 is known to those skilled in the art. It can be used to analyze the contents of a survey zone ZE located outside the system 10, in the path of the laser pulses, in front of the optical port 5. As is known, this survey zone ZE is the overlapping volume between the propagation trajectory of the laser pulses beyond the optical port 5 and the medium to be characterized, located outside the system 10. The distance d separating the proximal limit of the survey zone ZE from the optical port 5 is m The length l of the survey zone ZE may be defined by the time detection window applied by the detection path 10D (in particular, the acquisition card 8). e is also determined by the time detection window. Alternatively, the survey zone ZE can be determined by the convergence applied to the laser pulses to form the optical port 5, for example by using a converging lens. And, in the remaining part, each laser pulse has a pulse duration, which is denoted as τ. Under these conditions, the measurement signal SM contains the distance of separation of the targets 100 contained in the survey zone ZE according to each target relative to the optical port 5 (denoted as d e ) and the radial velocity of each target (denoted as v r ) characterization. In the remainder, the statistical separation distance d from the near side limit of the survey area ZE e . Furthermore, the radial velocity of one of the targets 100 is understood to mean the component of its velocity parallel to the direction of propagation of the laser pulse. It is positive when the target moves away from the system 10 and negative when the target moves towards the system 10. When the contents of the survey zone ZE are part of the atmosphere, the targets 100 consist of aerosol droplets, dust and / or ice particles suspended in the air and moved by, for example, air currents that may be present in the survey zone ZE. The radial velocity distribution v of these air currents is r The separation distance d within the survey zone ZE may be characterized based on measurements performed using the LIDAR system 10 as e This application of the LIDAR system 10 for determining wind force measurements is also known to those skilled in the art. The radiation pulse designated by the reference P is Figure 1 It is symbolically shown in as propagating between the optical port 5 of the system 10 and the investigation zone ZE.

[0060] The invention, which is the subject of this description, concerns the transformation of the measurement signal SM delivered by the acquisition card 8 into a radial velocity distribution v in the survey zone ZE. r The associated measurement result RM. This transformation can be performed by a calculation unit 11, denoted CPU. Typically, the measurement result RM comprises a backscatter amplitude distribution, which is a function of the separation distance d e and radial velocity v r A function of both, or equivalently the separation distance d e and the Doppler effect frequency shift Δf. The backscatter amplitude distribution of the measurement result RM is expressed as S(d e,v r ) or S(d e ,Δf). It constitutes the actual backscatter amplitude distribution S0(d e ,v r ) or S0(d e Reference 12 designates storage means referenced STOR, which are accessible by the computation unit 11. The structure of the storage means 12 may advantageously be chosen to facilitate convolution calculations using the values ​​stored in these means 12.

[0061] Figure 2 An example of a radiation pulse envelope corresponding to the shape of each pulse P generated, for example, by transmit path 10D of system 10 and effective downstream of optical port 5 is shown. The abscissa refers to time expressed in nanoseconds (ns), labeled t, and the ordinate refers to the prompt radiation power, labeled PW, normalized relative to its maximum value. The half-width of the pulse shown is approximately 75 ns, corresponding to a spatial length of the pulse of approximately 22 m when propagating in air outside system 10.

[0062] Figure 3 Display Target Figure 2 The pulse shape corresponds to Figure 1 Example of an impulse response of a LIDAR system. This pulse shape is obtained when a retroreflective target, such as a mirror, is arranged in the survey zone ZE.

[0063] By definition, the impulse response corresponds to the following condition:

[0064] - a retroreflective target such as a mirror is located in the survey zone ZE corresponding to the separation distance d e At a single position of a single value of ;

[0065] - The back reflection of the target corresponds to its radial velocity v r Preferably, but not necessarily, this radial velocity value of the retroreflective target may be zero;

[0066] - the retroreflector target is the only one inside the survey zone ZE; and

[0067] - The pulse shapes and preferably other or even all operating parameters of the system 10 are identical between its operating parameters used to obtain the pulse responses and the determination of these pulse shapes and, for example, the operating parameters used for the effective acquisition sequence (which in the present case is used to perform wind-determining measurements).

