FR2725863A1 - DIGITAL SIGNAL DEMODULATOR BIDIMENTIONAL MODULES - Google Patents

Abstract

a) The invention relates to a demodulator for two-dimensional modulated digital signals. b) The demodulator comprises a zone decision circuit which fixes the exclusion zones by decision thresholds in the plane of the value of the two-dimensional signal; the decision maker receiving a signal detected in his exclusion zones, sends an acquisition signal indicating to the regulation circuit that they are excluded. Excluded Zones (LO) are designed to be unaffected by digital receive values, which move circular paths (SS) around signal set values due to combined sinusoidal spurious signals. Sinusoidal parasitic signals prevent switching to acquisition mode when the regulation circuits are not yet hooked up.

Description

I

  "Modulated digital signal demodulator, bidimention-

  The present invention relates to a two-dimensional modulated digital signal demodulator comprising a domain decision-maker that defines exclusion zones.

  by threshold decisions in the bidimensional value

  signals, the domain decision maker indicating a

  acquisition signal when it detects

  received in such exclusion areas, for the purpose of

  signal to the existing regulatory circuit that they are ex-

  cluded. Such a demodulator is for example described in documents DE-41 00 099 Cl or IEEE Transactions on

  Communications, Vol, COM-30, No. 10, Oct. 1982, S. 2385-

  2390. In both documents, this is the optimal placement of

  mum of exclusion zones that initiate acquired signals

  for the control circuits excluded from the

  demodulator pass quickly and surely to the hanging.

  A demodulator for two-dimensional modulated digital signals (such as signals

  quadrature modulation (QAM signals) generally includes

  several control circuits coupled to each other

  others, namely to regulate the level of reception, the

  recording of the clock and carrier phase of the

  input signals, the phase quadrature of the carrier frequency possibly also the coefficients of an adaptive correction circuit. There are two different operating states for the control circuits, namely the acquisition mode and the tracking mode. The acquisition mode ensures at the beginning of a data transmission or after a failure the hooking of the control circuit. Then, we switch the mode of pursuit in which the

  Regulatory circuits only regulate small differences

  these. In any case, the control circuits require precise information to know when they must operate in acquisition mode or in

  pursuit. The decision of the operating state to take

  account at this time, depends on the areas in which-

  which the values of the reception signals arrive in the

  plan of the two-dimensional values of the signal. When

  receiving signal values, for example in the immediate vicinity of the signal reference values, this means that the control circuits are hooked and

operate in tracking mode.

  When reception signal values are found in the areas of the signal value plane, which areas are further from the signal reference values, the mode operation should be initiated.

  for control circuits.

  However, it sometimes happens that signal values of

  ception occur in the vicinity of setpoint values where the tracking mode would be required but that combined spurious signals cause the values of the receive signals to be output out of the zones corresponding to the mode

  in the plane of the signal values and then

  if go into acquisition mode although the circuits of

  regulation are not really off the hook.

  The present invention aims to create a demodulator corresponding to the type defined above, which

  avoid as much as possible switching to the mode of

  quisition when the control circuits are still hooked. This problem is solved according to the invention by the features of claim 1. An embodiment

  Advantageous of the invention follows from the subclaims.

  For example, according to the invention, signals

  sinusoidal rasps combined with the signal values of

  ception cause the regulator circuits of the demodulator to switch to acquisition mode even though they are not

unhooked.

  An exemplary embodiment of the invention will be

  described below using the drawing.

  The single figure shows the plan of values of the

  signal for example in a QAM modulation system, 16.

  The 16 setpoints of the signal located in this plane

  are characterized by points. Each value of

  The signal generator consists of a component in phase XI and

of a quadrature component XQ.

  The operation of a zone decision maker will be

  explained using the signal values plan, represent-

  Tees. The decision maker must decide which value of the reference signal is associated with a reception signal value and in which operating state (acquisition mode or

  tracking mode) each time the

  cooked regulator of the demodulator.

  The equidistant decision thresholds SI and SQ parallel to the axis in phase I and the axis in quadrature Q, subdivide the whole plane of the values of the signal into fields whose center corresponds to a set value of the signal. Depending on the field in which the value of the reception signal falls, the zone decision-maker associates the value of the reception signal with the value of the reference signal of

this field.

