RADAR SYSTEM AND METHOD FOR DETECTING OCCUPIED FREQUENCY CHANNELS

Description

technical field

[0001] The present invention relates generally to the field of passive multi-channel antenna systems, and in particular to a radar system and a method for detecting occupied frequency channels.

[0002] An antenna system can be disrupted when used in a dense electromagnetic environment, that is, one containing numerous busy (or polluted) frequency channels, each associated with transmitted electromagnetic signals that can be received by the antenna system. Knowing which frequency channels are busy allows an operator of this antenna system to optimize the use of frequency channels for transmission and / or reception by selecting free frequency channels to transmit a new electromagnetic signal and / or process signals of interest.

[0003] Today, there are systems for detecting occupied frequency channels, such as signals intelligence (SIGINT) systems. Signals Intelligence according to the corresponding Anglo-Saxon acronym, or ESM for Electronic Support Measures (according to the corresponding Anglo-Saxon acronym) allowing the detection and localization of electromagnetic signals in a very wide range of frequencies.

[0004] However, such systems present significant implementation complexity and high manufacturing costs. They also impose stringent constraints in terms of installation on a larger system, such as an airborne vehicle. In particular, existing systems are bulky, heavy, and subject to restrictive specifications regarding their specific installation areas on a carrier aircraft like an airplane or helicopter.

[0005] Examples of busy frequency channel detection systems have been described in US documents 5 565 764 A, FR 3 073 627 A1, and EP 4 239 903 A1.

[0006] Furthermore, such systems installed on an airborne vehicle show limited results in terms of estimating the direction of arrival of a detected signal, particularly because the antennas used have limited directivity.

[0007] There is therefore a need for an improved system and method for detecting busy frequency channels. Summary of the invention

[0008] To this end, a radar system for detecting occupied frequency channels is proposed. The system comprises N receiving channels and a set (10) of radar antennas associated with each channel, N being an integer greater than or equal to 2, each channel being capable of receiving signals by means of its associated radar antenna. For each receiving channel, the radar system includes a conversion unit in the spectral domain of the received signals to form a converted signal defined on a given instantaneous frequency band associated with a set of frequency channels, and a correlation value determination module adapted to apply a power correlation between the spectrum of the converted signal propagated on the receiving channel and the spectrum of an arbitrary reference signal chosen from among the converted signals propagated on one of the receiving channels.The radar system further includes an occupied frequency channel evaluation module adapted to determine the sum channel values ​​from the correlation values ​​of the N receiving channels, the frequency channel being occupied if the associated sum channel value is greater than a predefined detection threshold value.

[0009] Advantageously, the correlation value determination module can be configured to determine a power spectral density of the signal propagated on the receiving path relative to the arbitrary reference signal.

[0010] In some embodiments, the correlation value determination module can be configured to apply integration over several successive frames of the signal propagated on the receiving channel.

[0011] In some embodiments, the radar system may further include a calibration device comprising N calibration channels. For each calibration channel, the calibration device can be configured to determine correction coefficient values ​​for the converted signal to balance the amplitude and phase of each of the radar system's receiving channels.

[0012] The radar system can be configured to apply a frequency scan by repeating the evaluation of the detection of busy frequency channels against the detection threshold value, to cover an entire frequency range of predefined interest.

[0013] Advantageously, the radar system can be configured to apply a pointing antenna direction. The busy frequency channel evaluation module can be further adapted to determine channel gap values ​​from the correlation values ​​of the N receive channels, and then channel gap ratio values ​​from the sum channel values ​​and the channel gap values, the pointing antenna direction corresponding to the direction of arrival of the received signals relative to the orientation of the radar system antennas if the channel gap ratio values ​​are less than a predefined lobe threshold value.

[0014] The busy frequency channel evaluation module can be further adapted to determine at least one refined direction of arrival of received signals relative to the orientation of the radar system antennas from the values ​​of the channel gap ratio and the pointing antenna direction.

[0015] In some embodiments, the radar system can be configured to apply a spatial scan by repeating the evaluation of the direction of arrival of the received signals against the lobe threshold value, to cover an entire predefined angular domain of interest.

[0016] The radar system may include an operationally "listening" configuration in the "X-band" type radio frequency domain.

[0017] Furthermore, a method for detecting frequency channels is proposed, implemented in a radar system comprising N receiving channels, where N is an integer greater than or equal to 2. The method includes the reception of signals at each channel. For each receiving channel, the method comprises the following steps: convert, in the spectral domain, the received signals to form N converted signals each defined on a given instantaneous frequency band associated with a set of frequency channels, determine correlation values ​​by applying a power correlation between the spectrum of the converted signal propagated on the receiving channel and the spectrum of an arbitrary reference signal chosen from one of the converted signals propagated on one of the receiving channels.