[0068] The impulse response can be obtained from the measurement signal SM by confining it to time windows that are gradually shifted in time, and by applying an apodization function, such as Gaussian shaping, to each confinement of the measurement signal SM in the corresponding time window. As is known, the goal of the apodization function is to eliminate or attenuate discontinuities present in the analyzed signal in order to eliminate or reduce artifacts that may be caused by these discontinuities. In the present case, the end of the time window may result in side lobes in the subsequent Fourier transform result. The apodization function eliminates these lobes. Each portion of the measurement signal SM confined to the time window and apodized is then processed by a time-Fourier transform, and the result of this Fourier transform is then squared to provide a constant separation distance value d. e Down Figure 3 The reference separation distance d e The origin of the horizontal axis of the values ​​of can be assigned to one of those values ​​corresponding to the maximum value of the impulse response. In this way, the horizontal axis reference offset value, Δd e , which is defined relative to this origin of the impulse response. In the following, each corresponds to Δd e The segments of the representation of the impulse response of constant value are called the spectrum of the impulse response, and the set of these spectra forms the spectrogram of this impulse response. Each spectrum is therefore a function of the frequency shift Δf and is assigned to the separation distance d according to the time of flight between the emission time of the pulse and the middle time of the time window applied to the measurement signal SM for this spectrum. e The value of and thus assigned to the offset Δd e value.

[0069] The impulse response of system 10 can be obtained experimentally, for example by actually placing a retroreflective mirror at an identified position in the survey zone ZE and by providing this retroreflective mirror with a radial velocity of motion that is also identified. Preferably, for this experimental determination of the impulse response, the retroreflective mirror can be fixed relative to system 10. An acquisition sequence is then performed by system 10, during which calculation unit 11 applies the operations of division and transformation just described to the measurement signal SM as output by acquisition card 8.

[0070] Alternatively, the impulse response may be obtained by numerical simulation of the operation of the system 10 during the acquisition sequence by simulating the effect of an emitted radiation pulse on a retroreflective mirror placed in the survey zone ZE.

[0071] Typically, obtaining the impulse response enables calibration of the system 10 in order to subsequently perform valid measurements using this system 10. The impulse response determined for the system 10 is stored in the storage means 12.

[0072] exist Figure 3In the diagram of , the horizontal axis may therefore be offset by reference to a separation distance Δd relative to the position of the retroreflective target used for calibration. e . It is expressed in meters. The vertical axis refers to the frequency shift Δf relative to the frequency of the radiation in the emission path 10E, which is expressed in megahertz. Finally, Figure 3 The third axis of the diagram refers to the amplitude of the impulse response, which can be expressed in arbitrary units (au). It corresponds to the horizontal lines with the indicated values ​​of 0.1, 0.3, 0.6 and 1.0. The impulse response shows that, due to the shape of each pulse and the signal processing applied in the detection path 10D, a single retroreflected target appears in the measurement signal SM as a group of diffuse retroreflected targets that move according to a radial velocity distribution scattered around the actual radial velocity value of the single target. This dispersion of the impulse response in terms of separation distance and radial velocity is specifically due to Figure 2 The finite width of the pulses is due to the detection time window applied by the acquisition card 8. Furthermore, the radial velocity distribution depends on the offset Δd relative to the actual separation distance of the single targets used to calibrate the system 10. e , in such a way that the impulse response is two variables Δd e and v r or equivalently Δd e and Δf. Hereinafter, h(Δd e ,Δf) represents this impulse response. In addition, for Figure 3 In the example shown in e The tilt stretch affecting the pulse response in the Δf axis reveals a drift of about 3 MHz in the frequency of each transmitted pulse between the start and end of this pulse. This transmit frequency drift is therefore automatically included in the process for obtaining the measurement result RM of the valid acquisition sequence, so that this drift does not introduce errors into this result.

