Method of and device for displaying moving targets, in particular in a variable PRF radar.

The object of the present invention is to display moving targets in radar equipment with a variable repetition period.

By visualization of moving targets is meant making visible, without significant holes, in the range of useful radial velocities of the targets, the mobile echoes belonging to a certain Doppler frequency band, by eliminating as much as possible, the fluctuating fixed echoes and their fluctuations caused by the possible rotation of the antenna, as well as parasitic echoes at low speed, the latter, better known by the English term "clutter".

To obtain this better visualization, the frequency of recurrence, that is to say the duration of the repetition period, of the Doppler radar considered is varied by hops.

We know that in a Doppler radar, certain speeds are said to be blind for certain values ​​of the repetition period. By varying the value of this repetition period from one recurrence to the next, certain objectives, which were invisible, reappear, while the fixed echoes themselves remain eliminated regardless of this value of the repetition period. However, this stepwise variation of the repetition period also has the effect of significantly reducing the attenuation rate of the low Doppler frequencies, which are mainly close to zero.

According to a prior art described by the Applicant, compensation is made for the spreading of the spectrum of fixed echoes caused by the variation of the repetition frequency in an electromagnetic detection equipment, by forming the useful signal coming from a target from the differences of the signals received successively from this target and by carrying out a weighted sum of these difference signals with weighting coefficients chosen so as to reduce these difference signals to the values ​​they would have in the absence precisely of this variation of the frequency of repetition.

In a corresponding embodiment, the signals from two coherent detectors and phase shifted by π / 2 relative to each other are preferably sampled in digital. In each of the channels considered, the so-called sine channel and the so-called cosine channel, two samples are taken, by echo, which are stored in memory so as to be able to simultaneously dispose of the samples corresponding to three successive recurrences.

The difference signals of the three signals thus defined are then applied to a weighting device which comprises two multipliers with variable coefficients, these coefficients being a function of the recurrence considered. The signals leaving these weighting devices are added and then processed in a known manner in a filtering and detection device.

It is noted, according to the method and the embodiment which have just been briefly described and which relate to the prior art, that a weighting is carried out on two difference signals, one of which must be delayed with respect to the other so that the signals corresponding to the same distance delivered after weighting can be added. This requires the presence in each channel of an additional memory which increases the cost of the equipment. In addition, it was found that the desired rejection of low frequency echoes was not very satisfactory.

According to the invention, the weighting operation is simplified by carrying out only one directly on the difference signal which is created at a given instant.

According to the invention, consequently, a storage of a signal is eliminated for the duration of a recurrence and the weighting operation which would take place on the delayed signal is eliminated. This solution, in addition to simplifying the process and lightening its material implementation, provides better or at least as good results as those obtained according to the prior art.

According to the invention, the weighting operation is carried out with a real and / or complex coefficient, the weighting circuit delivering on its two output channels compensation signals obtained by linear combinations of the difference signals delivered at the input of the circuit by the sine and cosine channels.

The invention will be better understood in the description which follows of an exemplary embodiment given with the aid of the figures which represent:<ul id="ul0001" list-style="none"><li>FIG. 1, a schematic diagram of the signal processing device according to the invention, and</li><li>- Figure 2, a schematic diagram of the weighting device.</li></ul>

It has been seen in the introduction to the present description that to make mobile targets visible in Doppler radar equipment with variable periods of recurrence, took into account the useful video signal as formed by a combination of the differences of the sampled signals with successive recurrences taken two by two and that a weighted sum of these differences was made with weighting coefficients varying in time as a function of the variation in recurrence periods.

To simplify the operations and reduce the necessary circuits, according to the invention, the weighted summation of the differences is no longer carried out, but only a single weighting is carried out on a single difference signal obtained over two successive repetition periods.

The weighting coefficient chosen, which varies according to the values ​​of the last repetition periods, can be either real or complex; in the case where the coefficient is complex, this amounts to operating on the vector representing the video signal considered, a homothety followed by a rotation so that the vector obtained at the output of the device implementing the invention is equipollent to the vector " ideal "which would be received at the input of the device for a frequency for which compensation is sought, in other words that the" corrected "vector would have the same amplitude and the same phase as this ideal vector. In the case where the weighting coefficient is real, the vector considered would undergo only one homothety.

The device in which the weighting takes place is advantageously placed between two filtering devices canceling the zero frequency. The elimination of the zero Doppler frequency being carried out in the first of these filtering devices is therefore not disturbed by any imperfections in the weighting. In addition, it will be noted that the compensation remains effective, for frequencies close to that chosen.

According to the invention also, the weighting device can be followed by a filtering device of any structure having a rejection zone which contains the frequency f on which the compensation system is adjusted.

When the choice has been made of the frequency f for which the compensation must be perfect, the calculation of the weighting coefficients is as follows.

We consider the signals U_i-1 and you_i which are applied to the filtering device located before the weighting device and which are therefore those received at successive recurrences, in this case at recurrences i-1 and i, and the filtering device of which will deliver the difference. These signals can be written:<maths id="math0001"><img file="EP0010481A1_D0001.tif" /></maths><maths id="math0002"><img file="EP0010481A1_D0002.tif" /></maths>equalities in which A is a constant and t_i = t ^{+ T}₁^{+ T}₂⁺ .., T_i, the symbols T corresponding to the repetition periods.

The difference between the two previous signals is written:<maths id="math0003"><img file="EP0010481A1_D0003.tif" /></maths>This latter equality represents the amplitude of the difference signal delivered by the first filtering device if it is observed that it is modulated by the time sin πfT_i. If we suppose that in the absence of variation of the recurrence period, this is T_m therefore constant, there is no more modulation.

