FR2729763A1 - BROADBAND PROGRAMMABLE HYPERFREQUENCY RECEIVER - Google Patents

Abstract

This wideband programmable microwave receiver comprises a transmission line (1) traversed in opposite directions by two signals of the same frequency, of which we are interested in the frequency and / or phase shift, of the detectors (Dro, ..., DrN -1) which are regularly distributed along the transmission line (1) and which carry out a spatial sampling of the standing wave regime established on the transmission line (1), programmable multipliers (20-24) affecting coefficients weighting the amplitudes of the output signals of the detectors (Dro, ..., DrN-1), a multiplier programming unit (MN01) and an adder (MA01) taking the sum of the weighted output signals of the detectors to generate a output signal of the receiver, performing the function programmed by the coefficients. This receiver can be made adaptive.

Description

HYPERPREQUENCY RECEPTOR

  The present invention relates to a broadband programmable microwave receiver, in particular for making a frequency meter or a phasemeter, and operating from a spatial sampling of a standing wave regime envelope developing in a transmission line. BACKGROUND OF THE INVENTION traveled in

  opposite directions by the signal (s) to be measured.

  Known frequency meters and phasemeters operating from a spatial sampling of a standing wave envelope envelope developing along a transmission line perform the calculation of the Fourier transform of the envelope of the wave regime stationary and look for either the significant maximums of the Fourier transform module for frequency measurement or the transform argument

  of Fourier at these points for phase measurement.

  If N is the number of detectors placed on the transmission line to perform the spatial sampling, the development of the Fourier transform is done either on N real points for a frequency meter or on N / 2 complex points for a phasemeter which results in a large amount of material for applications or the processing time must be very

  short, and by a fixed and unmodifiable architecture.

  The object of the present invention is to reduce the amount of material required for a discrete Fourier transform spectral analysis and also of

  to make the receiver easily programmable and adaptive.

  It relates to a broadband microwave receiver comprising: a transmission line traversed in opposite directions by two signals of the same frequency, whose frequency and / or phase shift are to be measured, generating a standing wave regime, - detectors distributed along the transmission line sampling the amplitude of the envelope of the standing wave regime established within the transmission line, - programmable weighting means affecting weighting coefficients the amplitudes of the output signals of the detectors a programming unit of the weighting means and an adder performing the sum of the weighted output signals of the detectors for generating a signal

output of the receiver.

  Other features and benefits of

  the invention will emerge from the following description of

  several embodiments given by way of example.

  This description will be made against the drawing in

  which: - a figure 1, shows a diagram of a broadband microwave receiver of the prior art, - a figure 2, shows a diagram of a broadband microwave receiver according to the invention adapted for use in phasemeter FIG. 3 represents a diagram of a broadband microwave receiver similar to that of FIG. 2 but adapted for frequency meter use; FIG. 4 represents a variant of FIGS.

  microwave and broadband receiver schemes

  FIGS. 2 and 3, and FIG. 5 represents the schematic diagram.

an adaptive receiver.

  FIG. 1 shows a transmission line 1 receiving at one end 2 a microwave signal of the form Eo exp (jwt) and at the other end 3 a microwave signal of the form Eo exp (jwt + P). Taking the end 2 for origin, the voltage E (x) at any point on the remote line of x of the end 2 is the sum of a component due to the signal Eo exp (jwt) expressed by: Eo exp jw (t - x) v

  v being the speed of the waves in the transmission line

  sion, and a component due to the signal Eo exp (jwt + p) expressed by: Eo exp j (wt - wL + wx + p v v L being the length of the transmission line, so

  that the voltage E (x) at each point of the transmission line

  mission is expressed by: E (x) = Eo exp (jwt) [exp (_jW x) + exp (jWx - jwL + jp)] vvv or else E (x) = 2 Eo cos (wx wL +) expj ( wt - wL +) v 2v 2 2v 2 N probes are regularly distributed along the transmission line 1 from its ends. They perform a quadratic detection of the standing wave regime established on the transmission line i and detect an envelope V (x) which has the value: V (x) = 4 Eo2 cos2 (wx _ wL + -) v 2v 2 or still V (x) = 2 Eo2 [1 + cos (2W (x-L) + p)] v 2 These N probes each consist of a quadratic detector DrO, Drl, ... DrN-l coupled to the line transmission 1 for example by a capacitor not shown and followed by a low-pass filter FI0, FIl, ... FIN-l. They deliver a sampling V (k) of the envelope V (x) which, assuming that they are numbered from O to N-1 starting from the end 2 of the transmission line 1 taken for origin of the distance x, that is to say that one has the relations x = k A x and L = (N-1) A x is of the form V (k) = 2 Eo2 [1 + cos (2w Ax (k - N-1) + p)] v 2

for the k c (0, ..., N-1).

