Method for detecting and locating targets in bistatic mode

Abstract

The invention relates to a method for detecting and locating (RE) targets (C), the method being designed to use, in bistatic mode, non-cooperative radio signals transmitted on separate channels by a transmitting source (EM), the device including a plurality of 6 to 10 receiving elements (EREi), i ∈ [1; N], as well as a processing unit (ECU) for processing the radio signals, characterised in that: - the receiving elements (EREi) are substantially aligned along an axis (x), and - the value of the spacing d(EREi, EREi+1) between two consecutive receiving elements (EREi, EREi+1) is greater than the half-wavelength of the radio signals, and - ∀ i ∈ [1; N] and ∀ k ∈ [1; N], the value of the spacing d(EREi, EREk) between two separate receiving elements (EREi, EREk) is unique.

Description

Method for detecting and locating targets in bistatic mode The present invention relates to passive methods and systems exploiting opportunistic radio and television signals (FM, Digital Audio Broadcasting, Digital Video Broadcasting) to detect and locate targets. These non-cooperative emissions are, for example, broadcast signals such as FM radio or television signals such as DVB (Digital Video Broadcasting). The invention relates in particular to passive bistatic radars operating in the UHF or VHF bands, dedicated to the detection and localization, particularly Cartesian localization, of small aerial targets operating primarily at low altitudes. These passive bi-static radars make it possible to detect intrusions even in the absence of a high density of transmitters, on sensitive sites such as military infrastructure and strategic industrial sites. These passive bi-static radars also allow for initial classification based on the specific signatures of intruders and / or for the characterization of targets of no interest. State of the art Passive systems are traditionally seen as discrete means of detection and localization, primarily for supplementary coverage such as low and very low altitude detection. However, today all these passive systems that exploit opportunistic transmissions such as television or radio broadcasts with a carrier frequency below 1 GHz face challenges with poor angular measurement. This low-frequency measurement is indeed limited in accuracy when the antenna array consists of a small number of antennas classically spaced half a wavelength apart between two successive antennas and presents ambiguities when the array uses spacings greater than half a wavelength without having been specially defined to limit the ambiguities due to these "too large" spacings. To address this ambiguity issue, the document "Target DoA estimation in passive radar using non-uniform linear arrays and multiple frequency channels," 2018 IEEE RADAR CONFERENCE (RADARCONF18), IEEE, April 23, 2018 (2018-04-23), proposes a non-uniform linear array (NULA) of antennas to expand the unambiguous angular sector. It also suggests the simultaneous use of multiple DVB-T (multi-frequency (MF)) channels with different frequencies in the UHF band (470–862 MHz), which create their own ambiguity but cancel each other out, in order to increase the signal-to-noise ratio and consequently the accuracy. This document specifically recommends using three frequencies on three antennas, resulting in a total of nine receiving channels, to resolve the ambiguities. The paper "A simple NULA design strategy for target detection and DoA estimation in mobile passive radar," published in IET Conference Proceedings, vol. 2022, October 24, 2022, addresses the design of non-uniform antenna arrays (NULA) to improve target detection and angle-of-arrival (DoA) estimation in passive radars mounted on mobile platforms (e.g., vehicles or aircraft) using DVB-T as an opportunity signal. This paper proposes a compromise between angular resolution, which requires a long linear antenna array, and the absence of ambiguities, which necessitates small antenna spacings. The proposed solution is to impose a minimum peak-to-secondary-lobe ratio (PSR) to select the antenna spacing configuration, and then validate this approach in a DVB-T scenario. This document imposes multiple half-wavelength spacings and also recommends 3, 4, 5 antennas with two frequencies. The document "DVB-T based passive radar for simultaneous counter-drone operations and civil air traffic surveillance," published by IET RADAR SONAR NAVIGATION, THE INSTITUTION OF ENGINEERING AND TECHNOLOGY, UK, vol. 14, no. 4, April 1, 2020, also addresses the design of non-uniform antenna arrays (NULA) to improve target detection and angle of arrival (DoA) estimation in passive radars. Specifically, this document focuses on the effectiveness of passive radar based on digital terrestrial television (DVB-T) for counter-drone operations, particularly for airport area surveillance. The proposed architecture uses two frequencies and three surveillance antennas to unambiguously monitor a sector. Thus, these documents use few antennas, which implies the use of several frequencies and makes measurements and calculations complex. Based on these observations, the inventors decided to define a sectoral surveillance method using a single bistatic pair, i.e. a transmitter and a receiver, the pair operating with a single frequency allowing direct and unambiguous localization of all targets. More specifically, the invention relates to a method for detecting and locating targets using a target detection and location device that exploits non-cooperative radio signals emitted on separate channels by a transmitting source, said device comprising a