FR2846099A1 - Radar apparatus, e.g. for missile, includes systems for producing first and second signals representing mean amplitude of train of modified pulses, each weighted according to first and second weighted functions - Google Patents

Abstract

The apparatus has system transmitting pulses of energy and receiving corresponding pulses returned in response to transmitted pulses. The system compares the time interval of occurrence of each returned pulse with that of further pulse having predetermined phase relationship with corresponding transmitted pulse, and the system producing first signal representing the mean amplitude of the train of modified pulses, each weighted according to first weighting function. The radar apparatus also has system producing second signals representing a mean amplitude of the train of modified pulses, each weighted according to a second weighting function.

Description

The present invention relates to radar equipment.

  The radar apparatus according to the invention comprises: - means for emitting energy pulses and for receiving corresponding pulses returned in return, - means for comparing the time interval of occurrence of each pulse returned in re10 turn with that of an additional pulse having a predetermined phase relationship with the corresponding transmitted pulse, so as to produce a train of modified pulses, each having a duration in relation to the time interval of occurrence common to 15 l 'corresponding returned pulse, and the corresponding additional pulse, and - means for producing a first signal representing the average amplitude of the train of modified pulses, each of them being weighted according to a first weighting function which is non-uniform , for at least one range of durations of each modified pulse. In one embodiment of the invention, further means are provided for producing a second signal representing the average amplitude of the train of modified pulses, each of which is weighted according to a second weighting function, and means for forming a report signal representing the

  ratio of the signals representing said weighted average amplitudes.

  In a preferred embodiment, the

  comparison means include a mixer capable of receiving each pulse returned and of

  as an additional pulse, receive an RF radio frequency pulse, so that the modified pulses are demodulated. In the preferred embodiment, the means or means for producing a signal representing the average amplitude of the train of modified pulses responds to a predetermined Doppler modulation of the train of modified pulses, such

  so that the signal represents the average amplitude of the Doppler modulation of the train of weighted pulses. Preferably, each additional pulse is derived from the corresponding transmitted pulse, and is almost totally in time concidence with it.

  In the preferred embodiment, the means or means for producing a signal representing the weighted average amplitude of the modified pulse comprises means for producing a weighting signal and means for mixing the weighting signal with the pulse changed. Preferably, the means for producing said second signal comprise means for obtaining a bipolar signal having at least

  a polarity transition for the duration of the additional pulse.

  In an alternative embodiment, the means

  Transmitting signals include a clock, for producing a clock pulse train having a preset repetition rate and means for deriving from the pulse train both the transmitted pulses and the additional pulses. In this embodiment, the weighting signal, or each weighting signal, is derived from the clock pulse train.

  A rough measurement of distance (or range) can be obtained by means which respond to the aforementioned signal representing a weighted average amplitude of the train of modified pulses, to vary the clock frequency in order to produce a train of modified pulses. , each with a preset duration. A developed variant of the invention com5 takes a series of means responding to the different Doppler modulations of the modified pulse train to produce respective first signals representing the average amplitudes of the Doppler modulations of the pulse train, weighted according to the first function 10 of weighting, a similar series of means responding, in correspondence, to the same Doppler modulations to produce respective second signals representing the average amplitudes of said Doppler modulations of the pulse train, weighted according to the second weighting function, and a similar series of means to form the reports of the first and second signals

  associated with the same Doppler modulations.

  The invention will be better understood on reading the appended drawings in which: - Figure 1 is a block diagram of a radar apparatus according to the invention; - Figures 2, 3a, 3b and 4 are diagrams explaining the operation of the apparatus of Figure 1; - Figure 5 is a block diagram of an improved variant of the apparatus of Figure 1; Figure 6 is a diagram illustrating the operation of the apparatus of Figure 5; Figure 7 is a block diagram of another refinement of the apparatus of Figure 1; - and Figures 8a to 8e are diagrams

  explaining the operation of the device in Figure 1.

  The apparatus of Figure 1 will now be described by way of example, as a component of a proximity sensing device capable of indicating the distance between the device and a small object that

will call the target.

