FR2638306A1 - Receiver - Google Patents

Abstract

The invention relates principally to a receiver. The invention relates principally to a receiver capable of separating, demodulating and decoding modulated wave transmissions using long pulses on a plurality of frequencies 31, 32, 33, ... 3N. Advantageously, each channel is orthogonal to the others, two consecutive frequencies being 1/T apart, T being the duration of each useful transmission interval. The invention applies to the reception of radiophone and television transmissions, to the devices for analog and/or digital information reception, to communications between computers, to telephone communications between exchanges, to telephone communications between radio telephones and communications stations, to radio communications between earth stations and satellites, to communications between satellites, to acoustic communications in air and/or in water and to the production of local computer networks. The invention is particularly well suited to the production of high fidelity radiophone receivers as well as to the production of high definition television (HDTV) receivers and/or digital television receivers.

Description

RECEIVER
The invention relates mainly to a receiver.

 It is known to transmit information using modulated waves, such as for example electromagnetic waves. It is known to try to increase the data rates transmitted. However, increasing the speed most often involves increasing the frequency band used. However, in the case of guided transmissions, one is limited by the bandwidth of the channel, for example a coaxial target or an optical fiber, while in aerial transmissions there are no longer enough frequencies to meet all needs.

On the other hand, patent application FR 86 09622 published under No. 2 601 210 describes a transmission method using symbols A (f, t). Each symbol corresponding to a given frequency and transmission time. Since the transmission time is not fixed, the device does not use fine synchronization means, which limits the spectral efficiency to approximately 0.7 bit / (Hz.s). On the other hand the French addition certificate 86 13271 published under the number 2 604 316 describes the use of devices for calculating the Transformed
Discrete Fourier for signal demodulation. This add-on certificate also suggests the use of a guard period between the symbols. However, the non-orthogonality of the channels used limits the spectral efficiency in the best case to 1 blt / (Hz.s).

 With the device according to the present invention, it is possible to exceed, under analogous conditions, 5 bitsSHz.

The present invention relates to a receiver adapted to transmission means and methods described in the
Patent applications FR 86 13937, FR 86 13938, FR 86 13939, FR 86 13940, FR 86 13941, FR 86 18351, FR 86 18352.

 In devices of known type, attempts have often been made to increase the data rate by reducing the transmission times allocated to each information unit (or by increasing the number of possible symbols). Thus, a broad spectrum was generated which, on transmission, had to filter the secondary lobes, which created a distortion of the signal. For a square signal of duration T we obtain a spectrum with many secondary lobes; the main lobe has a width of 2 / T. We will call in the rest of this patent self-distortion this distortion of the signal.

 The device according to the present invention has the originality of reducing or eliminating the self-distortion of the signal by using long transmission intervals for the information elements (often called symbols) to be transmitted. To obtain a high throughput, a plurality of pieces of information are transmitted simultaneously using orthogonal channels. A piece of information is, for example, a 6-bit digital word. Advantageously, one element of information is transmitted per transmission channel.

The orthogonality at the reception of the transmission channels allows the separation of the information belonging to various channels. The orthogonallsation on reception of the various channels comes from a choice of several modulation frequencies spaced regularly by k / T, k is a natural integer I and T the duration of the useful transmission interval. This type of transmission presupposes synchronous sampling on reception to obtain channel separation.

The main object of the invention is a receiver comprising sampling means synchronous with the signal, characterized in that it comprises means for demodulating a transmission of modulated waves using symbols transmitted for a duration T | + T on a plurality of frequencies, two emission frequencies being 1 / T apart,
T being the useful transmission interval and AT being the transition interval and that it includes a servo device ensuring synchronization of the receiver with the received signal.

 The invention also relates to a receiver characterized in that it comprises an automatic gain control device (AGC) controlled by a device for detecting the average power of at least part of the signal.

 The invention also relates to a receiver characterized in that it includes means for decoding a torque (amplitude, phase) or (real part, imaginary part) to convert them into digital words.

 The invention also relates to a receiver characterized by the fact that it comprises at least one device for calculating the fast Fourier transform (FFT).

 The subject of the invention is also a receiver, characterized in that it includes a test device capable of providing amplitudes and / or reference phases from calibration signals.

 The subject of the invention is also a receiver, characterized in that it includes an equalization device compensating for the disturbances of the signal originating from the transmission.

 The invention also relates to a receiver characterized in that it includes reorthogonalisation means using the transition interval of duration T to make a plurality of channels orthogonal.

 The invention also relates to a receiver characterized by the fact that said receiver is a receiver of radio transmissions.

 The invention also relates to a receiver characterized by the fact that said receiver is a receiver of television broadcasts.

