GB2601987A - An instantaneous phase meter receiver for an electromagnetic signal in a wide frequency band and signal processing method - Google Patents

Abstract

The phase meter-receiver receives, on reference and measurement inputs, ultrahigh frequency wideband signals (S_Re and S_v1). It comprises a mixer (19) of the output signal of the measurement channel (11-17) with the output signal of the reference channel (82-87), for delivering, to the processing means (41, 42, 45, 50), signals comprising information regarding the phase difference between the received signals. The processing comprises an analog-digital conversion (41-42) of quadrature components representing the phase of the measurement signal. The reference receiving channel (82-87) and measurement channels (11-17) are not subjected to any change of frequency. Before being applied to the mixer (19), the output of the reference channel is subjected to a phase modulation by a signal with several states switched at a time sequence linked to the sampling sequence of the analog-digital conversion. By time multiplex measurement of the phases of signals S_v1 and S_v2 received at two spaced serial inputs relative to the phases of the reference signals S_Re received at a reference serial, the direction of arrival of the (e.g. radar) signals may be determined.

Description

An instantaneous phase meter receiver for an electromagnetic signal in a wide freguencv band and signal processing method.

The present invention concerns the measurement of the phase of an electromagnetic signal after a radio-electric propagation, or simply taken up at different points of an electronic circuit, and capable of being situated at any point in a wide frequency band, in principle in an ultrahigh frequency band.

This measurement of the phase (therefore of time) has several applications. If two (or more) measurements are effected on the same signal after it has been subjected to different delays, it is possible to gain access to an instantaneous measurement of the frequency of the signal; if several measurements are effected on the same signal, but picked up by aerials of different locations, an interferometry may be obtained which makes it possible to gain access to the instantaneous arrival direction of the electromagnetic signal. The unit is advantageous for the location of transmitting sources, such as radar, and the determination of their frequency with respect to time. Thus the invention applies in particular, but not exclusively, to systems for electromagnetic countermeasures against radars.

We have already developed (and described in FR-90 01 762, not yet published) a receiver comprising in the main: - a reference antenna and at least two measurement antennas, these various antennas having a common spatial range for observation in a wide band, - a high frequency receiving channel for the reference antenna, and two high frequency receiving channels for the at least two respective measurement antennas.

In this prior receiver, each measurement channel comprises a mixing means with a single sideband (SSB), with the output of the reference channel as the local signal. It should be emphasized that each one of these mixing operations is effected between two received signals neither of which is therefore spectrally pure.

At the output of these two SSB mixing means two components (signals termed "complex" or sine/cosine signals) are obtained on each occasion. After an analog/digital conversion, a known processing makes it possible to obtain the phase difference between the signals picked up by these same antennas. Diverse means make it possible to measure the phase of the signal of interest in the observed wideband; since this signal is, for example, the most intense, limiters are placed into the measuring channels. The phase measurement nevertheless remains difficult, taking into account the fact that there remain many signal residues, in particular because the SSB mixing is effected between two received signals that are spectrally impure.

Hence, although a receiver of the above mentioned type provides worthwhile results, we wish to improve the performance still further, in particular, as far as the phase measurement is concerned.

The object of the present invention is to provide a solution for this problem.

According to one aspect of the present invention we provide a phase meter-receiver for ultrahigh wideband frequency signals 30 of the type comprising: a reference input and at least one measurement input for receiving wideband ultrahigh frequency signals coming from a propagation of electromagnetic waves and picked up by at least one antenna or aerial, - a high frequency receiving channel for the reference input, at least one high frequency receiving channel for the measurement input, mixing means capable of mixing the output signals of the measurement channel with the output signal of the reference channel, the output of these mixing means containing information regarding the phase difference between the signal coming from the reference input and the signal coming from the measurement input, and means for processing the signals coming from these mixing means, comprising an analog-digital conversion of the signals with two components representing the phase of the signal from the measurement input; wherein the outputs of the receiving channels have the same frequency as their inputs; and wherein, before being applied to the mixing means, the output of the reference channel is subjected to a phase modulation by a signal with several states switched at a chosen time sequence.