[0073] When a so-called valid acquisition sequence is performed with the system 10 on a portion of the atmosphere containing the survey zone ZE, then the measurement signals SM output by the acquisition card 8 are considered to be generated by the atmosphere in the survey zone ZE at a separation distance d. e The incoherent sum of the portion of the laser pulse backscattered by the target distributed at various values ​​of e The backscatter amplitude and radial velocity value are associated with each of the separation distance values ​​between . The calculation unit 11 then applies a process substantially the same as that described above to the measurement signal SM output by the acquisition card 8 in order to experimentally obtain the impulse response. This process comprises the following steps:

[0074] - limiting the measurement signal SM to time windows that are gradually shifted in time;

[0075] - applying the apodization function to the portion of the measurement signal SM contained in each time window;

[0076] - applying a time-Fourier transform to the apodized portion of the measurement signal SM contained in each time window;

[0077] - Calculating the square of the result of the Fourier transformation for each time window and each frequency value f in order to obtain the spectrum of this time window.

[0078] The set of spectra then forms an effective acquisition sequence spectrogram characterizing the content of the survey zone ZE. Within this effective acquisition sequence spectrogram, each spectrum is assigned to a separation distance d e The value of , which corresponds to the time of flight between the emission time of the pulse and the middle time of the time window applied to the measurement signal SM for this spectrum. The spectrum is written as Sp(d e , Δf), by taking a reference frequency so as to define a frequency shift Δf between each spectral component and this reference frequency. In the case of heterodyne detection, this reference frequency is the frequency of the monochromatic radiation generated by the laser source 1. Thus, without taking into account possible measurement noise, the spectrum Sp(d e ,Δf) can be obtained by [S0*h](d e ,Δf) model, where S0(d e , Δf) is again the amplitude distribution of the backscatter of the contents of the survey zone ZE during the active acquisition sequence, and * denotes a two-dimensional convolution operation.

[0079] The calculation unit 11 then calculates the spectrum Sp(d e ,Δf) to determine the backscattering amplitude distribution S0(d e ,Δf) is the estimated value of the backscatter amplitude distribution S0(d e This estimate of Δf) can constitute the measurement result RM and is denoted as S(d e ,Δf). To this end, several algorithms may alternatively be used, including the following two algorithms given as non-limiting examples.

[0080] Two-dimensional decomposition algorithm This algorithm uses the impulse response h(Δd e ,Δf) and applied to the spectrum Sp(d e , Δf). It is usually called a two-dimensional decomposition algorithm and can use the maximum likelihood method, which consists in minimizing a data attachment criterion (also called a cost function, such as the least squares criterion) that measures the spectrogram Sp(d e ,Δf) and [S*h](d e,Δf). A second possible approach, called the a posteriori maximum method, consists in adding a regularization criterion (also called a penalty criterion) to the data attachment criterion, said regularization criterion containing information from a priori knowledge of certain characteristics of the content of the survey area ZE. This information may be, for example, that each spectrum has a minimum spectral width. An example of a reference study on this a posteriori maximum method is "Bayesian Approach to Inverse Problems", edited by J. Idier, ISTE / John Wiley (London, 2008), pp. 243-283. An equally possible third approach may be a stochastic method, also known as a Monte Carlo method, in which the solution is sought by exploring a set of possible states of the content of the survey area ZE according to at least one random characteristic. This Monte Carlo method is further called a Markov chain, in which each new extraction of the random characteristic depends only on the result of the current extraction, and not on the results of earlier extractions. An example of a reference study on these stochastic methods is "Monte Carlo Statistical Methods" by Christian Robert and George Casella, Springer-Verlag, Springer Series in Statistics, 2010. Generally, the two-dimensional decomposition method is well known to those skilled in the art and therefore does not need to be described further here. The result of the two-dimensional decomposition is an estimate S(d e ,Δf), which depends on two parameters: the separation distance and the frequency shift. This estimate S(d is generated by applying a two-dimensional decomposition algorithm. e ,Δf) can be improved by an additional step, which consists in determining the separation distance d e Each value of reduces the radial velocity distribution v r or the width of the associated Doppler frequency shift Δf can in particular be determined by e For each value of , the maximum reduction is obtained by keeping the following: only the value of the frequency shift Δf - or the radial velocity v r The value of - corresponds to the estimated distribution S(d e ,Δf). Another possible maximum reduction method is to e , as a function of the frequency shift Δf and individually for the separation distance d e For each value of , only keep the value corresponding to the estimated distribution S(d e The center of the peak can be determined, for example, from a segment at half-maximum of the peak or by fitting the peak to its model.