  Referring to the drawings, Figure 2a (i.e. Figure 2, curve a) shows the envelope of a pair of r.fu pulses. emitted successively as part of a continuous pulse train emitted by the radar apparatus of Figure 1. The pulse duration d10 is for example 50 nanoseconds (ns), and the repetition period T is of 200 ns. The proximity detection device incorporating the radar apparatus of FIG. 1 is arranged so as to be sensitive to the proximity between the device and the target, as will be seen more fully below, and for this purpose, part of the energy of each of the rf pulses flowing along a feeder 2 for the emission serves as a local reference oscillation, being sent, via a coupling 6 to a mixer or demodulator 5, the other input of which is coupled to the receiver. The energy thus supplied is not subject to a delay, and it will therefore be understood that the maximum range of the apparatus in this example (in which d takes the value of 50 ns) is approximately 8 meters 25 due to the fact that the wave propagation speed is about 3 x 108 m / s. For larger ranges, the returned echo pulse does not overlap the corresponding transmitted pulse over time, and thus, no local oscillation was transmitted to the timing circuit at the instant considered, the received echo pulse does not is not out of fashion. It is understood from the above that an echo pulse received from a target at a distance of zero completely overlaps the corresponding transmitted pulse, as shown in FIG. 2b, which makes -J that the corresponding demodulated output of the mixer circuit is be an impulse whose duration is practically equal to d. If, however, the target is at a distance, for example about 4 meters 5 from the device, the received echo pulse will overlap half of the corresponding transmitted pulse, which gives a demodulated output pulse having a duration d / 2 at the output of the mixer circuit. Intermediate durations are produced by intermediate instantaneous distances, and it is clear, therefore,

  that the duration and phase of each pulse leaving the mixer are significant for the distance from a target to the device.

  According to the mother idea of the invention, this relation is exploited by obtaining distance information either from the duration alone, or from both

  duration and phase, demodulated pulses.

  FIG. 3a shows part of the amplitude - / frequency characteristic obtained for a fixed target at a distance of zero, that is to say when the echo pulses having the timing of the pulses whose envelope is represented in Figure 2b are demodulated by being mixed, with a suitable relative carrier phase, with coupled energy from the transmitted pulses shown in Figure 2a. As can be seen, the characteristic is presented as a linear spectrum, the lines being separated from each other by a frequency equal to the inverse of the repetition period of the pulses T, and the envelope of the falling spectrum to zero for 20 M1-z (1 / d) after which it rises again in a well known manner

for such characteristics.

  On the other hand, Figure 3b shows the character-

  corresponding teristic obtained when the impulses

  echoes received from a fixed target overlap the pulses emitted by an interval of d / 2. As can be seen, the characteristic of FIG. 3b shows at 20 MHz a component of significant amplitude different from zero and the envelope curve does not pass through zero before 40 MHz.

  Thus, we see that the amplitude of any spectrum line varies with the range. For example, if the 20 MHz component of the mixer output pulses 10 is selected (for example by means of a suitable filter, or by phase detection), then for the zero range, its measured amplitude is zero, it increases with the growth of the range, until it reaches a maximum value at mid-range, then, when the energy is dissipated over a larger bandwidth, the amplitude decreases, until the maximum range of 8 meters, for which, as indicated above, there is no overlap between the echo pulses

  received and the coupled energy synchronous with the transmitted pulses, so that there is no output pulse from the mixer circuit.

  In a preferred example of the invention, described in particular in French application No. 80 04583 filed in the name of the Applicant on February 29, 1980 for "Comparison radar apparatus on spectral lines", the characteristic is sampled at two different chosen frequencies, x and y, as seen in FIGS. 3a and 3b, and the ratio x / y of the amplitudes detected at the two frequencies, is taken and used to provide the indication of distance. In this example, where the rate of change of the range is appreciable, the amplitudes of the Doppler lateral bands resulting from the two chosen frequencies are measured, and their ratio is taken. The reason for doing this is a practical reason. Even in the absence of a target, a strong video signal is present at the output of the mixer,

  without modulation of the passage of the local oscillator.