According to the invention if the difference vector is multiplied by the expression<maths id="math0004"><img file="EP0010481A1_D0004.tif" /></maths>we find a vector of constant amplitude. It is this value which represents the real part of the weighting coefficient.

The imaginary part of this coefficient will be determined by considering the phase of the signals U_. and you_i-1.

The difference of the signals considered was found:<maths id="math0005"><img file="EP0010481A1_D0005.tif" /></maths>which can be written:<maths id="math0006"><img file="EP0010481A1_D0006.tif" /></maths>Under the same conditions as above, if T_m represents the duration of a repetition period in the absence of variation, we can write the previous expression in the form<maths id="math0007"><img file="EP0010481A1_D0007.tif" /></maths>By equalizing the two expressions obtained for the phase to within 2πf we have:<maths id="math0008"><img file="EP0010481A1_D0008.tif" /></maths>After simplification, the phase modulation coefficient which must be applied to the signal is determined so that the signal obtained after weighting has a phase identical to that which the difference vector would exhibit in the absence of variations in the duration of the recurrence periods, is :<maths id="math0009"><img file="EP0010481A1_D0009.tif" /></maths>From this formula we can determine the phase increment to apply from one period to another, either:<maths id="math0010"><img file="EP0010481A1_D0010.tif" /></maths>

After this weighting operation, the treated vector which in the general case has therefore undergone a homothetic operation followed by a rotation has become an equipollent vector to the vector applied to the input of the device which would correspond to a pure Doppler frequency. The signals obtained after weighting are then conventionally processed in filters, then rectified and applied for example to a display or operating device.

FIG. 1 gives an exemplary embodiment of the device for displaying mobile targets according to the invention, in which only the part of the receiver of the Doppler radar concerned has been shown.

The radar intermediate frequency receiver has an output S which is also the input of the device considered. This terminal S is connected to two demodulators la and Ib respectively which receive the signal leaving the receiver at intermediate frequency, as well as a reference signal by the - terminal R, reference signal which is phase shifted by n / 2 in circuit 11 before being applied to the demodulator lb. The signals leaving the demodulators 1a and 1b thus constitute the two components in quadrature of the vector representing the signal applied to the input S. It will be assumed that the different frequency transpositions carried out upstream of the intermediate frequency receiver are such that for a fixed echo, the signal applied in S has a constant phase with respect to the reference signals and also that a coherent oscillator is used in the case of a random phase transmission.

At the output of the demodulators, the signals called X and Y which are the components of the signals U considered previously are coded in digital at a sampling frequency f_e whose value is chosen according to the bandwidth of the radar receiver. It is posited that the numerical values ​​thus obtained, preferably in binary are X_AT and Y_AT. These signals are respectively applied to the delay device 3a, 3b followed by a subtractor 4a-4b. A direct connection between the output of the coding device 2a-2b is however carried out with the subtractor 4a-4b.

The delay devices can be either shift registers or a random access memory. The assembly which has just been described makes it possible to perform the subtraction of the signals X_AT and Y_AT, corresponding to two successive recurrence periods, it being understood that the signals present simultaneously at the two inputs of each of the two subtractors 4a-4b all correspond to the same distance, that is to say to the same delay relative to the transmission pulse which is at the origin of each of the signals.

The digital signals delivered by the subtractors thus appear to have been freed, in particular of the spectral line corresponding to the zero Doppler frequency.

These digital signals from the subtractors are applied to the single weighting device 5 which operates in the manner which has been seen previously.

The role of the weighter 5 is therefore to carry out an amplitude modulation and a phase modulation on the components E_x and E of a vector delivered respectively by subtractors 4a and 4b. If M and ϕ symbolize these modulations, the weighter performs the operation MVeⁱ in which V = E + jE_y.

The expression MVeⁱ turns into<maths id="math0011"><img file="EP0010481A1_D0011.tif" /></maths>

Figure 2 shows schematically how this operation is performed. The modulator has two direct channels and two crossed channels into which operators performing the operations according to the last expression are introduced. If A, B, C and D are the input and output terminals of the weighting device 5, we can assume that the AC path is the cosine path and the BD path is the sine path. The AC channel then includes an operator 12 carrying out the multiplication by cos ϕ and an adder '13 connected to an operator 17 carrying out the multiplication by - sin ϕ of the component E. Similarly, the channel AD includes an operator 14 performing the multiplication by cos ϕ and an adder 15 connected to an operator 16 performing the multiplication by sin ϕ of the component E_x. On the output terminals of the weighter 5 we collect the components P_x and P_y from the vector equipollent to the "ideal" vector.

These components are respectively treated in two sets 6a-7a, 6b, 7b and 8a-9a and 8b-9b constituting filters similar to that 3a-4a, 3b-4b connected to the input of the weighter and which reject the low frequencies located around the frequency 0 rejected by the filter placed at the input. In these assemblies, the circuits 6a, 6b, 8a, 8b are delay circuits and the circuits 7a, 7b, 9a, 9b are subtractors if the direct link avoiding the delay circuits is with transfer coefficient +1. A rectifier circuit 10 connected to the outputs of the last subtractors 9a-9b envisaged in the channels delivers signals which conventionally are transmitted to operating or display circuits.

The number of filter assemblies mentioned in the embodiment described above is not limited to two and moreover, as has been said, the structure of the filter assembly which follows the weighting can be arbitrary with a rejection zone which contains the frequency f on which the compensation system is adjusted.

In this case, experience shows that the compensation for variations in the repetition frequency of the radar is very good over a wide frequency range which surrounds the value f for which the compensation is theoretically perfect.

We have thus described a method and a device for displaying moving targets in radar equipment with a variable repetition period.