[0018] The process further includes an evaluation of occupied frequency channels, comprising the steps of: determine the sum channel values ​​from the correlation values ​​of the N receive channels, compare the associated sum channel value to a predefined detection threshold value, and determine that the frequency channel is occupied if the associated sum channel value is greater than the predefined detection threshold value.

[0019] The embodiments of the invention thus provide a radar system and a method for detecting and evaluating busy frequency channels according to a satisfactory estimation of the direction of arrival of a detected signal.

[0020] The embodiments of the invention also provide an efficient and affordable solution in terms of hardware complexity, while limiting manufacturing and installation costs, as well as the energy costs of implementation. The proposed system advantageously has reduced volumetric and mass compactness, as well as an optimized footprint. Description of the figures

[0021] Other features, details and advantages of the invention will become apparent from the description made with reference to the attached drawings given by way of example. [ Fig.1 ] There figure 1is a diagram representing a radar system for detecting occupied frequency channels, according to embodiments of the invention. Fig. 2 ] There figure 2 consists of two diagrams (a) and (b) representing a measurement module for a receiving channel of a radar system, according to embodiments of the invention. Fig.3 ] There figure 3 is a diagram representing a calibration device for a radar system, according to embodiments of the invention. Fig. 4 ] There figure 4 consists of two diagrams (a) and (b) representing respectively a noise correction measurement module for a noise processing channel and a correction measurement module for a calibration channel of a radar system calibration device, according to embodiments of the invention. Fig. 5 ] There figure 5 is a diagram representing a 2D, 4-way receiving antenna, according to embodiments of the invention. Fig. 6 ] There figure 6 is a flowchart representing a method for evaluating the frequency channels of a received signal, according to embodiments of the invention. Fig. 7 ] There figure 7 is a flowchart representing a method for estimating the direction of arrival of a received signal, according to embodiments of the invention.

[0022] Identical references are used in the figures to designate identical or analogous elements. For clarity, the elements shown are not to scale. Detailed description

[0023] There figure 1 schematically represents a radar system for detecting occupied frequency channels 1 comprising an array 10 of N radar electromagnetic antennas (or an array 10 of N receivers of a radar electromagnetic antenna) each associated with a receiving channel 20n and an evaluation module 30 of occupied frequency channels, according to embodiments of the invention.

[0024] The radar system 1 thus comprises a number N of receiving channels 20n, each channel being capable of receiving one or more received signals (i.e. received time signals), denoted Srn, coming (i.e. by means of) one of the radar antennas of the set 10. The index 'n', associated with the receiving channels, is an integer between 1 and N, and the value of N being an integer greater than or equal to 2.

[0025] The radar system 1 can be used in various civil or military fields, particularly for applications of detecting and locating electromagnetic emissions in a dense electromagnetic environment.

[0026] Advantageously, radar system 1 can be a multi-antenna (or multi-channel) system configured in "listening" mode, that is, an operational configuration that uses signals from the antennas (or antenna) capturing the external environment and from the various receiving channels without emitting any electromagnetic signal. Thus, as used here, an 'operationally "listening" configuration' refers to a configuration in which the radar system operates in "listening" mode.The operational frequency range of such a system can include a very wide range of frequencies, and be located in the radio frequency (RF) range corresponding, for example and without limitations, to an RF band of type "X band" typically between 8 GHz and 12 GHz, to an RF band of type "K band" typically between 12 GHz and 18 GHz, to an RF band of type "K band" typically between 18 GHz and 26 GHz, or even to an RF band of type "Ka band" typically between 26 GHz and 40 GHz.

[0027] The radar system 1 can in particular use a radar antenna device installed on board a carrier, of small size such as a helicopter for example, making it possible to provide an affordable system in terms of hardware implementation while limiting manufacturing and installation costs.

[0028] As depicted on the figure 1, each receiving channel 20 n of the radar system 1 comprises a conversion unit 22 n and a measurement module 24 n , according to embodiments of the invention.

[0029] The conversion unit 22n is configured to perform a conversion (or pass-through), in the spectral domain (i.e., the frequency domain), of the received signal(s) Srn to form a signal called the 'converted signal' and denoted Sn, to be propagated on the receiving channel 20n, and defined on a given instantaneous frequency band, denoted BF. Such an instantaneous frequency band is associated with a set of frequency channels, denoted F. For example, and without limitation, an instantaneous frequency band BF, defined in an RF band such as 'X-band', can be equal to 256 MHz, and comprise 256 associated frequency channels F, each frequency channel F being defined at 1 MHz. The remainder of the description will refer to the notation 'Sn(F)' as the value of the converted signal defined on a specific frequency channel F.

[0030] Such a 22n conversion unit is implemented using spectral analysis, defined in particular by applying a discrete Fourier transform to the received signal Sr n. A Fourier transform can be implemented by a fast Fourier transform (FFT) algorithm. Fast Fourier Transform (according to the corresponding Anglo-Saxon acronym).