  The fraction of the output which overlaps the signal emitted when a moving target is in range, has a small proportion of amplitude fluctuation, due to the Doppler sidebands. You can get rid of such a fluctuation by filtering, thus rejecting unwanted pulses, and by providing a signal

  with all the desired information about the target.

  Referring now to Figure 1, the apparatus shown by way of example comprises a CW continuous wave oscillator connected to an aerial antenna device 2 by a switch. The switch is controlled by a control

  switch using a timing circuit 1.

  Circuit 1 includes a clock which produces a bipolar square wave of mean zero (i.e. a rectangular wave with a form factor of 50%), of frequency f1 followed by a frequency divider to produce a square wave of frequency fl / 2, which is supplied to the input of an AND gate, 8, and to another input of gate 8 by another 2: 1 divider, so that the output of gate 8 is a wave

  square with frequency fl / 4 but with a pulse width equal to 1 / fl.

  Part of the energy in the feeder 2 is coupled to a mixer or demodulator 5 via a coupler 6. The pulses received are supplied to the mixer by the aerial antenna device 4, and the output of the mixer , which comprises a train of pulses having an amplitude modulated component at the Doppler frequency, that is to say an alternating video signal, is brought by a video amplifier to a mixing circuit 7. The circuit 7 comprises a first balanced mixer 71 which receives directly at one input the video signal (without filtering other than the capacitive coupling serving to block the intrusion component) from the video amplifier, and, at the other input, the bipolar square wave of zero frequency average f1 / 2 received directly (without filtering other than capacitive coupling) from timer circuit 1. The mixer circuit also includes another balanced mixer 72 capacitively directly coupled nt at one of its inputs to the video amplifier and to the other, at the clock of timer circuit 1, for excitation by the bipolar square wave of mean

zero frequency fi.

  The bipolar square wave of frequency fi includes a fundamental component of frequency fi plus many odd harmonics, the amplitudes of which are inversely proportional to their respective order. The square wave of frequency f1 / 2 likewise comprises a fundamental component of frequency f1 / 2

  plus many odd harmonics whose amplitude decreases with order.

  Thus, the components having the basic frequency m.fl (m being an odd integer) present at the input of the mixer 71 are demodulated therein and contribute to its output. Other basic frequency components (m / 2) fl present at

  the input are likewise demodulated in the mixer 72.

  In the case where there is a radial target speed 30 (that is to say a relative velocity along the line of sight), the Doppler effect causes each basic component to be divided into a pair of components, each of half amplitude and, shifted in frequency on each side of the nominal frequency by a Doppler deviation, 2V / XHz, in which vm / s is the speed of

  change of the relative range and X m is the wavelength of the carrier. In practice,. this frequency is much lower than the repetition frequency 1 / T Hz.

  This is shown at 41 in FIG. 3b for a spectrum line only, and in the case where simultaneous referrals from two targets give rise to pairs of lines S1 and & 2 respectively. For each target, the demodulation of each Doppler pair produces a zero mean output which varies at the Doppler frequency, all of the output components being mutually in phase, and therefore directly additive. Although the respective amplitudes of all of these output components are proportional to the strength of the received signal, they vary according to different laws with the change in distance (range) and, therefore, with the degree of overlap between the transmitted signals and the signals returned. As a result, the resulting output varies in a partially linear fashion with the scope, although

  is preserved the general form of variation indicated by Figures 3a and 3b relative to the main component.

  The function of the mixers 71 and 72 can be explained with reference to Figure 8. The waveform 8a shows the output portion of the demodulator 5 which results from reflections on a compact moving target; alteration of its envelope at the Doppler frequency

  is obvious. This signal, after video amplification, gives me an input for each of the balanced mixers (multipliers) 71 and 72.