plurality N of 6 to 10, preferably between 7 and 9, receiving elements, such as antennas, and a unit for processing said radio signals, wherein: The receiving elements are substantially aligned along an axis and, The value of the spacing between two successive receiving elements is greater than half the wavelength of said radio signals, and, The value of the spacing between two distinct receiving elements is unique.said method comprising at least: A reception step by each receiving element of the target detection and localization device, of a signal from the emitting source along a direct path, and of a signal backscattered by a target in the covered space, A selection step allowing, for each receiving element of the target detection and localization device, the separation and selection of a single frequency channel to be exploited, this channel being common to all N receiving elements, A step of determining the bistatic distance (d) and the radial velocity (v). r ) of the target, itself derived from said bistatic distance, A step of determining the angle of arrival of the target, based on measurements of the phase differences between the signals received by each receiving element of the target detection and localization device. This device allows for precise and unambiguous localization based on the use of a single frequency emitted by the source and received on the N receiving elements. The fact that this linear device is lacunary, that is to say that the spacing between two successive receiving elements is greater than half the wavelength of the radio signals, allows to obtain an improved resolution, due to the increase in the length of the antenna array (i.e. the distance between the two extreme receiving elements of the array). The fact that this linear device is irregular, that is to say that the spacings between any two receiving elements are all different from each other, helps to limit ambiguities. Various embodiments of the invention are envisaged, incorporating, according to all their possible combinations, the different optional features described herein. According to a preferred execution mode, the step of determining the bistatic distance (d) and the radial velocity (v r ) of the target, is implemented by means of a distance-Doppler correlation operation on the received signal, with replicas of the signal emitted by the emitting source, affected by time and frequency shifts. In particular, the step of determining the bistatic distance (d) and the radial velocity (v r ) of the target, implements the following sub-steps: The expression of the signal received by a receiving element (ERE i ) depending on the signal received by a reference receiving element (ERE ref ) shortlisted: With : signal received by the receiving element (ERE) i ), : signal emitted by the emitting source (EM), : frequency of the emitting source (EM), : bistatic delay (common to all receptor elements (ERE) i )) between the path from the emitting source (EM) to the target, and the path from the target to the receiving element (ERE) i ), : derivative of the delay Bistatic Doppler : spacing value between the receiving element (ERE) i ) and the reference receptor element (ERE ref ) predetermined : angle of arrival of the target common to all receiving elements (ERE) i )Parameter estimation And by means of finding the maximum power of the function : With conjugate of a replica of the signal emitted by the emitting source (EM). Calculation of the bistatic distance (d) and the radial velocity (v r ) of the target using the following formulas: And According to a preferred method of execution, the step of determining the angle of arrival of the target, proceeds to an estimation of the angle of arrival of the target by means of a search for the maximum signal-to-noise ratio of the following function: According to another alternative execution method, the step of determining the angle of arrival of the target, is an estimate of the angle of arrival of the target by means of a search for the maximum signal-to-noise ratio of the following function: The focused signal then to be compressed into bistatic distance, Doppler according to a step of determining the bistatic distance and the radial velocity of the target. Preferably, the minimum difference between the values ​​of the spacings between any two receiving elements is at least equal to 10%. Thus, the spacings between any two receiving elements are clearly differentiable, which makes it possible to eliminate ambiguities. Advantageously, the value of the spacing between two successive receiving elements is shaped to exploit radio signals used for broadcasting analog and digital television or radio, after adaptation to transpose the dimensioning carried out in the television frequency band to the radio frequency band. This allows easy access to sources of opportunity across the entire territory. Advantageously, the value of the spacing between the first receiving element and the last receiving element is less than 20 m, preferably between 8 and 10 m. The device according to the invention can thus be easily installed on a roof or a building terrace. Brief description of the FIGURES The invention will be better understood upon reading the following description, given solely by way of non-limiting example and with reference to the accompanying drawings, in which: Figure 1 is a schematic representation of a bistatic couple with a transmitter and a receiver on which the invention is based. Figure 2 is a schematic representation of a receiver of the bistatic couple, according to a non-limiting embodiment of the invention. Figure 3 is a comparative schematic representation