Patent Grant
8816897
This application claims priority to foreign French patent application No. FR 1004559, filed on Nov. 23, 2010, the disclosure of which is incorporated by reference in its entirety.
The present invention relates to the field of radars and in particular radars for searching mobile targets which are themselves equipping a mobile carrier. The invention applies advantageously to airborne maritime surveillance radars.
One objective of a surveillance radar is to detect and locate a target, mobile or not. The determination of the position of a target is done notably by estimating its angular position which is defined by the angle between a reference axis and the axis linking the radar, or its carrier, to the target. For example, when the reference axis points towards the geographic north, the angular position of the target is defined by the azimuth angle. When the reference axis is the axis of the carrier, the angular position of the target is defined by the relative bearing angle.
The general principle of a surveillance radar consists in emitting a series of pulses via a steerable antenna which scans the angular space to detect the presence of a potential target. If a target is actually present in the beam of the antenna, the signal pulses are reflected on this target and return echoes to the radar which includes suitable reception means. Means for processing and analysing the received echoes are implemented in order to deduce therefrom information concerning the location of the target, notably its angular position, its distance relative to the radar or even its speed.
Three methods for analysing the received signal to determine the angular position of a target using a radar are notably known.
A first known method uses a single-channel radar which performs an analysis of the amplitude (or of the power) of the received signal during the scanning of the antenna. The amplitude of the signal received by the radar varies according to the azimuth of the antenna following the appearance of the antenna pattern. The azimuth of the detected target is obtained by searching for the maximum amplitude. This principle is illustrated in
The efficiency of this method, and in particular the accuracy of the estimation of the azimuth of the target, is greatly dependent on the signal-to-noise ratio in reception. The lower this ratio, the more difficult it is to detect the target because the noise can generate spurious amplitude spikes.
Another known method consists in performing an analysis of the received signals in the frequency domain. This type of method is implemented by single-channel radars said to be working in “Doppler mode”. In such cases, the radar emits a series of pulses at different successive instants. The echoes received after reflection on the target are grouped together in blocks, a block comprising several pulses. For each pulse received, the azimuth, or the relative bearing, of the antenna is known to the radar, so an average azimuth is then defined for a block of pulses. By way of an analysis in the frequency domain of the pulses of a block, notably through a Fourier transform applied to this block, it is possible to detect the presence of the target by the level of the amplitude of the echoes received by the radar. Each block has an associated average azimuth of the antenna pointing direction. The angular locating of the target then consists in assigning, as estimate of the azimuth of the target, the average azimuth of the block for which a target is detected. The target may be detected on several consecutive blocks, a direction finding method can then be used in order to average the azimuth of the blocks by weighting them by the amplitude of the received signal.
This type of method has the drawback of limiting the azimuthal detection resolution to the angular resolution separating the average azimuth of two blocks and therefore indirectly to the size of a block. This drawback affects the efficiency of the detection with respect to the accuracy of the estimate of the angular position of the target. It is possible to remedy this drawback by implementing an azimuthal overlap between two or more consecutive blocks, but, the greater the rate of overlap, the more complex the calculation becomes, which affects the implementation cost.
A third known method consists in using a radar that has at least two reception channels. These two channels are commonly designated by the terms sum channel and delta channel, and the azimuth of the target is obtained by calculating the angle error measurement, for example taken to be equal to the scalar product of the signals obtained from each of the two channels. The angle error measurement value makes it possible to deduce therefrom the difference that there is between the angular position of the antenna and the real position of the target.
This third method presents the advantage of considerably increasing the angular location accuracy compared to the abovementioned second method, but it entails the use of a radar with two reception channels, a factor that increases the cost of the equipment.
The invention aims to remedy the abovementioned limitations of the known methods by proposing a method for determining the angular position of a target which substantially enhances the accuracy and in particular the variance of the estimations made. The invention requires the use of only one reception channel and also makes it possible to determine the radial speed of the target.
The main advantages of the invention are that it enhances the angular extraction accuracy and reduces fluctuations on the extracted azimuth.
To this end, the subject of the invention is a method for estimating the angular position θc of a target detected by a radar equipping a mobile carrier, said radar emitting, via an steerable antenna, a signal, in the form of pulses, towards the target and receiving echoes from the reflection of said signal on the target, said method being characterized in that it comprises at least the following steps:
in which λ is the wavelength of the radar, Va is the norm of the speed of the carrier and Vr,c is the radial speed of the target.