  Waveform 8b shows the continuous square input wave of frequency fl / 2 going to mixer 71, while 8c shows the product waveform

- 2846099

  (a). (b) supplied by 71. Because, whenever a is not equal to zero, b has a constant value, the curves 8a and 8c are of identical shapes. The dotted line on curve 8c indicates its fundamental component at the Doppler frequency, the value of which at each instant essentially depends on the surface of the neighboring pulses. It will therefore be understood that

  its amplitude is proportional to the duration of the pulses a, as weighted by a uniform weighting function defined by the signal (b).

  The waveform 8d shows the continuous square input wave of frequency fi going to the mixer 72, while 8e shows the waveform of the product (a). (D) supplied by 72. We see that the first part of each pulse a is inverted, which reduces the resulting pulse area and thus introduces phase sensitivity of the pulses a. The fundamental component, represented by dotted lines on the curve 8e is reduced correspondingly. The signal d 20 does not uniformly weight the signal a, because the

  pulse duration of signal a varies with range.

  Mixer outputs are supplied to a

Doppler selection circuit, 90.

  The circuit 90 comprises two identical channels (or channels) 25 connected to the mixers 71 and 72. Each channel comprises a Doppler amplifier, an envelope detector and an integrator. Doppler amplifiers

  reject all the output components of the mixers whose frequency falls outside the expected range of Doppler frequencies.

  The purpose of the envelope detector and integrators is to provide continuous measurement

  (D.C.) of the strength of the alternating outputs of the Doppler amplifiers.

  il The continuous outputs (DC) DS1 and DS2 of the channels of circuits 90 are applied to a reporting circuit, 80. Circuit 80 provides a measurement of the ratio of the amplitudes of the continuous signals DS1 and DS2, by subtracting in a subtractor 10 the outputs logarithmic amplifiers, of a type known in

  itself, which amplify signals DS1 DS2, so that their instantaneous outputs vary linearly with the logarithm of their respective instantaneous inputs.

  The partly linear manner in which the DS1 and DS2 signals vary with the range of the target (time delay) and the manner in which the

  DS2 / DS1 ratio are shown in Figure 4.

  In terms of spectrum, the mixers inside circuit 7 choose, weight and demodulate two different series of spectral lines coming from the bipolar video signal, allowing each half of circuit 90 to measure the square of the weighted average power of each series, which allows the

  circuit 80 to determine the relationship between the measurements.

  The weighting derives from the relative amplitudes of the harmonics of the square input waves going to 7.

  From another point of view, the circuit 90 provides25 a measurement of the average amplitude (as weighted by the signal b or d) of the component induced by the target of each of the signals Si or S2 via their content at the Doppler frequency. When the bipolar video input to the mixer 71 exists, the square wave at its second input remains unchanged, which leaves the video signal unmodified as output signal Si. During the first part T / 8 of each cycle d the mixer 72 changes in the signal

  S2 the polarity of the video signal, reducing the average amplitude of the target returns, with an equivalent range of less than T / 8 as shown in Figure 4.

  By determining the amplitude ratio of the signals DS2 and DS1 in the circuit 80, the effect of the intensity of the reflection of the target on the

strength of the two signals DS1 and DS2.

  A graph showing the variation of the DS1 / DS2 ratio with the range, in the case of a single target, 10 is represented in FIG. 4; we see that for all

  range less than the maximum range (corresponding to T / 8) there is a single value defining the range.

  Timing circuit 1 allows the form factor of the pulses sent to be maintained

  substantially constant by using a suitable clock frequency and drawing both the width of the pulses emitted and the repetition frequency.