A second aspect of the invention provides a method of processing of wideband ultrahigh frequency signals coming from a propagation 25 of electromagnetic waves of the type wherein: a) there is a reference receiving channel, and at least one measurement receiving channel, b) the output signals of the measurement channels are selectively mixed with the output signal of the reference channel, 5 and c) the signals coming from this mixing are processed, after an analog-digital conversion of these signals, in the form of two components representing the phase of the signal from the measurement input relative to the signal from the reference input; d) the reference and measurement receiving channels are not subjected to any change of frequency; and e) before being used for the mixing, the output of the reference channel is subjected to a phase modulation by a signal with several states switched at a chosen time sequence.

In accordance with the present invention, no initial provision is made for any change of frequency in the high frequency reference and measurement channels, so that the mixing in the single side-band is effected at the frequency of the received electromagnetic signals, which is frequently called uhomodyne mixing". The out-put of this SSB mixing is therefore a direct representation of the phase of the measurement signal considered, in its as received state, relative to the reference channel.

One difficulty emerges in that an SSB mixer has a "false zero" (offset") in the same way, moreover, as the analog-digital 25 converters which follow it. Moreover, the limiter of the measuring channel generates harmonics; the same applies to the SSB mixer.

To avoid this difficulty, before it is applied to the SSB mixer the output of the reference channel is subjected to a phase modulation at a time sequence defined by the processing unit. Preferably, this time sequence is linked to the sequence of the sampling clock of the analog/digital conversion means.

The phase modulating signal may be a zero/180° two phase signal, synchronous with the sampling clock. We have observed that a special processing of the respective signals obtained during these two states of the two-phase signal makes it possible to eliminate the offset errors, as well as the errors linked to the first harmonic.

More generally, the phase modulating signal may be defined on the basis of a regular subdivision of a trigonometric circle into P parts, the various phase states obtained being taken modulo 180°. For P=2, one obtains the above mentioned two-phase signal. For P=3, one obtains 00, 120° and 240° (= 60° modulo 180°) and so on. At present, it is considered that values of P in excess of 8 are no longer worthwhile.

Moreover, since no change of frequency is effected in the reference and measurement channels, it becomes possible to use a single measurement channel that is temporally multiplexed so as to process successively the signals to be measured.

Such a device may serve in particular as a frequency meter, or an interferometer, or for both these functions at the same time, as will be seen below.

Other characteristics and advantages of the invention will emerge on examining the detailed description given below with reference to the attached drawings wherein: Figure 1 is a schematic diagram of a device in accordance 5 with the invention, used as an interferometer; Figure 2 is a schematic diagram comparable to that of Figure 1, but wherein the apparatus includes additionally a test oscillator; Figure 3 illustrates a second embodiment wherein the apparatus serves at the same time as a frequency meter and an interferometer; Figure 4 illustrates a variant of Figure 3 wherein a single channel serves at the same time for measuring the frequency and effecting the interferometry; Figure 5 is a block diagram illustrating the operating stages of the device in accordance with the invention; Figure 6 is a schematic diagram of the adjustable time delay that can be used in conjunction with the embodiment with a test oscillator; and Figure 7 is a schematic diagram of a variable delay gate used in conjunction with the diagram of Figure 6.

The elements of the attached drawings are in essence of a definitive nature. This also holds good for the Annex of Formulae at the end of the description. These elements are therefore to be considered as an integral part of the description and may not only contribute to a better understanding of the invention, but also take a part in the definition of the invention.

Figure 1 concerns the case of an interferometer wherein a reference aerial is provided (not shown) which supplies an electric signal S_Re representing electromagnetic waves picked up by the aerial. Two measurement aerials are added thereto to supply, in the same way, two respective electric signals S_V1 and S_V2. These signals are in the ultrahigh frequency range. It will also be noted that these signals have a wide band. Inside this band, there is situated a wanted signal with electrical characteristics (frequency and phase) which are unknown beforehand.

It is known (from the unpublished FR-90 01762, already referred to above) that a receiving channel is used for the signal S_Re, as well as a plurality of receiving channels used for the measurement signals such as S_V1 and S_V2. According to the prior art, these receiving chains comprise frequency changes. Each of the measuring chains processing the signals S_V1 and S_V2 is subject to mixing with the signal coming from the reference channel, after which a demodulation is proceeded with which provides from the output of the demodulator components in phase and in quadrature, i.e. "complex" components, which are applied to analog-digital converters for subsequently undergoing known processing operations which make it possible to measure the phase of each one of the two measurement channels S_V1 and S V2 relative to the reference channel.