  Taking into account that the spacing

  unit A x probes along the transmission line

  sion 1 corresponds, to satisfy the exchange theorem

  at least one-half of the minimum wavelength of the detected quadratic envelope, that is to say one quarter of the wavelength corresponding to the maximum frequency of the input signals Ax = m = v 4 4fmax sampling of the envelope V (k) is expressed in the normalized form: V (k) = 2 Eo2 [l + cos (<f (k - Nl) + p)] f max 2

  It is possible to reach by this sample-

  the frequency f and phase shift p of the signals

  at the ends of the transmission line. This is

  in the usual way, through

  of functions defined from weighted sums

  points of the discrete Fourier transform computed from this sampling. To do this, the detectors outputs are connected to a Fourier transformer 10 followed by a multiplier array Srl, Sil, Srm, Sim and a MAO1 summator. The Fourier transformer 10 delivers on 2m outputs of the voltages Re (fl) to Re (fm) and Im (fl) to Im (fm) representing a sampling

  real and imaginary parts of the frequency spectrum

  quences of the quadratic envelope V (x). The bench

  multipliers Srl, Sil, ..., Srm, Sim affect the

  The outputs of the Fourier transformer 10 have weighting coefficients cr1, c1 1, ... crm, cim so as to produce at the output of the adder MA01 a function Z of the form: IZ = E [crk Re (fk) + cik Im (fk)] (1) k = l In the case where the weighting coefficients are quantized to 0 or 1, the Srl to Sim multipliers are only controllable switches and the function

  performed is that of a phasemeter programmable bandwidth in a discrete manner.

  Noting that the output voltages of the Fourier transformer 10 are linear combinations.

  res of the output voltages Vn of the detectors: A Re (fk) = E Vn cos (2r k p (n)) (2) n N Im (fk) = - E Vn sin (2w k p (n)) (3)

  n N p (n) being a function representing the normal position

  of the detector n and taking the values: - N-1 to + N-1, in phasemeter with a reference to the center

2 2

  of the propagation line, it is conceivable that it is possible to directly obtain the function Z from a weighted sum of the output voltages Vn of the detectors. The weighting coefficients applied C'0, ..., C'N-1 are deduced from the coefficients crl,... Cim and the voltages of the outputs of the Fourier transformer. Indeed the relation of definition (1) of the function Z is expressed considering the relations (2) and (3) by: Z = E [crk E Vn cos (2a kp (n)) kn N - cik E Vn sin (2w kp (n))] n N which is written again by permuting the summations on n and k Z = E Vn E [crk cos (2r kp (n)) nk N - cik sin (2 X kp ( n))] N by identification with: Z = E Vn C'n n we get c'n = E [crk cos (2r kp (n)) k N - cik sin (2w kp (n))] N The coefficients c'n can be both positive and negative while the output voltages Vn of the detectors are either all positive or all negative depending on the direction of the diodes Dro, DrN-1 so that it is possible to perform the weights output voltages Vn of the detectors by the coefficients

  using two-quadrant multipliers.

  The Z function is thus performed directly by a weighted summation of the output signals of the detectors. Some applications require the implementation of several distinct Z functions. If the operating time is critical, these different Z functions will be performed using several weighted summation systems connected in parallel to the outputs of the detectors rather than successively using a single summation system whose sequential

weights.

  For example, in a phasemeter application it will be wise to use two weighted summation systems, one developing a signal proportional to the cosine of the phase to be measured (or "UREEL" system) and the other a signal proportional to the sinus of phase to be measured (or "imaginary" system) .These two channels will then be digitized and a digital system will extract the

phase to be measured.

  By planning to vary the Z functions by programming their weighting coefficients, it is possible to envisage numerous applications, in particular: a programmable receiver whose performances (sensitivity, accuracy, etc.) are optimized for a few precise signals (characterized by example by their frequency); improvement of the measurement accuracy by gradual shrinking of the microwave band around the information to be measured (frequency adaptivity); - regulation of the operating threshold according to the spectral characteristics of noise or spurious signals; - Rejection of one or more undesirable signals by modulation of the response curves; operating in the manner of a heterodyne receiver by scanning all or part of the microwave band. In the previously mentioned case of a phasemeter, it is possible, for example, to make the phasemeter insensitive to a given frequency signal by generating, by a suitable choice of the coefficients of the two real and imaginary weighted summation systems, a hole of sensitivity at the desired frequency. in the microwave bandwidth of the phasemeter, the frequency position of this sensitivity hole can be displaced by a change

programmed said coefficients.

  FIG. 2 illustrates a broadband microwave receiver circuit according to the invention in which the function Z is obtained directly from a weighted sum of the output voltages Vn of the detectors produced by means of two quadrant programmable multipliers and

a summative montage.

  FIG. 2 shows the transmission line 1 receiving at one end 2 a microwave signal of the form E0 exp (jwt) and at the other end 3 a microwave signal of the form E exp (wt + p) and the probes regularly. distributed along this transmission line 1 consisting of quadratic detectors DrO, ... DrN-1 followed by filters

low pass FIo, ... FIN_1.