[0031] The measurement module 24n (also called 'determination module' or 'calculation module') is configured to implement operations, in the spectral domain, to determine the power correlation, denoted Rn, between the spectrum of the converted signal Sn propagated on the receiving channel 20n and an arbitrary reference signal, denoted Sref. For each frequency channel F of the given instantaneous frequency band BF, the associated correlation value can be denoted Rn(F).

[0032] The arbitrary reference signal S ref is chosen from the set of N converted signals S n propagated on the receiving channels of the radar system 1. For example and without limitations, the arbitrary reference signal S ref can correspond to the converted signal S 1 to be propagated on a first receiving channel denoted 20 1 . Each of the converted signals from another receiving channel can alternatively be used as the arbitrary reference signal S ref .

[0033] In some embodiments, the input signal of the measuring unit 24n can be the converted signal Sn propagated on the receiving channel 20n, as shown in the figure 2(a) .

[0034] In other embodiments, the input signal of the measuring unit 24n can be a signal calibrated in amplitude and phase, to be propagated on the receiving channel 20n and denoted Scn, as shown in the figure 2(b) .

[0035] The busy frequency channel evaluation module 30 of radar system 1 is configured to determine (or reconstruct) the sum channel Σ of the N receive channels. In particular, the busy frequency channel evaluation module 30 is configured to determine, for each frequency channel F of the given instantaneous frequency band BF, the value of the sum channel, denoted Σ(F), from the correlation values ​​R n (F) defined for each frequency channel F of the set of N receive channels, as defined by the following equation (01): ∑ F = ∑ n R n F

[0036] Module 30 is further configured to evaluate (i.e., detect or determine) whether a specific frequency channel F is occupied by comparing the value of the associated sum channel Σ(F) to a detection threshold. τ thresholdpredefined. The frequency channel F can then be considered "occupied" (or "polluted") if the value of the sum channel Σ(F) is greater than the detection threshold τ threshold . Alternatively, the specific frequency channel F can be considered "free" if the value of the sum channel Σ(F) is less than the detection threshold τ threshold For example, and without limitation, depending on the value of the detection threshold τ threshold predefined, if the value of the sum channel Σ(F) is strictly greater than (or alternatively greater than or equal to) the detection threshold τ threshold , then the frequency channel F can then be considered "occupied"; and if the value of the sum channel Σ(F) is less than or equal to (or alternatively strictly less than) the detection threshold τ threshold, then the frequency channel F can be considered "free".

[0037] Advantageously, module 30 is configured to apply this occupancy assessment of all frequency channels F defined in the given instantaneous frequency band BF.

[0038] It should be noted that a signal S(t) received by the antenna of system 1 is characterized by an arrival direction θ (or DOA for Direction of Arrival according to the corresponding Anglo-Saxon acronym, also called 'direction of origin') defined in relation to the orientation of the antenna, that is to say, the plane of the receivers of assembly 10, as schematically shown on the figure 1 .

[0039] For example, in the case of a one-dimensional (1D) antenna composed of receivers spaced a distance d apart, the receivers in the set 10 are configured to process the time-domain signal S(t) received from the arrival direction θ. The path difference xk of the received signal between a first receiver and a kth receiver can be defined according to the following equation (02): x k = k . d . sin θ

[0040] The index 'k' represents an integer between 2 and N. The phase difference Δϕ between the first and k-th receptors considered, and therefore the associated reception pathways, can thus be expressed according to the following equation (03): Δϕ = 2 π λ × x k = k . Δφ

[0041] In equation (03), the coefficient Δφ corresponds to the phase difference coefficient equal to the following equation (04): Δφ = 2 π λ . d . sin θ

[0042] The signal received by radar system 1, after demodulation and then Fourier transform, i.e. converted into the frequency domain, can be represented for each frequency channel according to the following matrix expression (05), for each reception channel 20n: S 1 S 2 … S N = A 1 . e iφ 0 A 2 . e i Δφ + φ 0 … A N . e i N − 1 × Δφ + φ 0

[0043] In formula (05), the values ​​A n correspond to the level of the received signal S(t) multiplied by the antenna gain, denoted G n, predefined for each of the reception channels 20 n of the radar system 1. Using the converted signal S 1 to be propagated on the first reception channel 20 1 as an arbitrary reference signal S ref , the set R relating to the power correlations R n of the reception channels 20 n can be represented according to the following relation (06): R = R 1 R 2 … R N = E S 1 ∗ S 1 ∗ E S 2 ∗ S 1 ∗ … E S N ∗ S 1 ∗ = A 1 2 A 1 . A 2 . e iΔφ … A 1 . A N . e i N − 1 × Δφ