The device in accordance with the invention is distinguished from that set out above, first of all in that there is no 30 change of frequency in the receiving channels. Consequently these signals retain their original frequency. Before effecting the mixing between the output of the reference channel and the outputs of the measurement channels, the reference channel will be subjected to a phase modulation according to several predetermined phase levels. The simplest version lies in obtaining this phase modulation with two states which correspond to the phases 0° and 180°.

Although it is conceivable to retain separate structures for each of the measurement channels assigned to the signals S_V1 and S_V2, the invention makes it moreover possible to process them in a multiplexed mode in time in a single receiving channel (which would not be compatible in practice with the use of mixers in this receiving channel).

Thus in Figure 1, the two measurements signals S_V1 and S_V2 are applied to a selector switch 11 which applies one of them to an input amplifier 12. There may then be provided a filtering stage 13, following which the measurement channel comprises an amplifier 17 which is, in principle, a limiter. Taking into account the response of the filter 13, the amplifiers 12 and 17 are chosen in such a way that the input applied to the succeeding modulator 19 practically attains the saturation limits of the modulator.

The reference channel has a similar structure (elements corresponding to each other having the same numbering but offset by 70). Thus the signal S_Re is applied to an input amplifier 82 which may be followed by filters 83, then by an amplifier 87 (which, in contrast to the amplifier 17, is linear) and then by the phase modulator 91 added in accordance with the present invention. The other input of this phase modulator receives a signal 92 with two states which correspond to the 0° and 180° phases already referred to. The output of the modulator 91 is applied by means of an amplifier-limiter 95 as the second input of the modulator 19 of the measurement channel.

The outputs of this modulator 19 comprise a sine channel on the line 21 which is applied to an analog-digital converter 5 41, and a cosine channel on the line 22 which is applied to an analog-digital converter 42.

The outputs of the units 41 and 42 are applied to a digital circuit 50 which effects the fast processing of the phase data thus obtained, and also defines the sampling frequency 10 of the analog-digital converters 41 and 42.

This unit 50 has a second function which is to synchronize the switching with the modulation and the sampling of the analog-digital converters 41 and 42. It is indeed preferable for this synchronization to be carefully ensured.

Moreover, the unit 50 may define the switching order for the switch 11 by means of a line 511, this order being marked SW/l/2.

The unit 50 defines digital outputs representing the phases of each of the measurement channels relative to the reference 20 channel. It is also connected to a control bus which may form part of the control unit of a countermeasures system.

Since the case in question concerns a simple interferometry, it is desirable for the unit 50 to be informed of the existence of a frequency to be measured, in order to avoid making phase measurements on erratic signals which are of no significance.

The embodiment of Figure I may be either without any of the filters 18 and 33, or only with one filter category (in principle, in this case, the filter 83), or with the two categories of filters at the same time.

Moreover, the filters used may be switchable, as indicated in Figure 1, and may comprise, for example, either a band-pass filter 833 and 133, or a band-stop filter 835 and 135. The choice between these different filter categories may depend on the electromagnetic environment in which the signal of interest is located, for example, so as to eliminate jammers.

The advantage of such a device is that it makes possible phase measurements on a signal, unknown beforehand, in a wide frequency band, this phase measurement being virtually instantaneous (that is to say, that it does not require a long term stability of the signal whose measurement is desired).

Since the reference channel operates in a linear mode as far as the input of the mixer 91, the 0/180° phase modulation applied through the other input of the mixer is applied only to the fundamental frequency of the signal passing through this reference channel, before any creation of harmonics.

The limiter stage 95 which intervenes after the modulation will, for its part, create harmonics, but these Are dephased according to their ranks, for example twice by 1800 for the harmonic 2 (first harmonic) which renders them virtually interference-free.

It will, moreover, be noted that it is the phase difference between the two measurement channels S V1 and 8V2 that is advantageous. Consequently, the absolute phase of the reference channel S Re does not intervene in the accuracy of the measurement. It is only necessary that it should remain stable for a short term, that is to say, during one scanning cycle of the signals 1 and 2, under the action of the switch 11.