  The outputs of the N probes are individually connected

  to the analog inputs of N multipliers two quadrants 20, 21, 22, 23, 24 which are realized

  conventionally using digital converters-

  analogous CNA and which receive on their digital inputs the values of weighting coefficients C'O, C'N-1 provided in digital form by a programming unit MNO1. The MXO-to-MXN-1 analog outputs of the 20 to 24 digital to analog converters on which the weighted voltages of the detector outputs are available are connected by a resistor network ROO to RON-1 of the same values at the input of an amplifier. MAO1 which is provided with a resistor R10 placed in counter-reaction and which delivers a voltage corresponding to the sum of the weighted output voltages

  detectors with a coefficient - R10 / ROO.

  FIG. 3 illustrates a broadband microwave receiver circuit similar to that of FIG. 2, adapted for use as a frequency meter with a transmission line 1 short-circuited at its second end 3, which makes it possible to be interested only in to the

  frequency and at the input signal level.

  FIG. 4 illustrates a variant of the broadband microwave receiver of FIG. 3 using quadrant multipliers to weight the voltages of FIG.

output of the detectors.

  In order to be able to use multipliers one quadrant, the output voltages Vn of the detectors must have the same sign and the same for the weighting coefficients. We have seen that this was the case of the output voltages of the detectors including the DrO diodes,

  ., DrN-1 are all connected in the same direction. As regards the weighting coefficients, it is noted that the summation of the function Z can be put in the form: Z = E (c'n + X) Vn - XE Vn nn with a parameter X chosen so that the coefficients c'n + X are all of the same sign regardless of n. This makes it possible to adopt a new series of weighting coefficients: C "n = C'n + X all of the same sign, provided that the sum of the voltages of the detectors thus weighted is subtracted from the term..DTD: X E Vn.

  n is distinguished in FIG. 4 of a quadrant digital-to-analog converters 30, 31, 32, 33 which weight the output voltages Vn of the detectors by the weighting coefficients C "O,... CnN-1 and which have current outputs MX'O, .... XX'N-1 summed by an amplifier MA'01 provided with a resistor R'10 placed in counter-reaction A resistor network R'OO, ... R ' ON-1 connects the outputs of the detectors to the input of an amplifier MA02 which is provided with a resistor R11 placed in feedback and

  which delivers a voltage representing the term - X nEVn.

  A resistor R12 interposed at the output of the amplifier MA02 transforms the output voltage of the latter into a current added to those of the outputs of the different digital-analog multipliers to a quadrant 30, 31, 32, 33 to give an output of the amplifier MA'01 a voltage proportional to the value of the desired Z function. The programming unit MN01 which provides the

  weighting coefficients C'n or C "N may be con-

  a set of memory registers inscribed once and for all with invariable values of coefficients or else programmed via a microprocessor with values of coefficients varying according to a particular law as a function of the value of the output signal of the amplifier MA01 or MA'01 which

makes the receiver adaptive.

  FIG. 5 illustrates an adaptive receiver scheme where the output is re-injected into the programming unit which modifies the weighting coefficients

  according to a given reference law.

REVEUDICATIONS

  1. Broadband programmable microwave receiver characterized in that it comprises: a transmission line (1) traversed in opposite directions by two signals of the same frequency, the frequency and / or phase of which are measured, generating a regime stationary waves; detectors (Dro, ... DrN-1) distributed along the transmission line (1) sampling the amplitude of the envelope of the standing wave regime established within the transmission line (1); programmable weighting means (20, 21, 22, 23, 24) assigning weighting coefficients to the amplitudes of the output signals of the detectors (Dro, ... DrN-1); a programming unit (MNO1) of programmable weighting means (20, 21, 22, 23, 24) and a first summator (MA01) performing the sum of the weighted output signals of the detectors (Dro,

  DrN-1) to generate an output signal of the receiver.

tor.

  Receiver according to claim 1, characterized

  characterized in that the programmable weighting means are digital-to-analog converters (20, 21, 22,

23, 24).

  Receiver according to claim 1, characterized

  in that the programmable weighting means

  are two quadrant multipliers (20, 21, 22, 23, 24).

  Receiver according to claim 1, characterized

  characterized in that the programmable weighting means are quadrant multipliers (30, 31, 32, 33) and in that it further comprises a second summator (MA02) which is connected by a resistor network of equal value ( R'00, ..., R'ON-1) at the outputs of the detectors and which delivers a sum of the output signals of the detectors weighted by a parameter X subtracted by the first summator (MA'01) to the sum of the signals of weighted output of the detectors, the parameter X having a value such that the weighting coefficients (C "n) by which the multipliers (30, 31, 32, 33) multiply the amplitudes of the output signals of the detectors are

all of the same sign.

  5. Receiver according to claim 1, characterized in that it is adaptive, the programming unit (MN01) receiving among its control signals the output signal of the first summator (MA01)

  which reacts on the programming of the coefficients.