[0044] Thus, for each of the reception channels 20n, the power correlation Rn can be defined, with respect to the arbitrary reference signal chosen (i.e. for example to the single reference channel 201), up to a multiplicative constant, according to the amplitude An and the phase difference coefficient Δφ of the received signal, as defined according to the following expression (07): R n = A 1 . A n . e i n − 1 × Δφ

[0045] Advantageously, as shown on the figures 2(a) and 2(b) The measurement module 24n of a receive channel 20n may include a first unit 24-1n configured to perform an operation consisting of performing a correlation product between the converted signal Sn and the arbitrary reference signal Sref. Such an operation, also called a 'product operation', can correspond to a power spectral density of the signal propagated on the receive channel 20n with respect to an arbitrary reference signal Sref, and can be represented by the expression " S n ∗ S 1 ∗ "presented in the preceding expression (06). The measurement module 24n of a receive channel 20n may further include a second unit 24-3n configured to apply a summation operation to said correlation product from the first unit 24-1n to generate the power correlation Rn. Such a summation operation may correspond to a post-integration (or integration) operation over several successive frames.

[0046] In embodiments, for each receiving channel 20n, the signal calibrated in amplitude and phase Scn can be used as input to the measuring unit 24n, as shown in the figure 2(b), can be generated from the correction of the converted signal S n propagated on the receiving channel 20 n with respect to a correction coefficient, denoted C n (also called the correction signal). In particular, the radar system 1 may include a beamforming system calibration device 1C configured to generate such a correction signal C n for each calibration channel 20 n.

[0047] As depicted on the figure 3 , the calibration device 1C may include a number N of calibration channels 40 n , each channel being capable of receiving one or more received signals Sr n from one of the antennas of the assembly 10 (or from one of the radar antenna receivers), a noise signal generation unit 50 and a noise processing channel 60.

[0048] Such a calibration device 1C makes it possible to calibrate the radar system 1, on all frequency channels of a given instantaneous frequency band BF and / or on a set of instantaneous frequency bands associated with a specific RF band, by equalizing in amplitude and phase the different converted signal(s) S n to be propagated on the receiving channels.

[0049] Unit 50 can thus be configured to generate a noise signal, denoted Sr 0, broadband, i.e. covering for example the entire instantaneous LF frequency band of the receiving radar antennas of set 10.

[0050] As depicted on the figure 3 , each calibration channel 40 n of the calibration device 1C can include a correction conversion unit 42 n and a correction measurement module 44 n , according to embodiments of the invention.

[0051] Similarly, as shown on the figure 3, the noise treatment channel 60 of the calibration device 1C may include a noise correction conversion unit 62 and a noise correction measurement module 64, according to embodiments of the invention.

[0052] The correction conversion unit 42n of each calibration channel 40n can correspond (i.e. be assimilated and / or be equivalent) to a conversion unit 22n of the receiving channel 20n of the radar system 1. The unit 42n can therefore be configured to apply a conversion (or a pass), in the spectral domain, of the received signals Srn to form the converted signal Sn to be propagated on the receiving channel 40n and defined on a predetermined instantaneous frequency band BF, and associated with the set of frequency channels F.

[0053] Advantageously, the noise correction conversion unit 62 of the noise treatment channel 60 can also correspond to a conversion unit 22n, that is, be configured to apply a conversion (or a pass), in the spectral domain, of the noise signal Sr0 to form the converted noise signal S0 to be propagated and defined on a predetermined instantaneous frequency band BF, and associated with the set of frequency channels F. The rest of the description will refer to the notation "S0(F)" relating to the converted noise signal defined on a specific frequency channel F.

[0054] The noise correction measurement module 64 (also called the 'noise correction determination module' or 'noise correction calculation module') of the noise processing channel 60 can be configured to implement spectral domain operations to determine the noise correction coefficient, denoted SC 0. In particular, this noise correction coefficient SC 0 corresponds to the power cross-correlation of the converted noise signal S 0.

[0055] The correction measurement module 44n (also called the 'correction determination module' or 'correction calculation module') of each calibration channel 40n of the calibration device 1C can be configured to perform spectral domain operations to determine the correction coefficient Cn. In particular, this correction coefficient Cn corresponds to the power correlation between the spectrum of the converted signal Sn to be propagated on the calibration channel 40n and the converted noise signal S0. It should be noted that the complex correction coefficient values, denoted Cn(F), can correspond to the correction estimates of the amplitude and phase deviation of each of the receiving channels 20n of the radar system 1 for each frequency channel F in the spectral domain.Such correction coefficients C n make it possible to balance in amplitude and phase each of the reception channels 20 n over the entire predetermined instantaneous frequency band BF.

[0056] For each receiving channel 20n, the signals calibrated in amplitude and phase Scn defined from the correction coefficients Cn applied to the converted signals Sn can then be used in the signal processing of the receiving antennas of the set 10.