Preferably, the phase modulation effected by the input 92 of the modulator 91 switches analog-digital converters 41 and 42 in a synchronous manner with the sampling clock Fech. This makes it possible to eliminate the errors due to the offset of the analog-digital converters after the digital processing, and also the errors due to the even harmonics which may be generated by the amplification chains, it being recalled that the measurement chains operate in a saturation mode, or near the saturation limit.

A scanning cycle (Figure 5) here takes place in four elementary periods: - at time Ti, the measurement M10 is effected which relates to the signal S_V1 with a 0° phase modulation; - at time T2, a signal Mil is determined which also relates to 5_V1, but with the phase of 180°; - at time T3, a signal M20 is obtained which this time relates to the signal S_V2 with the 00 modulation; and at time T4, a signal M21 is obtained which relates to the signal S_V2, with the modulation of 180°.

A priori, the sequence wherein the steps T1 to T4 are effected is not important. It simply suffices for these steps to be sufficiently close to one another, taking into account the stability of the signals that are of interest.

Subsequently, a step TT6 will make it possible to gain access to the respective phases of the S_V1 and S V2 signals in relation to the reference.

Finally, as will be seen below, at TT7 it is either possible to determine the relative phase of the channel V1 relative to the channel V2, or at TT8 to determine the frequency of the reference signal, in the case where the measurement signals of the channels V1 and V2 are not obtained by independent aerials, but are them-selves also derived from the reference channel through delay lines of different durations.

Reference will now be made to the first page of the formulae given in the Annex.

In the calculations, a single sideband (SSB) has been considered, which obtains the complex product. In practice, an SSE has been used which obtains the product of a real value multiplied by the same complex value. The result is the same, taking into account the low-pass filtering obtained at the output of the SSB, as will be explained below.

The equation 1 expresses a notation, whereby the symbol A_F represents the real number 2.n.F.t.

Depending on whether the phase modulation amounts to 0 or 180° at the level of the mixer 91, the output of the reference channel is expressed at this stage by Sa_Re (equation 2A) or Sb_Re (equation 25 28).

In these two equations, H/ represents the fundamental (for the coherence of the notations) and 112 represents the harmonic 2.

In the case of the equation 2B, where the fundamental is dephased by w, the harmonic 2 is dephased by 2w.

The equation 3 expresses the output signal of the channel V1 (input of the SSB), that is to say, S_Vl. To obtain the phase difference between the reference channel and the signal channel, one takes the conjugate of the signal channel, whence the formulation of the exp(-j.xx) type in this channel. Its content for the fundamental and the harmonic 2 is thus expressed by the equation 3.

At the output of the demodulator 19 (which may be likened to an SSB mixer), low pass filtering makes it possible to extract the continuous component of the signal which is expressed by the equations 4A and 4B, according to the modulation state at the level of the modulator 91. On each occasion, the signal is equal to the sum of (i) the false zero, or offset, of the demodulator, and (ii) the product of the output signal of the reference channel multiplied by that of the measurement channel (here S V1).

A comparison of the equations 4A and 48 shows that the contribution of the fundamental does not have the same sign in the two expressions of the equations 4A and 48.

Therefore, as expressed by the equation 5, by adding the signal Sa_Ssb to the signal Sb_Ssb dephased by 1800, the value of the modulus of the sought fundamental frequency is obtained twice.

At the output of the SSB, the real and imaginary parts of the second term of the equation 5 are measured separately.

These two parts are given respectively by the expressions 6A and 6B of the Annex.

From these real and imaginary parts, one may derive the wanted phase _V1, which is given explicitly by the expression 6C.

Thus the calculation shows, and experience confirms, that the 5 use of a two-state modulation makes it possible to cancel the offsets of the demodulator 19, as well as the contributions of the even harmonics of the measurement channel.

It is similarly demonstrated that the offsets of the analog-digital converters are eliminated, as has already been mentioned, subject to an appropriate synchronization of the sampling frequency with the changes of state of the phase modulation applied to the modulator 91.

Moreover, the filters illustrated as an option in Figure 1 may serve not only to eliminate jamming, but also to process 15 the case of several concomitant signals.