In a particular embodiment of the invention, the equation
is solved by minimizing the mean square error
in which N is the number of pulses or of groups of pulses, the angular position θc and the radial speed Vr,c of the target being assumed constant over the horizon N.
In another particular embodiment of the invention, the radial speed of the target Vr,c is set to an a priori given value Vr,capriori and that the angular position θc(i) of the target is taken to be equal to
for each pulse or group of pulses of time index i.
In a variant embodiment of the invention, the angular position θc(i) of the target is corrected by a measurement bias
in which ΔVr,c is the offset between the a priori value Vr,capriori of the radial speed of the target and its real value.
In a variant embodiment of the invention, the measurement bias δθ is determined from the difference between the average values of the angular positions of the antenna and the average values of the angular positions of the target.
In a variant embodiment of the invention, the measurement bias δθ is determined from a value of the radial speed of the target supplied by target tracking means.
In a variant embodiment of the invention, the angular positions are relative bearing or azimuth angles.
In a variant embodiment of the invention, the carrier is an aircraft.
In a variant embodiment of the invention, the radial speed of the carrier is estimated by an inertial unit equipping the carrier.
Also the subject of the invention is a surveillance radar comprising at least one steerable antenna, emission means for emitting a radar signal in the form of pulses, reception means for receiving echoes generated by the reflection of said pulses on the environment, means for estimating the angular position of the antenna, means for estimating the Doppler frequency of the received echo or echoes, means for pairing the angular position of the antenna and the Doppler frequency, said radar also being adapted to implement the steps of the method according to the invention.
Other features and advantages of the invention will become apparent from reading the description, given as an example, given in light of the appended drawings in which:
Each of the three diagrams of
The aircraft 300 has a speed {right arrow over (V)}a, of norm Va, the direction of which may differ from the axis 302 of displacement of the aircraft. The antenna of the radar (not represented) equipping the aircraft has, at a given instant, a pointing direction 303 which forms an angular deviation Δθ with the axis 304 linking the aircraft 300 and the target 301. The target 301 is mobile and itself has a speed {right arrow over (V)}c of norm Vc. Vr,a and Vr,c are used to designate the respective radial speeds of the aircraft and of the target which correspond to their projections on the line of sight of the radar.
The relative radial speed Vr of the wave received and measured by the radar in the direction of pointing of the antenna is linked to the Doppler frequency fD by the following relationship:
Vr=λfD/2 (1)
with λ being the wavelength of the signal emitted by the radar. The radial speed Vr corresponds to the sum of three terms, the radial speed Vr,a of the aircraft, the radial speed Vr,c of the target and a component ΔVr which results from the angular deviation Δθ between the direction of pointing of the antenna and the direction of the target. The total radial speed Vr is therefore expressed using the following relationship:
Vr=Vr,a+ΔVr+Vr,c (2)
The radial speed Vr,a of the aircraft can be estimated by means external to the invention, for example by an inertial unit equipping board the craft. It can thus be compensated by suitable processing operations such that the radial speed measured by the radar becomes:
Vr=ΔVr+Vr,c (3)
The component ΔVr is expressed as a function of the angular deviation Δθ and of the speed of the carrier by the following relationship, by making the likely assumption that the angular deviation Δθ is low:
ΔVr=Va sin θΔθ (4)
ΔVr, which translates the radial speed offset induced by the angular deviation between the line of sight and the direction of the target, is obtained by deriving the expression of the radial speed Vr,a=Va cos θ of the carrier relative to the relative bearing angle.
By combining the relationships (1), (3) and (4), the following equation is obtained:
The equation (5) is an equation with two unknowns, the angular deviation Δθ, which indirectly gives the angular position of the target in terms of azimuth or in terms of relative bearing and the radial speed Vr,c of the target. The wavelength λ of the radar and the speed of the aircraft are data that are a priori known. The angle θ, which corresponds to the relative bearing of the antenna of the radar, is also known and the Doppler frequency fD of the echoes received by the radar is measured for each value of the angle θ during the scanning of the antenna.
In a first particular embodiment of the invention, if the target to be detected is fixed, its radial speed Vr,c and the angular deviation Δθ can be estimated from the following relationship
for each pair of values (fD, θ) obtained from a detection determined by the radar. The precise relative bearing of the target is then obtained by adding the relative bearing angle θ of the antenna to the angular deviation Δθ. The relative bearing values of the target obtained can be averaged in order to deduce a precise angular position of the target therefrom.