of the results obtained with a receiver according to the invention and receivers of the prior art. Figures 4 and 5 show comparative results between the localization performed using a detection and localization device according to the invention and the actual localization provided by the GPS position extracted from cooperative drones. It is understood that the embodiments described below are by no means exhaustive. In particular, variants of the invention may be conceived comprising only a selection of the features described below, isolated from the other features described, if this selection of features is sufficient to confer a technical advantage or to differentiate the invention from the prior art. This selection includes at least one preferably functional feature without structural details, or with only a portion of the structural details if this portion alone is sufficient to confer a technical advantage or to differentiate the invention from the prior art. In particular, all the variants and embodiments described can be combined with each other if there are no technical obstacles to this combination. In the figures and in the rest of the description, elements common to several figures retain the same reference. Detailed description of the FIGURES We have represented on the diagram a bistatic couple consisting of an emitting source EM and a detection and localization device RE separated by a distance d(EM,RE), as well as a target C, distant from the emitting source EM by a distance d(EM,C), and distant from the device RE by a distance d(RE,C). Passive detection in bistatic mode therefore involves the emitting source EM on the one hand and the detection and localization device RE on the other. The EM emitting source is considered non-cooperative in that it emits signals whose destination and nature are not related to the use made of them by the RE device for detecting and locating C targets. These are, for example, FM band broadcasting signals intended for the broadcasting of public radio programs, or UHF or VHF broadcasting signals of the television type for example, or any other general-purpose emission. A characteristic of these non-cooperative transmissions is that they are generally carried out within a defined frequency band, which is subdivided into disjoint sub-bands also called channels. For a given geographic region, each channel is allocated to a specific use, corresponding to a particular type of transmission. For example, in the case of a transmitter intended for broadcasting FM programs, the frequency band on which the transmitter is likely to transmit extends from 87.5 MHz to 108 MHz and is divided into channels approximately 100 kHz wide. Within this band, all or part of the channels may be used. Each channel used can also be intended for one or more uses such as the broadcasting of speech programs, music, images or multiplexed digital data. Thus, the bandwidth of the FM signal transmitted on each channel is independent of that of the signals transmitted on the other channels and can vary over time depending on the signal or information being transmitted. For this reason, the EM transmitting source is described as non-cooperative and random. The RE detection and localization device utilizes this EM emitting source by implementing a bistatic detection process. This process involves separately receiving the direct signal emitted by the EM source and receiving the backscattered waves from targets within the area covered by the RE detection and localization device. Generally, for ease of implementation, this reception is limited to a single channel emitted by the EM source. Indeed, the signals emitted on the different channels are not necessarily synchronous, and their simultaneous processing requires appropriate processing. The signal directly received from the emitting source EM is used as a reference by the detection and localization device RE to determine the time difference that the signal backscattered to the detection and localization device RE by the target C presents with respect to this reference. This difference allows, after processing, the determination of the bistatic distance of the target C, the velocity of the target C being determined elsewhere, in a conventional way, by Doppler processing. Separate reception of the direct signal and the backscattered signal can be achieved by using a receiving element, such as a directional reference antenna, pointed in the direction of the emitting source or not, and another receiving element, such as an antenna, pointed in a given direction. Separate reception of the direct signal and the backscattered signal can also be achieved by the same receiving element. As represented in, the RE target detection and localization device C is configured to exploit in bistatic mode non-cooperative radio signals emitted on separate channels by an EM emitting source. The RE device comprises a plurality N of 7 to 9 receiving elements ERE i , i ∈ [1 ; N], as well as an ECU processing unit for received radio signals. ERE receptor elements i are substantially aligned along an axis (x) so as to define a broken line L, close to the main axis (x). The term "approximately aligned" refers to the fact that the ERE receptor elements i form a linear network that is almost rectilinear in the sense that each receiving element does not deviate from the (x) axis by a distance greater than 25% of the length of the broken line L. According to the invention, the ERE receiving elements iform a linear lacunar network insofar as the value of the spacing d(ERE i ,ERA i+1 ) between two ERE receptor elements i , ERE i+1 The successive wavelength