[0081] Figure 4a is a spectrum Sp (d e ,Δf). The horizontal axis of the diagram refers to the separation distance d between 0 and 250 m. e The vertical axis of the diagram refers to the frequency shift Δf between -250 MHz and 0, and the third axis of the diagram associated with the indicated horizontal line refers to the spectrum amplitude values. This spectrum has a frequency shift of Δf between -250 MHz and 0, and the third axis of the diagram associated with the indicated horizontal line refers to the spectrum amplitude values. e The apparent width of each value of is parallel to the frequency shift axis Δf.

[0082] Figure 4b and Figure 4c The horizontal and vertical axes of the diagram are Figure 4a same.

[0083] Figure 4b is the second three-dimensional graph, which shows the Figure 1 The estimated distribution S(d e ,Δf). S(d e ,Δf) is therefore referenced by the third axis of the diagram, which is perpendicular to the separation distance d e and the axis with frequency shift Δf. For the separation distance d e For each value of , the width of the frequency shift Δf decreases dramatically.

[0084] Figure 4c is a third three-dimensional diagram showing the separation distance d e Each value of the estimated distribution S(d e ,Δf) has been limited to its maximum value parallel to the frequency shift axis Δf. This measurement RM shows that the air is passing through a d corresponding to about 45 m. e The system 10 is moved locally in the first zone of the system 10 and at the same time at a d corresponding to about 100 m. e The laser beam is moved into a second region of the system 10 but now by moving away from the system 10. These regions of air movement correspond to two approximately synchronized vortices of a low-state von Karman turbulence structure, the vortex axes of which are perpendicular to the direction from which the laser pulses are emitted by the system 10, and through which the laser pulses propagate at a distance from the vortex axes. Figure 4a-4c The general frequency shift of -120 MHz appearing in the three figures of FIG is the frequency shift added by the AOM 2, making it possible to distinguish between positive and negative radial velocity values ​​based on the measurement signal SM.

[0085] Algorithm for decomposition with sparse adjustmentThis algorithm consists in searching directly for a single frequency shift value or a single radial velocity value v for each value of the separation distance in the survey zone ZE. r The backscattering amplitude distribution S0(d e ,Δf). It then performs these values ​​Δf(d e ) and are also assigned one-to-one to the separation distance value d e At each iteration, the estimated distribution S(d e ,Δf) and impulse response h(Δd e ,Δf) and the spectrum Sp(d e ,Δf) for comparison. Then, when reducing [S*h](d e ,Δf) and Sp(d e ,Δf) between the frequency shift and the backscatter amplitude value assigned to at least one of the separation distance values. Such iterations are repeated until a convergence criterion is met. As is known, a threshold criterion may be applied to [S*h](d e ,Δf) and Sp(d e Alternatively, another possible criterion involves reducing [S*h](d e ,Δf) and Sp(d e ,Δf). The use of this decomposition algorithm for the sparsity adjustment allows the measurement result RM to better resolve the variations in the radial velocity as a function of the separation distance corresponding to high spatial frequencies.

[0086] To provide a more complete representation of the three-dimensional atmospheric portion, the system 10 can be moved, for example, with a variable orientation of the laser pulse emission direction, to perform a scan in the atmospheric portion. The acquisition sequence is then repeated each time with a new orientation of the system 10.

[0087] It will be appreciated that the present invention can be reproduced by modifying minor aspects of the embodiments described in detail above while maintaining at least some of the advantages indicated above. Specifically, all numerical values ​​given are for illustrative purposes only and may vary depending on the application in question. Although an application to fluid flow characterization is used as an example, the present invention is applicable to other applications. Furthermore, the present invention can be applied to pulsed radars operating, for example, in the radio frequency domain with a wavelength between 1 mm and 7.5 mm for each pulse of radiation.