In the absence of filtering, it will be seen that the accuracy of the proposed phase meter will decrease according to the ratio between the level of the useful signal and that of the interference or secondary signals. By eliminating the secondary signals by means of an appropriate filtering, the full accuracy of the phase meter is restored.

Preferably, the filters can be tuned to any value of the frequency of the processed sideband. For most of the applications, the rejected or passing-band, as the case may be, is generally small as compared with the total band of the device.

The band-pass filters are mainly used when several interference signals are spread over the band, thus making it possible to concentrate only on the useful signal.

The band-stop filters are mainly used to eliminate a single or several interference signals provided that these have a small band, less than or equal to, the band eliminated by the 5 filter.

By providing filtering only in the reference channel, and not in the measurement channels, one may be faced with relatively large measurement errors when the ratio between the useful signal and the interference signal in the measurement channel comes to be low or negative. On the other hand, as a result of the reduction in mass and cost implied by the elimination of filters in the measurement channels, practical advantages are obtained which may justify providing the filters only in the reference channel.

Figure 2 will now be considered.

One of the important aspects of the present invention is the fast speed of the switching of the switch 11 which must, moreover, be effected in synchronization with the sampling signals of the analog-digital converters, as well as the changes of state of the modulation signal applied to the modulator 91.

Although the synchronization can be effected beforehand in the absence of the filters 13 and 83, this is not the case in the presence of all or one of these filters, because their 25 passing time may be variable.

To resolve this problem, an additional position is provided on the switch 11, as well as a switch 81 is provided, which makes it possible either to choose the reference channel or to choose another contact position. The two free contact positions receive the output of a test oscillator 70 on which the reference channel and the measurement channel will then be caused to operate. Correspondingly, the control signal on the line 111 takes charge of the test switching and becomes SW/1/2/T. A control line 181 is added for the switch 81, which permits the change of state SW/R/T between the real reference channel and the output of the test oscillator.

As for the rest, by means of this test oscillator it is 10 possible to measure the time delays occurring within the system in all the situations, with a view to compensating them thereafter.

It is, of course, with the object of simplifying the drawings that the filters do not appear in Figure 2, but it should be 15 emphasized that the assembly with a test oscillator 70 is most advantageous in the presence of such filters.

Figure 6 illustrates one way of applying the temporal correction thus determined. On the left-hand side, the signals are illustrated at 45-0, 92-0 and 111-0, which serve respectively for the sampling of the converters 41 and 42, for controlling the modulation of the modulator 91, and for controlling the state of the switch 11 (in the case of Figure 1). Additional lines are added if switching is also provided in the reference channel (Figure 2).

Figure 7 illustrates the principle of a delay command. Between a positive voltage V+ and a negative voltage V-, four field effect transistors Ql to Q4 are mounted in a cascade, the first two with a polarity opposed to the second two (01 and Q2 P channel; 03 and 04 N channel).

The two median transistors receive the non-delayed command and apply, at their common output (drain) connected to a capacitor C, a delayed signal to the following stage. The delay defined at the level of the inputs of the grids of the transistors Ql and 04 is controlled by a stage LS.

The control circuit LS acts on the delay by acting on the polarization of Ql and Q4, entailing a modification of the time constant with C. Time delays stored in the memory can thus be applied to amplifier chains, AA1 to AAn, AB1 to ABn and AC1 to ACn respectively which, starting from the signals 45-0, 92-0 and 111-0, supply delayed signals 45-R, 92-R and 111-R serving for the effective actuation of the elements concerned on each occasion.

It is, of course, not necessary to use the test oscillator permanently. It is possible to determine the time delays for a working session which attach to the position of the various filters in the electromagnetic environment in question. The time delays are then stored in the memory and can be applied as they are without any need for placing the test oscillator into action again, unless the working conditions of the device change, or if, for example, an abnormal state is noted at the level of the phase measurements effected.

The second embodiment illustrated in Figure 3 serves at the same time as an interferometer and as a frequency meter.

The reference channel has the same structure as in Figure 1 or Figure 2.

The frequency meter chain has the same structure as the measurement channel in Figure 1 or Figure 2. But its input is switchable so as to receive the same signal as that of the reference channel, either directly (position 0 of the input switch 61), or after it has passed through the delay lines Li and L2 having time delays of different durations.