is greater than half the wavelength of said radio signals. This characteristic allows for good resolution. According to the invention, the ERE receiving elements i form an irregular linear lacunar network insofar as ∀ i ∈ [1 ; N] and ∀ k ∈ [1 ; N], the value of the spacing d(ERE i ,ERA k ) between two distinct receptor elements ERE i , ERE k is unique. Thus, in this irregular network, the same spacing value is never found twice between two distinct receiving elements. This feature minimizes ambiguities in the target's angle of arrival when the spacing value d(ERE) i ,ERA i+1) between two ERE receptor elements i , ERE i+1 successive is greater than half the wavelength of said radio signals. Advantageously, in order to minimize ambiguities as much as possible on the angle of arrival of the target, the minimum difference between the values ​​of the spacings is at least equal to 10%. Preferably, and for practical implementation reasons, the value of the spacing d(ERE i ,ERA i+1 ) between two ERE receptor elements i , ERE i+1 successive is configured to exploit radio signals used for the broadcasting of analog and digital television or radio, after adaptation to transpose the dimensioning carried out in the television frequency band to the radio frequency band. In order to allow the installation of the RE detection and location device on the roof of a building, the spacing value d(ERE1,ERE) N ) between the first ERE1 and the last ERE N receiving elements of the irregular gapless linear network, is preferably less than 20 m. The target detection and localization device comprises at least 6 ERE receiver elements i at most 10 ERE receptor elements i , preferably between 7 and 9 ERE receptor elements i This allows for an efficient system while minimizing the number of components, ensuring controlled cost and footprint with reasonable deployment complexity. In order to cover all directions, i.e. an angular sector of 360°, several RE target detection and localization devices conforming to the invention can be installed, the whole then forming an SD target detection and localization system C. In summary, this target detection and localization device, which constitutes a passive sector surveillance system, can be coupled with other similar devices to ensure omnidirectional protection, each device being capable of directly and precisely detecting and locating targets of interest. The primary, but potentially expandable, application is in counter-drone operations using DVB-T / DVB-T2 transmitters. DVB-T broadcasts have the advantage of a long integration time, typically between 0.5 seconds and 1 second. The target detection and location system uses a limited number of receiving elements (antennas), for example eight, and covers a sector of approximately + / -30°. Other limitations may also be considered. The target detection and localization system is essentially linear and requires a small footprint, with a maximum range of approximately ten meters. This limitation allows for installation on any type of flat building terrace. It is compatible with DVB-T frequencies between approximately 500 MHz and 700 MHz, corresponding to a wavelength of around 50 centimeters. The target detection and localization system consists of an irregular, patchy antenna network. The implementation of a target detection and localization device according to the invention will now be described. The target detection and localization method C according to the invention uses a target detection and localization device exploiting non-cooperative radio signals emitted on separate channels by an EM emitting source, according to the invention. The process includes at least: A reception step 100 by each receiving element ERE i of the target detection and localization device, a signal from the emitting EM source along a direct path, and a signal backscattered by a target C in the covered space. A selection step 200 allowing, for each receiving element ERE i of the target detection and localization device, the separation and selection of a single frequency channel to be used, a step of determining the bistatic distance d and the radial velocity v r of the target, A step of determining the angle of arrival 400 of the target, based on measurements of the phase differences between the signals received by each ERE receiving element i of the target detection and localization device. Preferably, but not exclusively, the step 300 of determining the bistatic distance d and the radial velocity v r of the target, is implemented by means of a distance-Doppler correlation operation on the received signal, with replicas of the signal emitted by the EM emitting source, affected by time and frequency shifts. The implementation of step 300 makes it possible to determine the delay values ​​for a signal received at a given time and frequency shift for which a correlation peak is obtained between the measurement signal and the delayed reference signal, which is affected by a corresponding Doppler. A correlation peak is obtained when a target C, located at a given distance from the detection and localization device RE and moving at a given speed, reflects the wave emitted by the emitting source EM back towards the device RE. The delay Furthermore, it allows the corresponding bistatic distance to be determined directly. This bistatic distance allows the target C to be located on an ellipsoid whose foci are the emitting source EM and the target detection and localization device. The determination of the target's Cartesian coordinates is itself obtained by knowing the angle arrival of the target towards the target detection and localization device RE.