[0057] In embodiments, as shown in the figure 4(a)The noise correction measurement module 64 of the noise processing channel 60 may include a first noise correction unit 64-1 and a second noise correction unit 64-3. For example, and without limitation, the first noise correction unit 64 may be configured to perform an operation consisting of cross-correlation of the converted noise signal S0 (from the noise correction conversion unit 62 of the noise processing channel 60), corresponding, for example, to the product operation defined by the expression " S 0 ∗ S 0 ∗ "). The second noise correction unit 64-3 can be configured to apply a post-integration operation of said cross-correlation product from the first noise correction unit 64-1, to generate the noise correction coefficient SC 0.

[0058] In embodiments, as shown in the figure 4(a)The correction measurement module 44n of a calibration channel 40n may include a first correction unit 44-1n and a second correction unit 44-3n. For example, and without limitation, the first correction unit 44-1n may be configured to perform an operation consisting of a correlation product between the converted signal Sn (from the conversion unit 42n) and the converted noise signal S0 (from the noise correction conversion unit 62 of the noise processing channel 60). The second correction unit 44-3n may be configured to perform a post-integration operation on said correlation product from the first correction unit 44-1n.

[0059] The correction measurement module 44n may further include a third correction unit 44-5n configured to apply an operation consisting of performing a division between the post-integration signal from the second correction unit 44-3n and the noise correction coefficient SC0 (from the noise correction measurement module 64 of the noise processing channel 60), to generate the correction coefficient Cn of the receiving channel 20n.

[0060] Furthermore, in the case where the value of the sum channel Σ(F), defined for a frequency channel F, is greater than the detection threshold τ threshold, that is to say that the frequency channel F is occupied according to a received signal S(t), module 30 of radar system 1 can further be configured to evaluate the direction of arrival θ of the signal S(t) received by the antenna of system 1, with respect to the antenna receivers of assembly 10.

[0061] Advantageously, the radar system 1 can be configured to apply and modify an operational pointing direction of the receivers of the assembly 10. In particular, for a two-dimensional (2D) antenna, for example, the radar system 1 can be initially configured to apply an elevation pointing antenna direction θ E0 and a circular pointing antenna direction θ C0, prior to the acquisition of the signal S(t) by the antenna.

[0062] According to variants of the invention, to evaluate the arrival direction θ of the received signal S(t), the module 30 of the radar system 1 can be adapted to determine whether, for the frequency channel F then occupied, the operational pointing direction of the antenna 10 coincides (in particular coincides approximately) with this arrival direction, and then to estimate (or determine) a refined arrival direction of the received signal S(t).

[0063] For this purpose, the evaluation module 30 can be configured to determine (or reconstruct) the channel gap of the set of N receiving channels from the power correlations R n.

[0064] For a 1D antenna, the evaluation module 30 can, for example, be configured to determine, for each frequency channel F of the given instantaneous frequency band LF, a channel deviation value, denoted Δ(F). For a 2D antenna, the evaluation module 30 can be configured to determine, for each frequency channel F of the given instantaneous frequency band LF, a circular channel deviation value, denoted ΔC(F), as well as an elevation channel deviation value, denoted ΔE(F). Such channel deviation values ​​for a 2D antenna can be defined according to the following equations (08) and (09): ΔC F = ∑ n w c n . R n F ΔE F = ∑ n w E n . R n F

[0065] In equations (08) and (09), the elements wc(n) and wE(n) correspond to the antenna coefficients in circularity and elevation, respectively, relating to the receiving channel 20n and the nth antenna receiver associated with the radar system 1. For example, and without limitation, the figure 5 represents a schematic illustration of a 2D, 4-way receiving antenna. In this case, such antenna coefficients can be expressed according to the following equations (10) and (11): w c = + i , − i , + i , − i w E = + i , + i , − i , − i

[0066] The evaluation module 30 can also be configured to determine the ratio of the gap channel over the set of N receiving channels, from the gap channel and the sum channel Σ(F).

[0067] In particular, the evaluation module 30 can also be configured to determine, for each frequency channel F of the given instantaneous frequency band BF, one or more ratio values ​​from the value or values ​​of the channel gap and the value of the channel sum Σ(F) defined each on the set of N receive channels.

[0068] To facilitate understanding of the invention and for the sake of simplicity, the following description will be given primarily with reference to a 2D antenna. Those skilled in the art will readily understand that the invention can be applied to any type of antenna, including a 1D antenna. Thus, in the specific case of a 2D antenna, the evaluation module 30 can be configured to determine, for each frequency channel F of the given instantaneous frequency band BF, a value of the circular deviation channel ratio C(F), as well as a value of the elevation deviation channel ratio E(F), as defined by the following equations (12) and (13): ratio C F = ΔC F Σ F ratio E F = ΔE F Σ F

[0069] The evaluation module 30 can then be adapted to determine whether the elevation and circular pointing direction (θ E0 , θ C0 ) of the antenna 10, for each frequency channel F of the given instantaneous frequency band BF, corresponds to the arrival direction of the received signal S(t), from the comparison of the value(s) of the channel deviation to a predefined threshold value, called 'lobe threshold' and denoted ε lobe , and in particular, from the comparison of the value of the lane deviation ratio in circularity ratio C (F) and the value of the lane deviation ratio in elevation ratio E (F) at the predefined lobe threshold ε lobe .