  The ratio of the pulse width to the pulse spacing is thus kept substantially constant, even if the source of the clock pulses drifts in frequency. In addition, the frequency referenced y in FIG. 3a, for a zero amplitude of the envelope, is the clock frequency itself and this is significant, since if the clock frequency varies, causing a variation of the actual frequencies of the spectral lines, however, the y line remains at the clock frequency, the x line remains at half the clock frequency, and the other harmonics of the repetition frequency remain consistent, that is, ie at fixed proportions of the frequency of the line y. This means that any change in the output level at the clock frequency is due to a change in the characteristics of the spectrum which is produced by a cause other than variations in the degree of overlap between the transmitted pulses and the received pulses, itself caused by a change in the distance between the missile and the echo source (for example the sea). It is thus possible to use the value of the output of one of the channels, for example the upper channel of the circuit 90, whose output is referenced DS1, to change the clock frequency in a direction 10 tending to keep the The overlap interval and, therefore, the duration of the demodulated pulses, to a low value, with the result that variations in the clock frequency can be monitored to give an indication of altitude. Such an altimeter can be used in conjunction with a proximity sensing apparatus, of the type described above, to provide a coarse / fine procedure. When the altitude is considerable, an indication of the altitude, although of reduced accuracy, can be obtained by monitoring the clock frequency, as mentioned above. For example, the clock of circuit 1 can be a voltage controlled oscillator. A control device il (FIG. 1) can compare the outputs DS1 of the channel Sl-DS1 of circuit 90, with a predefined threshold device 25 terminated, any excess being amplified and used to increase the frequency of the oscillator and for Therefore, make the overlapping pulses tend towards zero. The oscillator control voltage would thus indicate a rough range. When the altitude has decreased to a degree compatible with the above-mentioned maximum range, of the order of 8 meters, the clock frequency is brought back to its initial value, and the method of pulse overlap above. above described is then used to give precise results at short range. In general, the intensity of an echo signal coming from a small target, relative to the range, follows a law corresponding to the inverse of the

  fourth power, but, in the case of a stretched target, for example the sea, this law is more exactly a law according to the inverse of the square.

  In the apparatus described above in FIG. 1, use is made of the fact that the outputs Si and S2 of the mixers 71 and 72 contain a significant component which varies at the Doppler frequency corresponding to the movement of the target relative to the radar equipment, to determine the DS1 / DS2 ratio defining the range. The Doppler selection circuit 90 chooses the Doppler frequency components Si and S2 at the outputs of the mixers 71 and 72 intervening in a planned range of frequencies (and consequently in a speed range) & 1, using the Doppler amplifiers and the DS1 / DS2 ratio (DS1 being the amplitude of this component extracted from the output of the mixer 71, and DS2 20 being the amplitude of the same component extracted from the output of the mixer 72) is determined in circuit 80. The report indicates the scope. Such an apparatus has only one channel to cover the Doppler frequency band S and can determine only the range of a target, producing a Doppler frequency in the range Si. According to the exemplary improvement of the present invention which is given at FIG. 5, a series of parallel channels, each comprising a baked cir30 for selecting the Doppler bandwidth, and a report circuit (90.80), etc. is provided following the mixer circuit 7, the channels responding respectively to a plurality of components, each of them varying at a different frequency. So only one

  switchgear can operate simultaneously with different roles.

  Referring to FIG. 6, it can be seen that a remote-controlled vehicle RPV flies over the ground in the direction of an obstacle 0 on which a beacon B is placed, in the vicinity of a moving target T. The vehicle is controlled from a ground station CS, via a data chain DL, which is not shown in FIG. 6. If we refer to FIG. 5, the vehicle driven at distance transports organs 1, 2, 4 and 7 with the CW oscillator, the switch, the switch control, the video amplifier, and part of the DL system. The ground station includes the rest of the DL system, and channels 80 to 84, 15 etc ... As shown schematically in Figure 5,

  the CL data link includes two channels C1 and C2.

  The target T produces a Doppler frequency in the range S, and thus, its instantaneous range from the RPV vehicle is determined as described above, by the channel comprising the Doppler selection circuit and the reporting circuit 80. The aerial device 2, 4, of the VPN vehicle is adapted to allow the emission of the pulses, as well as their reception, both forward and downward. This allows the altitude of the RPV to be determined. The Doppler frequency due to the component of the relative movement between an element of the ground L and the RPV according to the corresponding line of sight falls in a second distinct range S2, and the channel 91, 81 selects, by means of the selection circuit 91 the components having a frequency in the range & 2 modulating the outputs of the mixers 71 and 72, and determines, by means of the ratio circuit 81, the ratio of their amplitudes.