[0034] The value f of the frequency shift for which a correlation peak is obtained corresponds to the Doppler frequency of the signal reflected by object 17 and is directly related to the Doppler velocity Vd of object 17, relative to the detection system 12. In particular, step 300 of determining the bistatic distance d and the radial velocity v r of the target, implements the sub-steps described below. First, we express the signal received by a receiving element ERE i depending on the signal received by a reference receiver element ERE ref shortlisted: With : signal received by the ERE receiving element i , : signal emitted by the EM emitting source, : frequency of the EM emitting source, : bistatic delay (common to all ERE receptor elements) i ) between the path from the emitting EM source to the target, and the path from the target to the receiving ERE i , : derivative of the delay Bistatic Doppler : spacing value between the ERE receiving element i and the reference receptor element ERE ref predetermined : angle of arrival of the target common to all ERE receiving elements i Indeed, the analysis step 300 allows us to recover the parameters of bistatic distance τ, bistatic Doppler ν which is linked to the derivative of the bistatic distance and angle of arrival with respect to the detection and localization device RE by proceeding as follows. Within the received signal, the signal backscattered by the target can be approximated and schematically represented as follows on each of the ERE receiving elements i of the RE system: With : signal received by the ERE receiving element i , : signal emitted by the EM emitting source, : frequency of the EM emitting source, : bistatic delay (common to all ERE receptor elements) i ) between the path from the emitting EM source to the target, and the path from the target to the receiving ERE i , : derivative of the delay Bistatic Doppler : difference in path length between the path from target C to the ERE receptor element iand the path from target C to the reference receptor element ERE ref The ERE reference receptor element ref is chosen arbitrarily. As can be seen in FIGURE 2, = d(ERE i , ERA ref ) * cos . We can therefore rewrite the expression for the received signal as follows: Next, to estimate the parameters And (common to all ERE receptor elements) i ), it suffices to perform a so-called distance correlation operation, Doppler, according to the principle of finding the maximum power of the following function depending on the assumptions as follows: With combined with a replica of the signal emitted by the EM emitting source. Since we are looking for a maximum in terms of energy ratio, the phase term in does not need to be compensated. Finally, the calculation of the bistatic distance d and the radial velocity v r The target can be determined using the following formulas: And The 400 step for determining the angle of arrival of the target, consists of an estimation of the angle of arrival of the target. To do this, we perform a search operation to find the maximum signal-to-noise ratio of the following function: The relevance of the estimate of the angle of arrival The target's capacity is intimately linked to the spacing values ​​between the ERE receptor elements. i . Alternatively, step 400 determines the angle of arrival of the target, is an estimate of the angle of arrival of the target by means of a search for the maximum signal-to-noise ratio of the following function: , the focused signal then to be compressed into bistatic distance, Doppler according to a determination step 300 of the bistatic distance d and the radial velocity v r of the target. When the array is extended lengthwise while keeping the number of antennas unchanged, the antennas are spaced further apart, which leads to a greater risk of angular ambiguity, since overall angular resolution is inversely proportional to the total array length. This risk can be mitigated by using different frequency channels, as the position of ambiguities varies with wavelength. Angular resolution is indeed inversely proportional to the total array length because the phase shift term is: Laestablishes the contribution of irregular gap linear networks (dotted lines) according to the invention compared to classic linear networks (solid black line) and regular gap linear networks (solid grey line), in terms of isolation, resolution and ambiguity. We recall that isolation represents the difference in levels between the true direction of a target and other assumptions, resolution represents the ability to separate two "close" targets of the same level, and ambiguities represent other angle assumptions with similar levels. It appears that classical linear networks exhibit good isolation but average resolution, while regular linear gap networks exhibit improved resolution, but with ambiguities. We then see that the irregular gap linear networks according to the invention have an improved resolution, free from ambiguities. Tests were conducted for a DVB-T frequency approximately between 500 MHz and 700 MHz with a wavelength of approximately 50 centimeters. A conventional linear radar array would require spacing them by half a wavelength, resulting in a total span of 7 * 0.25 = 1.75 meters. Furthermore, if, according to the conventional approach, an omnidirectional surveillance approach were considered, using a "circular" array with target detection and localization devices, this array span would be reduced by a factor of two in terms of effectiveness for a given direction. For a regular network of large extent, that is to say for an inter-antenna distance greater than half the wavelength, network lobes then appear which correspond to ambiguities on the direction of arrival of the targets. With irregular gapless linear arrays according to the invention, with a maximum extension of the order of 8 to 9 meters, and a minimum spacing between two successive antennas of 60 centimeters, i.e. greater than lambda, only ambiguities of array lobes are observed outside the relatively constant elementary antenna lobe of the antennas. The basic antenna lobe is generally limited to -5 or -6 dB below the main peak level within