Finally, the top part (interferometer chain) illustrates a measurement channel similar to that of Figure 1 or of Figure 2, but extended this time to n input measurement signals, 10 instead of two as before.

The expert knows that, by knowing the phase shift associated with several different delays in the same signal, it is possible to gain access to data regarding the frequency of this signal, as described in particular in the unpublished FR-90 01762 already referred to above.

Finally, the embodiment of Figure 4 has a structure similar to that of Figure 3 except that the two measurement channels, associated respectively with the interferametry and the frequency measurement, are now combined into a single one, which makes it possible to obtain a particularly simple device for effecting at the same time the functions of interferometry and of frequency measurement. This device is suitable when the duration of the presence of the signals is sufficiently long, so that one has time to effect the whole set of the switching operations required at the level of the switch 11.

The present invention is not limited in its application to the case of signals obtained by antennas. It may be used in network analyses, in which case the measurement amplification 30 chain will not be brought to saturation, but will remain linear if it is intended to retain the level measurement. The fact that a two-phase modulation is used at the level of the modulator 91 makes it possible to eliminate the false zeros or offsets of the single sideband mixer 19.

It may also be used in interferometry and/or frequency measurements, as has already been described.

For applications requiring fast response times, it is desirable for the switch 11 to be a fast switch made using integrated gallium arsenide circuit technology, so to have a switching time of the order of 2 nanoseconds, that is to say, 20% of the time of scanning each input of the measurement channel. To filter the transitions which may appear at the output of this switch, and which may be situated in the operating band of the amplifiers of the chain, a high-pass filter may be inserted in the chain between the switch 11 and the amplifier 12, or after the amplifier 12.

One of the advantages of the invention is to make such arrangements that making the double mixer 19 with two outputs in quadrature does not require any special performance specifications, since these offsets are compensated by the device in accordance with the present invention, while its errors of quadrature are partly compensated by its principle.

The mixer 91 may be a conventional mixer with elimination of the carrier frequency whose input, to which the 0/180° phase modulation is applied, advantageously has a passing band which extends from the continuous state to approximately 1 GHz. By using a modulation input which does pass the continuous state, it becomes possible to simply switch the phase from 0 to 1800 by polarising the mixer quasi-permanently at 0 or 180° in the case of need, for example for an application where it might be desired to shorten the scanning cycle for processing very short pulses.

The digital circuit 50 is preferably made in the form of a specific circuit of the application, or "ASIC", taking into account the complexity of the necessary calculations, and mainly the speed required for them, the general principle (except the operation of the delays linked to the test oscillator) being known.

In a special application, this unit has the function: - of managing the time sequencing of the elementary measure (analog-digital conversions every ten nanoseconds, for example): - of accurately regulating the instants of the sampling, of the switching and of changing the states of the 15 modulation; - of calculating the phase differences from the measurements effected by the analog-digital converters; and - in the interferometry function, if the frequency of the signal is transmitted to it, or if the frequency meter function is associated, of calculating the arrival direction (since the frequency and the phase are then known).

The calculations are started by the external command indicating the presence of the signal. In the case where this presence is of a duration equal to or greater than one scanning cycle the unit 50 can, of course, effect a calculation of the mean on each phase difference obtained. This use of the mean value function may be actuated by means of the bus illustrated in the drawings under the designation of "control bus".

In the preceding discussion only the case of a 0/1800 modulation has been considered.

The invention is not limited to this case of modulation and may apply to other types of modulations, comprising a subassembly corresponding to a regular subdivision of the trigonometric circle.

Diverse embodiments of this type may be envisaged by going up to a 10 rather high number of the elementary sectors of the trigonometric circle. However, we consider at present that subdivisions into more than eight portions would be difficult to use.

By way of example, reference will now be made to the second page of the attached formulae, which illustrates the case of a 0/120/ 15 2400 subdivision.

The notations used in these formulae are the same as those used before.

Equations 7A to 7C illustrate the trend of the signals Skye, Sb_Re and Sc_Re for the three phase states mentioned above.

The notation is here a little different, by representing the harmonics in the form of a signal sum for an index greater than, or equal to 1.

Similarly, the input signal 8_V1 is defined by equation 8 with the same notation. The equation 9A expresses the continuous component 25 of the single sideband (SSE) for the signal Sa_Re and expresses the latter in a simplified form.