Claims

1. A method for detection and telemetry using electromagnetic radiation pulses to determine the separation distance (d) within a zone of investigation (ZE) e ) is a function that characterizes the radial velocity distribution (v r ), which comprises the following steps: 1) Acquiring a system (10) for detection and telemetry by using electromagnetic radiation pulses, adapted to emit during an acquisition sequence at least one electromagnetic radiation pulse (P), to detect a portion of said at least one radiation pulse backscattered by at least one target (100) present in said survey zone (ZE), and to generate a measurement signal (SM) corresponding to said backscattered and detected portion of the radiation pulse, wherein said measurement signal contains information about a separation distance (d e ) and the radial velocity of each target (v r ), the information about the radial velocity of each target corresponding to the frequency shift due to the Doppler effect that occurs when the radiation is backscattered by this target; and 2) performing an acquisition sequence by controlling the system (10) to transmit the at least one pulse (P) into the survey zone (ZE), The method is characterized in that it further comprises the following additional steps: 3) obtaining a characterization of an impulse response of the system (10), wherein the impulse response corresponds to a pulse response generated by the system (10) during an acquisition sequence and when a single backscatter element is located in the survey zone (ZE), the single backscatter element corresponding to a single separation distance value (d e ) and has a known radial velocity value relative to the system (v r ) and the measurement signal (SM) generated when the survey area has no backscatter elements other than the single backscatter element; as well as 4) by considering the measurement signals (SM) generated in step 2) as corresponding to the values ​​of the separation distance (d e ) and radial velocity value (v r ) and multiplied by the backscatter amplitude value, to obtain measurement results (RM) in the form of pairs by decomposing the measurement signal into several weights, each pair consisting of a backscatter amplitude value and a radial velocity value, and which are respectively assigned to several separation distance values ​​within the survey area.

2. The method according to claim 1, characterized in that The impulse response is the separation distance (d e ) or a round trip time of the radiation from the optical exit of the system (10), and is also a function of one of: - frequencies of spectral components of the portion of the radiation pulse that is backscattered and subsequently detected by the system (10); - a frequency shift (Δf) between the spectral components of the portion of the radiation pulse that is backscattered and then detected by the system (10) and the radiation of each pulse emitted by the system, or The radial velocity value (v r );as well as - the frequencies of the spectrum of said measurement signal (SM) used in step 3).

3. The method according to claim 1 or 2, characterized in that In step 3) by using a system (10) for detection and telemetry using electromagnetic radiation pulses and using the determined separation distance (d e ) performing at least one acquisition sequence for a single backscatter element at ) to obtain said characterization of said impulse response, Alternatively, the characterization of the impulse response is obtained by performing a numerical simulation of the operation of the system (10) when a single backscatter element is present at the determined separation distance in the survey region.

4. The method according to claim 1 or 2, characterized in that The system (10) for detection and telemetry using electromagnetic radiation pulses is adapted to implement a heterodyne detection mode, and the measurement signal (SM) generated at each acquisition sequence and used in step 4) is a heterodyne measurement signal.

5. The method according to claim 1 or 2, characterized in that said system (10) for detection and telemetry using pulses of electromagnetic radiation is of the LIDAR type and said radiation of each pulse (P) emitted by said system is laser radiation, or The system (10) for detection and telemetry using pulses of electromagnetic radiation is of the RADAR type, and the radiation of each pulse (P) emitted by the system has a vacuum wavelength between 1 mm and 7.5 mm.

6. The method according to claim 5, characterized in that The system (10) for detection and telemetry using electromagnetic radiation pulses is of the LIDAR type, and wherein each radiation pulse (P) has at least one of the following characteristics: - the wavelength of the radiation within the pulse (P) is between 250 nm and 10 μm; - the duration of the pulse (P) is between 50 ns and 1 μs; and - the pulses (P) have a half-peak frequency width of less than 1 GHz.