the + / -60° main lobe, this lobe being twice the size of the surveillance sector. The surveillance sector is the area where target detection is sought, i.e., a -3 dB lobe. The extended lobe is the area where it is considered possible to still receive contributions from other targets. An antenna deployment then makes it possible to ensure these constraints over the entire DVB-T band, whose frequency channel bandwidth is 7.5 MHz, in order to facilitate implementation during deployments on various sites. Irregular linear gap arrays according to the invention, with a maximum extension of approximately 8 to 9 meters and a minimum spacing of 60 centimeters between two successive antennas, were specified for a total extension of 8.28 meters, or 4.75 times the extension that would have been obtained with a conventional approach. The gain in accuracy obtained on the angular measurement is then close to this factor of 4.75 for the same system complexity. A good resolution of 40 meters for the bistatic range (measured in DVB) was achieved, and the integration time (typically between 0.5 and 1 second for counter-drone applications) leads to a good resolution for the bistatic velocity (derived from the range) on the order of 1 meter / second. These performance levels for both range and velocity are generally accurate enough to contain only a single target, assuming a bistatic range and its derivative. A good estimation of the target's angle of arrival, with a resolution of around 4 degrees linked to the 8-meter extension network, was also obtained. Thus, it is possible to separate two targets sharing the same assumption of bistatic distance and its derivative if they are sufficiently separated angularly. We can therefore obtain detection plots with resolutions for the bistatic distance, the derivative of the bistatic distance, angle relative to the detection and target localization device, of the order of respectively 40 meters, 1m / s, 4 degrees. Accuracy is then deduced from these improved resolutions in inverse proportion to the square root of the signal-to-noise ratio of the targets. The high accuracy of bistatic measurements, particularly of the bistatic distance and angle relative to the RE target detection and localization device, allows for the direct and precise determination, through a simple geometric transformation, of the target's Cartesian coordinates relative to the RE target detection and localization device, or any other reference point. The result of this direct transformation enables Cartesian localization, as opposed to tracking, which also utilizes velocity measurements to determine the target's Cartesian position and its kinematics. It should be noted that Cartesian tracking is also possible using measurements associated with a single bistatic pair, given their high resolution, especially in terms of distance and angle. Figures 4 and 5 show comparative results between the localization carried out by means of a detection and localization device according to the invention and the actual localization provided by the GPS position extracted from cooperative drones. Figures 4 and 5 illustrate, respectively, the angle relative to the antenna array and the Cartesian distance from the receiver. The measurements obtained are compared to estimates derived from a GPS drone trajectory reference. Among the main advantages of this invention, it is noted that the target detection and localization devices according to the invention reduce the impact of the blind sector related to the axis of the emitter. These devices also allow direct bistatic localization in the device's frame of reference with good horizontal accuracy, on the order of 30 m along the two axes of the horizontal plane. Furthermore, the implementation of the devices according to the invention allows for instant localization as soon as a detection occurs, which makes it possible to launch an alert instantly since it does not require simultaneous detections on other transmitter / receiver pairs. The installation and implementation of monitoring using this principle is simpler than systems based on multistatic association, since it is possible to centralize the detection system. Localization is directly accessible from a single transmitter / receiver pair, which simplifies the deployment of the devices according to the invention compared to a conventional passive solution that requires the combination of three bistatic pairs. Indeed, the density of DVB transmitters does not necessarily allow for the use of three separate transmitters for certain areas of interest. It is also possible to combine several devices according to the invention for omnidirectional coverage needs while preserving the independence of the performance of each sector. The detection and localization method and device according to the invention simplify and improve the performance of the principle of associating bistatic markers with a physical object when dealing with a Single Frequency Network. Indeed, the results can be based on track proximity assumptions rather than marker association assumptions. The level of bistatic localization accuracy obtained with the detection and localization method and device according to the invention exhibits a measurement precision sensitive to wind direction, which can allow a certain level of classification. The detection and localization method and device according to the invention is adaptable to a 3D extension, by adding the vertical component. Tracking performed using simple bistatic information makes it possible to overcome potential imperfections when associating poorly georeferenced data. Given the accuracy of the measurement, the kinematic parameters of the targets are particularly well estimated, which helps to sort the plots between the desired plots and the others even if some false alarms may remain. Similarly, the increased accuracy of the measurement makes it possible to: Limit the elimination of road detections to a smaller sector, Limit false alarms related to the selection of hypotheses (distance, angle) possessing a drone-like signature, Of course, the invention is not limited to the examples that have just been described.