[0070] Advantageously, the elevation and circular pointing direction (θ E0 , θ C0 ) of the antenna 10 corresponds to the arrival direction of the received signal S(t), if the absolute value of the circular deviation channel ratio |ratio C (F)|, as well as the absolute value of the elevation deviation channel ratio |ratio E (F)| are each less than the predefined lobe threshold ε lobe Alternatively, the elevation and circular pointing direction (θ E0 , θ C0 ) of antenna 10 does not correspond to the arrival direction of the received signal S(t), if the absolute value of the circular deviation channel ratio |ratio C (F)| and / or the absolute value of the elevation deviation channel ratio |ratio E (F)| is greater than the predefined lobe threshold ε lobe For example, depending on the lobe threshold value ε lobepredefined, if the absolute value of the circular deviation lane ratio |ratio C (F)| and / or the absolute value of the elevation deviation lane ratio |ratio E (F)| are each strictly less than (or alternatively less than or equal to) the lobe threshold ε lobe , then the elevation and circular pointing direction can correspond to the arrival direction of the received signal S(t); and if the absolute value of the circular deviation channel ratio |ratio C (F)| and / or the absolute value of the elevation deviation channel ratio |ratio E (F)| is greater than or equal to (or alternatively strictly greater than) the lobe threshold ε lobe , then the direction of pointing in elevation and in circle can correspond to the direction of arrival of the received signal S(t).

[0071] In the case where the absolute values ​​of the lane ratio, circular deviation |ratio C (F)| and elevation deviation |ratio E (F)| are less than the lobe threshold ε lobe ,That is, if the elevation and circular pointing direction (θE0, θC0) of antenna 10 corresponds to the arrival direction of the received signal S(t), module 30 can be adapted to evaluate (i.e., estimate or determine) a refined elevation and circular pointing direction (θEestim, θCestim) of arrival of the received signal S(t). Such a refined direction can be estimated from the value of the circular deviation channel ratio C(F), the value of the elevation deviation channel ratio E(F), and the elevation and circular pointing direction (θE0, θC0) of antenna 10.

[0072] Thus, in some embodiments, the evaluation module 30 can be adapted to determine the value of the refined elevation direction θ Eestim according to the following equation (14): θ E estim = asin sin θ E 0 + ∑ i = 0 J a i × ratio E F i

[0073] Furthermore, the evaluation module 30 can be adapted to determine the value of the refined circular direction θ Cestim as a function of the value of the refined elevation direction θ Eestim according to the following equation (15): θ C estim = asin 1 cos θ E estim × cos θ E 0 . sin θ C 0 + ∑ i = 0 J a i × ratio C F i

[0074] In the preceding mathematical equations (14) and (15), the elements ai are polynomial coefficients depending on predefined characteristics of antenna 10, and the integer J corresponds to an order of the chosen polynomial.

[0075] It should be noted that the frequency band associated with the frequency channel F can advantageously correspond to the instantaneous frequency band centered on the predefined carrier frequency of the radar antenna, for example, characterized by a particular pointing direction. Advantageously, the radar system 1 can be configured to apply the evaluation of the frequency channel F and / or the estimation of the direction of arrival of the received signal S(t) for a plurality of instantaneous low-frequency bands, including several carrier frequencies of the radar within its agility band. This allows the radar system 1 to perform a frequency scan for signal detection by the antenna to cover an entire frequency range of predefined interest (for example, an entire frequency range associated with the RF band of the "X-band" type).

[0076] When estimating the arrival direction of the signal S(t) received by the antenna of system 1, if the elevation and circular pointing direction (θ E0, θ C0) of the antenna 10 does not correspond to this arrival direction, for example, the radar system 1 can be configured to apply a new antenna elevation and circular pointing direction before acquiring a new signal S(t) with the antenna and then performing another estimation of the signal's arrival direction. This allows the radar system 1 to perform a spatial scan of the signal detected by the antenna to cover a predefined angular domain of interest (antenna pointing direction).