7. The method according to claim 1 or 2, characterized in that In step 4) the measurement result (RM) is inferred from the measurement signal (SM) by applying a two-dimensional decomposition algorithm using a method selected from a list comprising a posteriori maximum method, a maximum likelihood method and a random method.

8. The method according to claim 7, further comprising an additional step, performed on the basis of the measurement results (RM) provided by the two-dimensional decomposition algorithm, in order to reduce the separation distances (d e ) for several values ​​of radial velocity (v r )The width of the distribution.

9. The method according to claim 1 or 2, characterized in that Step 4) consists of converting a single radial velocity value (v r ) and a single backscatter amplitude value is assigned to the separation distance (d e ) for each value of the separation distance value inside the survey zone (ZE), and then calculating a reconstruction of the measurement signal for all separation distance values ​​in the sampling as the sum of weights each equal to the impulse response applied to the separation distance value and the radial velocity value assigned to the separation distance value, multiplied by the backscatter amplitude value also assigned to the same separation distance value, and then performing a series of iterative adjustments of the assigned radial velocity and backscatter amplitude values ​​in order to reduce the deviation between the measurement signal produced in step 2) and the reconstruction of the measurement signal produced by the assigned radial velocity and backscatter amplitude values, wherein the separation distance value (d e ) of the radial velocity (v r ) and the backscatter amplitude value form the measurement result (RM).

10. The method according to claim 1 or 2, which is used in at least one of the following applications: - Meteorological measurements; -Measurement of the dispersion of atmospheric pollutants; - measurement of the local concentration of backscattered particles suspended in the environment or chemical compounds that absorb and re-emit the radiation of the pulse; -Measurement of shear force of atmospheric air; - measurement of the position and / or lifetime of at least one vortex present in the fluid flow; - measurements of wind speed from an aircraft in flight; - measurements of the determined wind force performed to optimize the operation of the wind turbine; as well as -Measurements of wind speeds performed to adjust aircraft flight formations or to adjust the flight of drones. The method according to claim 6 , wherein the half-peak width is less than 20 MHz.

12. The method according to claim 7, wherein the stochastic method is a Markov chain Monte Carlo method.

13. The method of claim 10, wherein the meteorological measurement is a measurement of atmospheric turbulence. The method of claim 10 , wherein the measurement of the shear force of the atmospheric airflow is a measurement of the shear force of the atmospheric airflow at an airport. The method according to claim 10 , wherein the aircraft is a spacecraft or a drone.

16. A system (10) for detection and telemetry using electromagnetic radiation pulses, comprising: - a transmission path (10E) adapted to transmit at least one electromagnetic radiation pulse (P) into the survey zone (ZE) when performing an acquisition sequence; a detection path (10D) adapted to detect a portion of the at least one emitted radiation pulse (P) after said portion of the radiation pulse is backscattered by at least one target (100) present in the survey zone (ZE) during the acquisition sequence, wherein the detection path is further adapted to generate a measurement signal (SM) containing information about the separation distance (d e ) and the radial velocity (v r ), the information about the radial velocity of each target corresponding to a frequency shift due to the Doppler effect that occurs when the radiation is backscattered by the target; and - a controller (9) arranged for activating said emission path (10E) and detection path (10D) according to said acquisition sequence, Characterized in that it further comprises: - storage means (12) for characterization of an impulse response of the system (10), wherein the impulse response corresponds to a pulse response generated by the system during an acquisition sequence and when a single backscatter element is located in the survey zone (ZE), the single backscatter element corresponding to a single separation distance value (d e ) and has a radial velocity value relative to the system (v r ) and said measurement signal (SM) generated when said survey area has no backscatter elements other than said single backscatter element; and - calculation means (11) adapted to decompose said measurement signal (SM) into components each corresponding to a voltage applied to said separation distance (d e ) and the radial velocity (v r ) and multiplied by the backscatter amplitude value so as to output measurement results (RM) in the form of pairs, each pair consisting of a backscatter amplitude value and a radial velocity value, and respectively assigned to several separation distance values ​​within the survey zone (ZE).

17. The system (10) according to claim 16, adapted to implement the method according to any one of claims 1 to 15.

Images