  In addition, other decision thresholds delimit several LO exclusion zones (shaded areas with dots). If reception signal values are in such exclusion zones LO, the zone decision maker

  points out that control circuits are not

  that they must switch to operation in acquisition mode. The exclusion zones LO are placed according to the following criterion:

  When a reception signal value has

  dye a set signal value and that at this value

  reception signal is combined with a spurious signal,

  nusoidal, it moves on an SS circle around the va-

  their signal setpoint. The diameter of this circle SS predetermined by the amplitude of the sinusoidal parasitic signal must not exceed as much as possible the decision thresholds SI and SQ delimiting the value of the reference signal, so

  in fact not to cut it (see figure). For a trip

  their signal of reception to which is combined a signal

  sinusoidal parasite up to an amplitude at the maximum

  indicated, one can keep the tracking mode. Only when the amplitude of the spurious sinusoidal signal exceeds this maximum value must be switched to acquisition mode. For this purpose, the LO exclusion zones must include only those areas of the signal value plan that

  are not cut by SS circles with the maximum radius

  mum allowed. We thus obtain the exclusion zones represented

  shown in the figure, in the intersection areas of

  SI and SQ decision thresholds and outside the constella-

  setpoint values of the signal. For LO exclusion zones to be as close as possible to the SS circle, they must be delimited by decision thresholds that are partly parallel and partly inclined at 45 relative to the axis in phase I and the axis. in quadrature Q. The zone decision maker generates a signal defined as follows: for (| eI + eQj> emax) V (XI> Xmax) V (XQ> Xmax) G = exclusion zone

I0 for other cases.

  The symbol V represents the logical OR combination.

  XI is the quadrature component of the value of the received, detected signal.

XQ is its quadrature component.

  If we define as transmission alphabet in QAM mode, the values (1, 3 ...), we obtain for m-QAM, the constellation boundary Xmax = \, which, for the

16-QAM system, gives Xmax = 4.

  eI and eQ are the differences of the components in phase and

  in quadrature of the value of a reception signal

  relative to the corresponding components of the value of the decision command signal

  emax a dimension parameter; it defines the extension

  exclusion zones. It can at the same time be sized for holding

  maximum noise level to the pa-

  sinusoidal rasites (emax <2.02 = 0.58), as well as for maximum robustness compared to a reception level tainted with errors Court 16 QAM: emax> 20- 4.0, 0 = 0.47)

  The zone decision maker averages several

  it generates G signals and it finally compares the average signal G to a predetermined threshold GS. For G> Gs, the

  zone decoder emits an acquisition signal. As long

  time that the inequality G <Gs is satisfied, the decision-maker of

zone indicates the tracking mode.

  We need to size Gs to make it possible

  safe between the hooked state and the non-hooked state

  and at the same time the switching point is

  produce for error rates as high as possible.

  For 16-QAM, a non-hooked control circuit gives G = 0.25. Half of the value gives an orientation for the threshold Gs = 0.125. This dimensioning gives the point of

  switching between the tracking mode and the acquisition mode.

  sition, an error rate of about 1.5. 10. This dimensioning gives the switching point between the tracking mode and the acquisition mode an error rate of

  1.5. 10-2 which is a sufficiently large value.

  The zone decision maker is advantageously a

  moire (PROM) whose memory locations include the entire plane of the signal values; the memory locations associated with the exclusion zones contain the

signal G = 1.

Claims (2)

R E V E N D I C A T I 0 N S   1) Two-dimensional modulated digital signal demodulator comprising an operating decision-maker which sets exclusion zones by decision thresholds in the plane of the two-dimensional values of the signal, the zone decoder when it detects received signal values in such exclusion zones, transmits an acquisition signal which signals to the existing control circuit that they are excluded, characterized in that the excluded zones   (LO) are located and dimensioned so as not to be concerned   by the values of the received signals moving on circular paths (SS) around the   setpoint of the signal, because of sinusoidal daux.   2) The demodulator according to claim 1,   characterized in that the zone decision maker transmits a signal (G) defined as follows: lr for (eI + I eQ> emax) V (XI |> Xmax) V (IXQI> Xmax) G = _ exclusion zone [o for other cases (logical V: OR) in this relation:   XI represents the component in phase and XQ the component   in quadrature of a received signal value,   eI and eQ represent the differences between   in phase and in quadrature with a value of   received in relation to the components of the decision set signal value,   Xmax = -, (m is the number of possible values   setpoint signal) and emax <0.58, the de-   zone coefficient averaging several signals (G) and comparing the average signal (G) with a predetermined threshold (GS) and in the case   (G)> (Gs) it transmits an acquisition signal.