Claims

1. A method for detecting and locating targets (C) by means of a target detection and locating device (RE) exploiting non-cooperative radio signals emitted on separate frequency channels by a transmitting source (EM), said device comprising a plurality N of 6 to 10, preferably between 7 and 9, receiving elements (ERE) i ), i ∈ [1 ; N], as well as a radio signal processing unit (ECU), in which: The receiving elements (ERE i ) are substantially aligned along an axis (x) and, The value of the spacing d(ERE i ,ERA i+1 ) between two receptor elements (ERE i , ERE i+1 successive ) is greater than half the wavelength of said radio signals, and, ∀ i ∈ [1; N] and ∀ k ∈ [1; N], the value of the spacing d(ERE i ,ERA k ) between two distinct receptor elements (ERE i , ERE k) is unique. Said process comprises at least: A reception step (100) by each receiving element (ERE) i ) of the target detection and localization device, a signal from the emitting source (EM) along a direct path, and a signal backscattered by a target (C) in the covered area. A selection step (200) allowing, for each receiving element (ERE i ) of the device (RE), the separation and selection of a single frequency channel to be used, this channel being common to all N receiving elements, A step of determining (300) the bistatic distance (d) and the radial velocity (v r ) of the target, itself derived from said bistatic distance (d), A determination step (400) of the angle of arrival of the target, based on measurements of the phase differences between the signals received by each receiving element (ERE) i ) of the device (RE).

2. A method for detecting and locating targets (C) according to claim 1, characterized in that the step of determining (300) the bistatic distance (d) and the radial velocity (v r ) of the target, is implemented by means of a distance-Doppler correlation operation on the received signal, with replicas of the signal emitted by the emitting source (EM), affected by time and frequency shifts.

3. A method for detecting and locating targets (C) according to claim 2, characterized in that the step of determining (300) the bistatic distance (d) and the radial velocity (v r ) of the target, implements the following substeps: The expression of the signal received by a receiving element (ERE i ) depending on the signal received by a reference receiving element (ERE ref ) shortlisted: With : signal received by the receiving element (ERE) i ), : signal emitted by the emitting source (EM), : frequency of the emitting source (EM), : bistatic delay (common to all receptor elements (ERE) i )) between the path from the emitting source (EM) to the target, and the path from the target to the receiving element (ERE) i ), : derivative of the delay Bistatic Doppler : spacing value between the receiving element (ERE) i ) and the reference receptor element (ERE ref ) predetermined : angle of arrival of the target common to all receiving elements (ERE) i )Parameter estimation And by means of finding the maximum power of the function : With conjugate of a replica of the signal emitted by the emitting source (EM). Calculation of the bistatic distance (d) and the radial velocity (v r) of the target using the following formulas: And 4. Target detection and localization method (C) according to claim 3, characterized in that the step of determining the angle of arrival (400) of the target, proceeds to an estimation of the angle of arrival of the target by means of a search for the maximum signal-to-noise ratio of the following function:

5. Target detection and localization method (C) according to claim 3, characterized in that the step of determining the angle of arrival (400) of the target, is an estimate of the angle of arrival of the target by means of a search for the maximum signal-to-noise ratio of the following function: The focused signal then to be compressed into bistatic distance, Doppler according to a determination step (300) of the bistatic distance (d) and the radial velocity (v r ) of the target.

6. A method for detecting and locating targets (C) according to any one of the preceding claims, characterized in that the minimum difference between the spacing values ​​of the receiving elements (ERE) i ) is at least equal to 10%.

7. A method for detecting and locating targets (C) according to any one of the preceding claims, characterized in that the value of the spacing d(ERE i ,ERA i+1 ) between two receptor elements (ERE i , ERE i+1 ) successive is configured to exploit radio signals used for broadcasting analog and digital television or radio.

8. A method for detecting and locating targets (C) according to any one of the preceding claims, characterized in that the value of the spacing d(ERE1, ERE N ) between the first (ERE1) and the last (ERE N ) receiving elements is less than 20 m, preferably between 8 and 10 m.

Images