 The invention also relates to a method characterized in that the modulated waves are electromagnetic waves.

 The invention also relates to a method characterized in that the step of separating N orthogonal channels comprises a step of calculating the fast Fourier transform (FFT) of the signal.

 The invention also relates to a method characterized in that it comprises a step of reconstituting a television signal from the signals received in them. N channels.

The invention will be better understood by means of the description below and the appended figures given as nonlimiting examples among which
- Figure 1 is a diagram illustrating the phenomenon of broadening of spectrum
- Figure 2 is a diagram illustrating the transmission of a carrier frequency
- Figure 3 is a diagram explaining the operating principle of the device according to the present invention
- Figure 4 is a diagram explaining the operating principle of the device according to the present invention
- Figure 5 is a diagram explaining the operating principle of the device according to the present invention
- Figure 6 includes timing diagrams illustrating the sequence of periods or intervals of emission
- Figure 7 is a diagram of an example of coding capable of being implemented in the device according to the present invention
- 18 Figure 8 is a diagram of an exemplary embodiment of a receiver according to the present invention
- Figure 9 is a diagram of an exemplary embodiment of a television receiver according to the present invention
- Figure 10 is a diagram illustrating an example of equalization capable of being implemented in the device according to the present invention
- Figure 11 is a diagram of an architecture likely c. be implemented in the device according to the present invention
- Figure 12 is a diagram of an embodiment of a detail of the receiver according to the present invention
- Figure 13 is a curve illustrating an example of analog embodiment of the transmitter-receiver synchronization
- Figure 14 is an exemplary embodiment of a device capable of being implemented in the device according to the present invention;
- Figure 15 is a curve illustrating the information rate obtained as a function of the number of coding states for a duration of the useful transmission interval T and the number of channels used given
In FIGS. 1 to 15, the same references have been used to designate the same elements.

 In FIG. 1, we can see a curve 3 representing, on reception, the amplitude A 2 of the spectrum of a wave of constant amplitude emitted during a limited time interval of duration T. Curve 3 has a shape in sin x / x.

 In frequency, in addition to a main lobe, secondary lobes are transmitted which decrease as one moves away from the central frequency f0 The amplitude A passes through zero at two points symmetrical with respect to the frequency f0 referenced 50 and 51. The passages by the zero amplitude are regularly distributed, distant by 1 / T.

 The widening of the frequencies depends mainly on the duration of the pulses emitted. Short t broadcasts cause wider frequency broadening.

In devices of known type, broadening the spectrum with a limited bandwidth allocated to transmissions lengthens the duration of the impulse response of the signal, thereby creating interference between the pulses (called "intersymbol interference"). The amount of separable information was thus limited.

 In FIG. 2a, one can see a curve 6 corresponding to the emission from an instant 16 of a pure sinusoidal frequency. The signal 6 can, for example correspond to a carrier. Curve 6 represents the amplitude as a function of time.

 In FIG. 2b, we can see a curve 7 representing the amplitude as a function of time of the wave 6 received by a receiver. Insofar as the receiver is fixed relative to the transmitter, the received wave 7 has the same frequency as the transmitted wave 6. However, the amplitude and the phase have varied.

In FIG. 2b, the start of reception bears the reference 17. The instant 17 is after the instant 16, the difference corresponding to the propagation time of the waves between the transmitter and the receiver. From an instant 18, the signal 7 has the same appearance as the signal 6. Between instant 17 and instant 18 we assist in the establishment of the signal during which various disturbances are observed. The disturbances in the time interval between instant 17 and instant 18 mainly arise from the distortions caused by the equipment for transmitting and receiving disturbances, due to multiple echoes, as well as from the limitation of the bandwidth of As long as the frequency of signal 6 and signal 7 is retained, it is possible, by performing a calibration of the amplitude and phase, to recover the information transmitted on reception. Certain stable short-term frequency changes, such as the frequency change due to the Doppler effect, for example as a result of the displacement at the receiver relative to the transmitter, can be compensated by an adequate calibration.

 To carry out the calibration we consider for example that everything that happens between the transmitter and the receiver is a filter having a stable frequency response over durations much longer than the duration of the symbols. By transmitting signals of known type, it is possible to determine the frequency response of the filter. Thus, by applying the inverse frequency response, the signal transmitted can be regenerated on reception.

 In Figure 3, we can see a diagram illustrating the operating principle of the device according to the present invention. In FIG. 3, we can see a first curve 31 and a second curve 32 centered on frequencies respectively f0 and f0 + 1 / T, T being the duration of a useful transmission interval. The amplitude A of curve 31 goes through zero at points 51.53.