-_ 2846099

  The obstacle O will produce yet another Doppler frequency g3, to which the channel 92.82 will be adapted,

  but none of the channels 90.80 or 91.81.

  The beacon B includes a repeater ("trans5 ponder") which responds to a pulse emitted by the radar system of the vehicle RPV, to emit a return pulse whose carrier frequency is displaced from that of the emission by a "Doppler" frequency artificial, in a range J 4 much larger than that of a true target. A tag identification code

  could be superimposed on the artificial Doppler frequency. Channel 93, 83 only responds to the frequency range i 4.

  Other routes can be added, as needed. For example, a 94.84 channel can be used

  detecting the passage of projectiles launched from the ground.

  Detection of Doppler frequency bands

  if, 2, S3 and 4, and, consequently, the determination of the distance or range of each individual target is made simultaneously for all the targets.

  The RPV vehicle is equipped with a data link, of a type known in itself, for receiving data from the ground control station CS and for transmitting it. Since the information concerning the range of the target T, the altitude, the range or distance of the obstacle O, and the information coming from the beacon B, are obtained after filtering in the vehicle, the width of The link band (at least from the RPV vehicle to the CS ground station) remains very reduced, if we compare it to what it is necessary to use with a conventional radar in the RPV vehicle. If we were using a conventional radar, the bandwidth would be

  in the video range as shown at 40 in FIG. 3b, while using a radar according to the invention

  tion, the bandwidth necessary for the audio range is reduced, as shown in 41, which is necessary to take account of the inevitable variations of a line in FIG. 3b. The data collected by the radar matching device according to the invention are analyzed at the ground station. The equipment necessary for the vehicle to collect this data is relatively simple and

  inexpensive, while the installation of data analysis is relatively complicated and expensive.

  In the improved variant of Figure 1 which is shown in Figure 7, there are a plurality of channels or blocks responding to targets at different ranges. Each range block includes a mixer 5, a line selection circuit 6 and one or more Doppler channels (90,80), (91,81), etc., as shown. The mixers 5 are connected to an aerial aerial system which is common to them, 4, via an RF amplifier. However, the local oscillator signals LOi, L02, sent to the mixers 5 have different timings, which defines a series of coarse range doors. The rough position of the target is thus indicated by that of the

  channels that responds to it. Likewise, the signals f1, F1 / 2, sent to the mixer circuit 7 in the different range blocks 25 must also have different suitable timings.

  The ambiguity which is present and which is due to the fact that the output is symmetrical with respect to the edge of the pulse of the local oscillator can be removed by confirming the presence of a signal at the two gates. A fine indication of the position of the target within the range corresponding to a

  door is provided by the report of the sizes of the

  alternative resulting posants of Si and S2 at the outputs of mixers 71 and 72, as described above. Targets with the same range, but having significantly different Doppler values are detected by using the Doppler channels (90,80), ( 91.81), etc ...

  However, each range block must be provided with a single Doppler channel 80, so as to respond to a single Doppler frequency band, if desired, although in this case targets appear simultaneously in the same block range, but with different Doppler frequencies that cannot be

  distinguished, each of them damaging the extent of the reach of the other.

  Each staff block can be used independently. However, when a target goes from one block to the next, the signal goes from one channel to another. Therefore, one can conveniently increase the range by subtracting the logarithmic outputs of the two range blocks (i.e. by deriving the logarithm of the ratio of ratios) in a subtractor 12, to provide a final output voltage which is significant of the position of the target in the two blocks

range.

  In the above, the square waves have frequencies of fl and fl / 2. We can understand that other frequencies, for example fi and fl / 4 could

  be used, or else fl and the continuous component.

  In the example in Figure 1, a CW oscillator and a switch are used to produce a pulsed oscillation, which is consistent from one pulse to another. However, this is not essential, and an oscillator which is not coherent from one pulse to another could be used. Indeed, in this case, there would nonetheless be consistency between each

  pulse sent and the corresponding pulse received.

  The invention teaches the use of long

  impulses, with the discrimination of short impulses.

  The invention has been described with reference

  to the device in which the two-channel DS1 and DS2 output reports provide a measurement of the range.