[0077] In some embodiments, the evaluation module 30 can be configured to roughly estimate the arrival direction θ of the signal S(t) received by a 1D antenna, for example, from the measurement of the phase difference coefficient Δφ of the received signal, the wavelength values ​​λ of the signal S(t), and the spacing d of the known receivers. According to equation (04) defining this coefficient, the arrival direction θ can be deduced from the following expression (16): sin θ = Δφ . λ 2 π . d

[0078] It should be noted that since the phase difference coefficient Δφ is defined between -π and π, it can introduce ambiguity in the estimation of the arrival direction θ of the received signal S(t). In this case, such a rough estimate can be made for an antenna 10 comprising very close receivers (i.e., with a maximum spacing of half a wavelength of the signal) or by digitizing the channels at the radiating element in the case of an array antenna.

[0079] There figure 6 represents the method of evaluating a frequency channel of the received signal S(t), according to embodiments of the invention.

[0080] At step 600, each receiving channel 20 n receives signals Sr n from one of the antennas of the assembly 10 (or from one of the radar antenna receivers) and from a signal S(t).

[0081] At step 620, on each receiving channel 20n, the received signals Srn are converted into the spectral domain, to form the converted frequency signal Sn to be propagated on the channel 20n.

[0082] At step 640, for each receiving channel 20n, for each frequency channel F of a given instantaneous frequency band BF, the correlation values ​​Rn(F) are determined by applying a power correlation between the spectrum of the propagated signal Sn on the channel 20n and an arbitrary reference signal Sref.

[0083] At step 660, for each frequency channel F of the given instantaneous frequency band BF, the values ​​of the sum channel Σ(F) are determined from the correlation values ​​R n (F) defined for the set of N receiving channels as defined according to the preceding equation (01).

[0084] In step 680, for each frequency channel F of the given instantaneous frequency band BF, the values ​​of the sum channel Σ(F) are compared to a detection threshold value τ threshold predefined.

[0085] At step 682, if a value of the sum channel Σ(F) is greater than the detection threshold τ threshold, The associated frequency channel F is then considered "occupied". Alternatively, at step 684, if a value of the sum channel Σ(F) is less than the detection threshold τ threshold The associated frequency channel F is then considered "free".

[0086] Step 620 and / or 640 can subsequently be repeated to determine new correlation values ​​R n (F) according to another given instantaneous frequency band BF to perform a frequency sweep, on a predetermined RF band for example.

[0087] There figure 7 represents the method of estimating the direction of arrival of the received signal S(t), according to embodiments of the invention.

[0088] The process may include an initial step 700 of applying the pointing antenna direction (in elevation θ E0 and in circular θ C0).

[0089] At step 720, for each frequency channel F of the given instantaneous frequency band BF, the values ​​of the channel deviation (circular ΔC(F) and elevation ΔE(F)) can be determined from the correlation values ​​R n (F) of the set of N receiving channels as defined for example according to the preceding equations (08) and (09).

[0090] At step 740, for each frequency channel F of the given instantaneous frequency band BF, the values ​​of the channel deviation ratio (circular ratio C(F) and elevation ratio E(F)) can be determined from the values ​​of the sum channel Σ(F) and the values ​​of the channel deviation (circular ΔC(F) and elevation ΔE(F) respectively) as defined for example according to the preceding equations (12) and (13).

[0091] At step 760, for each frequency channel F of the given instantaneous frequency band BF, the values ​​of the channel gap ratio (circular ratio C(F) and elevation ratio E(F)) can be compared to a lobe threshold value ε lobe predefined.

[0092] At step 762, if the absolute values ​​of the lane deviation ratio (circular ratio C (F) and elevation ratio E (F)) are less than the lobe threshold ε lobe ,The antenna's pointing direction (elevation θE0 and circular θC0) can correspond to the arrival direction of the received signal S(t), and the values ​​of the refined direction (elevation θEestim and circular θCestim, respectively) can be determined from the absolute values ​​of the channel-gap ratio and the pointing direction as defined, for example, by the preceding equations (14) and (15). Alternatively, in step 764, if at least one of the absolute values ​​of the channel-gap ratio (circular ratio C(F) and elevation ratio E(F)) is greater than the lobe threshold ε lobe , The antenna pointing direction may not correspond to the arrival direction of the received signal S(t). The initial step 700 can then be repeated to apply a new antenna pointing direction.

[0093] A person skilled in the art will readily understand that certain steps in the processes relating to figures 6 And 7can be carried out simultaneously, sequentially, independently or not, and / or in a different order, for example in an order defined by radar system 1 or evaluation module 30.

[0094] The system and methods described above, according to embodiments of the invention or sub-elements thereof, can be implemented in various ways by hardware, software, or a combination of hardware and software, including in the form of program code that can be distributed as a program product in various forms. In particular, the methods of figures 6 And 7 can, for example, be implemented in a computer system.

[0095] The invention is not limited to the embodiments described above by way of non-limiting example. It encompasses all possible embodiments that could be envisioned by a person skilled in the art. In particular, a person skilled in the art will understand that the invention is not limited to the different units or configurations of the receiving channels described by way of non-limiting example. In particular, certain embodiments of the invention can be combined.