The equations 9B and 9C do the same, in the other two cases, by passing directly to the simplified form.

In the expression 10, the summation is effected of the three 5 signals obtained, by applying the respective phase shifts of 0°, -120° and 240° to them.

A certain number of the products cancel each other because of the properties of the trigonometric functions, with the result that the phase finally obtained is that expressed by the equation 11 which contains no more than the fundamental, the harmonic 4 and the harmonic 7.

When one operates with switching frequencies such that the transient has disappeared at the time of sampling, the devices described in Figures 1 to 4 are sufficient for the proper functioning of the device in accordance with the invention. However, when this condition is not verified, it is preferable to use additional filters.

Thus high-pass filters (not shown in the Figures) may be inserted: - as regards the first, between the switch 11 and the amplifier 12, as regards the second, between the mixer 91 and the amplifier 95.

The function of these filters is to largely eliminate the 25 transient phenomena due to the changes in state of the switch, and to the change of state of the modulation 0/w in the modulator.

Although the invention concerns the measurement of the phase of an electromagnetic signal received after a radio-electric propagation, or one which has simply been acquired at various points of an electronic circuit, the application to signal processing may also be envisaged, and more particularly to the processing concerning phase or time.

Moreover, as has already been indicated, one is not restricted to the division of the trigonometric circle along two or three sectors. It may be decided to divide it into as many as four or five sectors or even into more than eight sectors if this is required by the application.

Annex Formulae -1 (2) A_F = 2.n.F-t (2A) Sa_Re = Reill * exp(j.A_F) + ReH2 * exp(j.2.A_F) (22) Sb_Re = Real * exp[j.(A_F+n)] + ReH2 * exp(j.(2*A_F+2-n)] (3) S_V1 = V1H2 * expj-j.(A_F+Phi_V1H/)] + V1H2 * exp(-j.(-2-A_F+Fhi_V1H2)] (4A) Sa_Ssb = S_OFF*exp(j-Phi_OFF) + Sa_Re * S_V1 = S_OFF*exp(j*Fhi_OFF) + exp(-j-Phi_V1111)-ReH1iV1H1 + exp(-jaPhi_V1H2).ReH2.V1H2 (43) Sb_Ssb = Sb_Re * S_V1 + S_OFF * exp(j*Phi_OFF) = S_OFF*exp(j*Phi_OFF) -exp(-j*Phi_V1H1).ReH1*V1H1 + exp(-j*Phi_V1112).ReH2.V1H2 (5) Sa_Ssb + exp(j.n).Sb_Ssb = Sa_Ssb -Sb_Ssb = 2 * exp(-j*Phi_V1H1)-ReH1sV1H1 (6A) 2. ReHI.V1Hi.sin(Phi_V1H1) (68) 2. ReHI.V1HI.cos(Phi_V1H/) (6C) Phase Vihl =atanr -2 ReH V1H sin(Phi V1H) 2.Re111.172HI.cos(Phi_V1H0 (7A) Sa_Re = Annex Formulae -2 (73) (7C) Sb_Re = (8) Sc Re = (9A) s_yi = Sa_Ssb = E ReH *exp(j.n*A_F) (nal) n E ReH *exp(j.n.A_F)-exp(n.j.2.n/3) (nal) a E ReH *exp(j*n*A_F).exp(n.j.4.n/3) (nal) n V1H *exp(-j.k.A_F).exp(-j*Phi_V1Hk) Sa_Re * S_V1 + HO*exp(j*Phi_HO) = HO*exp(j*Phi_HO) + E ReHnsexp(j.n.A F) (nal) +(Li)viiik*exp(-j.k.A_F).exp(-j*Phi_V1Hk) = HO*exp(j*Phi_HO) +(&1)SV1Hn*exp(-j*Phi_SV1Hn) (93) Sb_Ssb = HO*exp(j*Phi_HO) +(al)SV1Hrieexp(-j*Phi_SV1HOsexp(nej.2.n/3) (9C) Sc_Ssb = HOsexp(j*Phi_HO) +( Enal) SV1Hicexp(-j*Phi-SV1Hn).exp(n.j-4.n/3) (10) Sa_Ssb + exp(-j-2.n/3)-Sb_Ssb + exp(-j.4.n/3).Sc_Ssb = HOsexp(5*Phi_H0).(1+exp(-j.2-7t/3)+exp(-j.4-a/3)) + E SV1H sexp(-j*PhiSV1Hn).(1+exp(j.2.(n-1).n/3)+ (nal) n exp(j-44.(n-1)-1T/3)) (11) Phase_V1 = Arg( V1H1*exp(-j*Phi_V1H1) + V1Heexp(-j*Phi_V1H4) + V1H7.exp(-j*Phi_V1H7) )