In another embodiment of the invention, for which the radial speed of the target is not zero, the equation (5) then includes two unknowns. If a sufficient number of measurements of the Doppler frequency performed by the radar is available in a sufficiently short time interval, the radial speed of the target is assumed constant over that interval. In such a case, an optimization algorithm is used to determine the values of the unknowns Δθ and Vr,c from the measurements of the Doppler frequency for different relative bearing angles θ of the radar antenna.
The method according to the invention then consists, for example, in determining the pair of values ({circumflex over (θ)}c, {circumflex over (V)}r,c) which minimizes the mean-square error between the two members of the equation (5). The mean square error is calculated as follows:
with N being the number of measurements performed on the one hand on the radial speed Vr(i) and on the other hand on the relative bearing of the radar antenna θ(i). The radial speed Vr(i) is measured from the Doppler frequency fD(i) of the received echo,
θc=θ(i)+Δθ is the relative bearing angle of the target that is to be determined.
It will be noted that, in addition to the angular position of the target, the method according to the invention also makes it possible to determine an estimate of the radial speed of the target. Also, the method is applicable only when the pointing axis 303 of the antenna is not collinear to the axis 302 of the carrier because, otherwise, the angle θ becomes zero and the minimizing of the squared error according to the relationship (6) is no longer possible.
The point 400 of coordinates {circumflex over (θ)}c=−27.1° and {circumflex over (V)}r,c=−0.8 m/s is the point which minimizes the mean square error function.
Without departing from the framework of the invention, other methods for solving the equation (5) can be envisaged, for example least squares methods.
In some cases, the radial speed of the target cannot be considered constant over the entire duration of the measurements performed by the radar. Such is the case, for example, when the measurements of the radial speed Vr are too spaced apart in time or the target is subject to significant fluctuations in its speed. In these conditions, the method according to the embodiment of the invention described above cannot be applied because it requires a constant radial speed throughout the duration of the measurements performed.
In another variant embodiment of the invention, the relative bearing angle of the target is determined by setting, a priori, the value of the radial speed of the target at a given value, for example zero. The equation (5) then has only a single unknown and the relative bearing angle is estimated using the following relationship
for each measurement i varying from 1 to N.
The value of the relative bearing angle θc is affected by a bias
in which ΔVr,c=Vr,c−Vr,capriori is the offset between the value given, a priori, to the radial speed of the target and its real value. However, one advantage of the method according to the invention lies in the angular correction Δθ which makes it possible to reduce the variance on the measurements of the relative bearing angle θc.
The bias δθ(i) resulting from not knowing the radial speed of the target can be compensated by calculating the difference between the average values of the measurements of the relative bearing of the antenna θ(i) and the measurements of the relative bearing of the target θc(i), and by adding said difference to the measurements of the relative bearing angle θc(i) obtained, the average preferably being calculated over a short time horizon. The abovementioned variant embodiment of the invention is applicable in the case where the radial speed of the target is not constant over the time horizon of the measurements performed by the radar. It also includes a benefit in the case where it is desirable to limit the complexity of the processing operations and where the desired accuracy relates only to the variance of the estimate θc and not to its absolute value.
The radar 600 according to the invention also comprises a module 611 for estimating the relative bearing, or more generally the angular position, of the target or targets detected using the method according to the invention as described previously in its different embodiments. The module 611 produces as output an estimate 612 θc of the angular position of the target, possibly accompanied by the estimated measurement bias when the radial speed of the target is set to an a priori value.
In a particular embodiment of the invention, these estimations 612 are supplied to a tracking module 613 which uses the estimated angular position of the target, at different successive instants, in order to track the displacement of this target. A tracking module 613 uses, for example, a Kalman filtering fed by the angular position estimates in order to perform the tracking of the displacement of the target.
In another embodiment of the invention, the tracking module 613 can provide an estimate 614 of the radial speed of the target which can be used to correct the measurement bias δθ.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10 04559 | Nov 2010 | FR | national |
| Number | Name | Date | Kind |
|---|---|---|---|
| 3713154 | Kummer | Jan 1973 | A |
| 4034373 | de Pierre et al. | Jul 1977 | A |
| 6204804 | Andersson | Mar 2001 | B1 |
| Entry |
|---|
| J. C. Curlander and R. N. McDonough: “Synthetic Aperture Radar—Systems and Signal Processing (passage)”, Jan. 1, 1991, John Wiley and Sons, New York, NY, USA, XP007921734, ISBN: 0-471-85770-X pp. 16-21, p. 17, line 4-line 16. |
| Number | Date | Country | |
|---|---|---|---|
| 20120127025 A1 | May 2012 | US |