Claims

1. Radar system (1) for detecting occupied frequency channels, the system comprising N receiving paths (20n) and a set (10) of radar antennas associated with each path, N being an integer greater than or equal to 2, each path being capable of receiving respectively signals (Srn) by means of the associated radar antenna, the radar system (1) comprising, for each receiving path (20n), a conversion unit (22n) in the spectral domain of said received signals (Srn) to form a converted signal (Sn) defined on a given instantaneous frequency band associated with a set of frequency channels (F), and a module (24n) for determining correlation values (Rn( F)) characterised in that it is adapted to apply a power correlation between the spectrum of the converted signal (Sn) propagated over said receiving path (20n) and the spectrum of an arbitrary reference signal (Sref ) chosen from among one of the converted signals propagated over one of said receiving paths, the radar system (1) further comprising a module (30) for evaluating occupied frequency channels adapted to determine the values of the sum path (Σ(F)) from the correlation values (Rn(F)) of said N receiving paths, the frequency channel (F) being occupied if the value of the associated sum path (Σ(F)) is greater than a predefined detection threshold value (τthreshold ).

2. Radar system (1) according to claim 1, wherein said determination module (24n) is configured to determine a power spectral density of the signal propagated over said receiving path (20n) with respect to said arbitrary reference signal (Sref).

3. Radar system (1) according to any one of the preceding claims, wherein said determination module (24n) is configured to apply an integration on several successive frames of the signal propagated over said receiving path (20n).

4. Radar system (1) according to any one of the preceding claims, wherein the radar system (1) further comprises a calibration device (1C) comprising N calibration paths (40n), and wherein, for each calibration path (40n), the calibration device (1C) is configured to determine correction coefficient values (Cn(F)) of said converted signal (Sn) to balance, in amplitude, and in phase, each of said receiving paths (20n) of the radar system (1).

5. Radar system (1) according to any one of the preceding claims, wherein said radar system (1) is configured to apply a frequency scan by repeating the evaluation of the detection of occupied frequency channels with respect to said detection threshold value (τthreshold ), to totally cover a predefined interest frequency domain.

6. Radar system (1) according to any one of the preceding claims, wherein the radar system (1) is configured to apply a pointing antenna direction (θE0, θC0), and wherein said module for evaluating occupied frequency channels (30) is further adapted to determine the values of the deviation path (ΔC(F),ΔE(F)) from the correlation values (Rn(F)) of said N receiving paths, then the values of the ratio of the deviation path (ratioC(F), ratioE(F)) from said values of the sum path (Σ(F) ) and of said values of the deviation path (ΔC(F),ΔE(F)), said pointing antenna direction (θE0, θC0) corresponding to the arrival direction of the received signals (Srn) with respect to the orientation of said antennas of the radar system (1) if said values of the ratio of the deviation path (ratioC(F), ratioE(F)) are less than a predefined lobe threshold value (εlobe ).

7. Radar system (1) according to claim 6, wherein said module for evaluating occupied frequency channels (30) is further adapted to determine at least one refined arrival direction (θEestim , θCestim ) of the received signals with respect to the orientation of said antennas of the radar system (1) from said values of the ratio of the deviation path (ratioC(F), ratioE(F)) and of said pointing antenna direction (θE0,θC0)·8. Radar system (1) according to any one of claims 6 or 7, wherein said radar system (1) is configured to apply a spatial scan by repeating the evaluation of the arrival direction of the received signals with respect to said lobe threshold value εlobe, to totally cover a predefined interest angular domain.

9. Radar system (1) according to any one of the preceding claims, wherein the radar system (1) comprises an operationally "listening" configuration in the "band X"-type radiofrequency frequency domain.

10. Method for detecting occupied frequency channels, implemented in a radar system (1) comprising N receiving paths (20n), N being an integer greater than or equal to 2, the method comprising the receiving, at each path, of signals (Srn), characterized in that said method comprises, for each receiving path (20n) the following steps consisting of: - converting (620), into the spectral domain, the received signals (Srn) to form N converted signals (Sn), each defined over a given instantaneous frequency band, associated with a set of frequency channels (F), - determining (640) correlation values (Rn(F)) by applying a power correlation between the spectrum of the converted signal (Sn) propagated over said receiving path (20n) and the spectrum of an arbitrary reference signal (Sref ) chosen from among one of the converted signals propagated over one of said receiving paths, the method further comprising an evaluation of the occupied frequency channels comprising the steps consisting of: - determining (660) the values of the sum path (Σ(F)) from the correlation values (Rn(F)) of said N receiving paths, - comparing (680) the value of the sum path (Σ(F)) associated with a predefined detection threshold value (τthreshold ), and - determining that the frequency channel (F) is occupied (682) if the value of the associated sum path (Σ(F)) is greater than said predefined detection threshold value (τthreshold).