Claims (16)

1. Claims 1. A phase meter-receiver for ultrahigh wideband frequency signals of the type comprising: - a reference input and at least one measurement input for receiving wideband ultrahigh frequency signals coming from a propagation of electromagnetic waves and picked up by at least one antenna or aerial, - a high frequency receiving channel for the reference input, - at least one high frequency receiving channel for the measurement input, - mixing means capable of mixing the output Signals of the measurement channel with the output signal of the reference channel, the output of these mixing means containing information regarding the phase difference between the signal coming from the reference input and the signal coming from the measurement input, and means for processing the signals coming from these mixing means, comprising an analog-digital conversion of the 20 signals with two components representing the phase of the measurement signals; wherein the outputs of the receiving channels have the same frequency as their inputs; and wherein, before being applied to the mixing means, the output of the reference channel is subjected to a phase modulation by a signal with several states switched at a chosen time sequence.
2. 2. A device according to claim 1, wherein the signal with several states is switched at a time sequence linked to the sampling time sequence of the analog-digital conversion.
3. 3. A device according to claim 2, wherein the signal with 5 several states is switched in synchronization with the sampling of the analog-digital conversion.
4. 4. A device according to any one of claims 1 to 3, wherein the signal with several states comprises two states respectively representing phase shifts of 0° and 1800.
5. 5. A device according to any one of claims 1 to 3, wherein the signal with several states comprises three states respectively representing phase shifts of 00, 120° and 240°.
6. 6. A device according to any one of the preceding claims, wherein at least one of the receiving channels comprises 15 switchable filters.
7. 7. A device according to any one of the preceding claims, wherein the reference channel comprises switchable filters.
8. 8. A device according to any one of the preceding claims, wherein the measurement receiving channel processes signals 20 coming from several measurement inputs in a way multiplexed in time, by means of a single physical channel.
9. 9. A device according to any one of the preceding claims, wherein it comprises a test oscillator whose signal is applied on command to the receiving channel and to the 25 reference channel.
10. 10. A device according to any one of the preceding claims, wherein at least some of the inputs of the receiving channels come from the same aerial as the input of the reference channel, but after the application of different delays which permits a frequency measurement.
11. 11. A device according to any one of the preceding claims, wherein at least some of the inputs of the receiving channels come from respective aerials that are different from the aerial from which the input of the reference channel is originating, which permits an interferometry.
12. 12. A device according to any one of the preceding claims, wherein the reference high frequency receiving channel is linear as far as its modulator, while the measurement receiving channel or channels comprise a limiter amplifier.
13. 13. A method of processing of wideband ultrahigh frequency 15 signals coming from a propagation of electromagnetic waves of the type wherein: a) there is a reference receiving channel, and at least one measurement receiving channel, the output signals of the measurement channels are 20 selectively mixed with the output signal of the.reference channel, and c) the signals coming from this mixing are processed, after an analog-digital conversion of these signals, in the form of two components representing the phase of the measurement signals relative to the reference; d) the reference and measurement receiving channels are not subjected to any change of frequency; and e) before being used for the mixing, the output of the reference channel is subjected to a phase modulation by a signal with several states switched at a chosen time sequence.
14. 14. A method according to claim 13, wherein the states of the phase modulation are defined by the limits taken modulo 1800 of a regular subdivision of the trigonometric circle into P parts, with P being at most approximately equal to 8.
15. 15. A phase meter-reservoir for UHF wide band frequency 10 signals constructed and adapted to operate substantially as hereinbefore described with reference to, and as illustrated in, the accompanying drawings.
16. 16. A method of processing wide band UHF signals substantially as hereinbefore described with reference to the 15 accompanying drawings.