  However, as is clear from Figure 1, it is possible to determine the range (subject to the reflective nature of the target) from the

  one way exit. In practice, the channel producing the signal DS2 would be used, due to the substantial weighting of the phase provided by the signal d of Figure 8.

Claims (17)

  1. A radar apparatus comprising - means for emitting energy pulses and for receiving corresponding pulses returned in return, - means for comparing the time interval of occurrence of each pulse returned in return with that of a additional pulse having a predetermined phase relationship with the corresponding transmitted pulse, so as to produce a train of modified pulses, each having a duration in relation to the time interval of occurrence common to the corresponding returned pulse and to the corresponding additional pulse, and - means for producing a first signal 15 representing the average amplitude of the train of modified pulses, each of them being weighted according to a first weighting function which is non-uniform, for at least one range duration of each modified pulse. 2. An apparatus according to claim 1, characterized in that it further comprises: - means for producing a second signal representing the average amplitude of the train of modified pulses, each being weighted according to a second weighting function, and - means for forming a report signal   representing the ratio of the signals representing said weighted average amplitudes.   3. An apparatus according to claim 2, characterized in that said second weighting function is uniform for all the durations of each pulse changed.   4. An apparatus according to one of claims 1 to 3, characterized in that each of the means   used to produce a signal representing the weighted average amplitude of the modified pulse, comprises means for producing a weighting signal and means for mixing the weighting signal with the modified pulse.   5. An apparatus according to claim 4, 10 characterized in that the means for producing said second signal comprise means for obtaining a   bipolar signal having at least one change of polarity during the duration of the next pulse.   6. An apparatus according to claim 5, characterized in that said bipolar signal comprises a continuous bipolar square wave.   7. Apparatus according to one of claims 1 to 6, characterized in that the transmission means comprise a clock for producing a train of clock pulses having a repetition rate   predetermined and means for determining from the pulse train both the pulses emitted and additional pulses.   8. Apparatus according to claim 7,   taken in combination with claims 4, 5 and 6,   characterized in that each of the weighting signals is obtained from said train of clock pulses.   9. Apparatus according to claim 7, characterized in that it comprises means responding to said signal representing the weighted average amplitude of the train of modified pulses, for varying the clock frequency in the direction tending to produce a train of modified pulses, each having a predetermined duration. 10. Apparatus according to claim 9,   characterized in that the predetermined duration is substantially zero.   11. Apparatus according to one of claims 1 to 10, characterized in that the additional pulse is equal in duration to the corresponding transmitted pulse.   12. Apparatus according to one of claims 1 to 11, characterized in that the additional pulse coincides in time almost entirely with the corresponding transmitted pulse.   13. Apparatus according to one of claims 1i or 12, characterized in that each pulse   additional is derived from the corresponding transmitted pulse.   14. Apparatus according to one of claims 1 to 13, characterized in that each of the means   for producing a signal representing the average amplitude of the train of modified pulses responds to a predetermined Doppler modulation of the train of modified pulses, so that said signal represents   the average amplitude of the Doppler modulation of the train of weighted pulses.   15. Apparatus according to one of claims 1 to 14, caractéiséenceque l.es 5ensde comparison   include a mixer arranged to receive each return pulse and to receive an RF pulse as an additional pulse, such that   the modified pulses are demodulated.   16. Apparatus according to one of claims 1 to 15, characterized in that the transmission means   include an aerial antenna, an oscillator   continuous wave and means for selectively transmitting the oscillation produced by the oscillator to the antenna.   17. Apparatus according to claim 2, characterized in that it comprises: - a series of means responding to different Doppler modulations of the pulse train modified to produce first signals representing the 10 average amplitudes of said Doppler modulations of the pulse train , weighted according to a first weighting function, - an analogous plurality of means responding to the same different Doppler modulations to produce second signals representing the average amplitudes of said Doppler modulations of the pulse train, weighted according to the second weighting function, and - a similar series of means for forming   the ratios of the first and second signals associated with the same Doppler modulations.