 The amplitude A of curve 32 goes through zero at points 52.54.

The point 52 corresponds to the maximum amplitude of the curve 31 and to the zero amplitude of the curve 32. A point 311 of the signal at the frequency f is not disturbed by the signal
o corresponding to curve 32.

Likewise, point 53 corresponds to the maximum amplitude of curve 32 and to the zero amplitude of curve 31. At point 53, at frequency f + 1 / T the signal belongs only to the
o curve 32.

 By sampling the spectrum at frequencies f0 and f0 + 1 / T, we can completely separate the frequencies corresponding to curves 31 and 32. We can easily use independent phase amplitude or amplitude / phase states on each of the frequencies f0 and f0 + 1 / T. The two coding being completely independent and separable on reception, it is possible to distribute the total information rate between several channels.

 In devices of known type, to increase the bit rate of information transmitted, the duration of the pulses and / or the time devoted to the transmission of each elementary information was reduced (or by increasing the number of possible symbols).

 On the other hand, in the device according to the present invention, for a given information bit rate, insofar as it is possible to divide the bit rate to be transmitted between several channels, it is possible to increase the duration T of the pulses and / or useful emission intervals corresponding to basic information. This significant flow being obtained by carrying out the sum of the elementary flows corresponding to each frequency. By increasing the duration T of the useful emission intervals, the broadening of the spectrum and the self-distortion of the signal are reduced. It is thus possible, as illustrated in FIG. 4, to use a large number of carriers 31 to 3N. The use of N frequencies 31, 3N allows very good filling of the passband B. As in the case of FIG. 3, successive curves are separated, in frequency of l / T. Thus, the maximum of the spectrum of each channel corresponds to the passage through the zero amplitude of the spectra of all the other channels.

 In FIG. 4, a curve 31 passes through the maximum at a frequency corresponding to point 5 (i + 1) and through the zero amplitude at the frequencies corresponding to points 55 with j) i + l. For the clarity of FIG. 4, only the secondary lobes of the curve 31 have been represented.

 Each transmission channel corresponding to a different carrier frequency carries information independent of the other channels. The total bit rate is equal to the sum of the bit rates of N channels.

 Increasing the number of channels, without decreasing the bit rate, increases the duration T of the useful transmission intervals.

 On the other hand, the increase in the number of channels requires more or better equipment on transmission and on reception.

 For correct operation of the device according to the present invention, it is imperative that the useful part of the received signal is stable. For this, the time intervals of duration AT during which the signal may not be stationary are eliminated on reception. This time mainly corresponds to the impulse responses from the transmitter and the receiver and to the multiple paths. In the remainder of this patent, the transition interval of duration ss T is called the interval during which the signal may not be stationary, useful interval of duration T the stationary part of the signal used by the receiver. The signal is advantageously transmitted during transmission intervals of duration T + AT. Thus, the spectra of each channel although distributed uniformly spaced 1 / T have a width of the main lobe equal to 2 / (T + ss T). On reception, only the useful interval of duration T is used, which makes it possible to reconstruct the spectrum of FIG. 4.

A non-limiting example of bit rate in M blt / s as a function of the number of possible states (that is to say of different symbols capable of being transmitted) is shown in FIG. 18. In addition on the abscissa we have indicated the number of coding bits. For example, coding on 4 bits corresponds to 24 = 16 different states. These curves are given for a
Identical transition interval of duration AT = 8 they.

 A first curve indicates the flow rates for N = 64 and T + AT = 16 zs. .

 A second curve indicates the flow rates obtained for N = 128 and T + A T = 24 s.

A third curve indicates the flow rates obtained for
N = 256 and T + AT = 40 lls.

A fourth curve indicates the flow rates obtained for
N = 512 and T + ss T = 72 lls.

 At a constant transition interval, the fact of increasing the useful part of the signal leads to a saturation phenomenon limiting the flow rate, not shown in FIG. 18.

 The curves in Figure 18 correspond to a bandwidth B of 8 MHz.

 Depending on, in particular the bandwidth available for the application and the propagation conditions, the person skilled in the art will choose the ideal compromise between the number of channels N and the useful transmission interval T.

 the bit rate can be increased within a certain limit by using a transition interval of duration 8 T very small compared to the duration T of the useful emission interval.

It is advantageous to use the fast inverse Fourier transform (FFT 1 in Snglo-Saxon technology) to effect the modulation of the channels on transmission and the
Fast Fourier transform (FFT in English terminology) to perform demodulation at reception.

The use of fast Fourier Transform algorithm requires performing calculations on a number of samples equal to a power of two. During television transmissions, for example 256,512,1024 or 2,048 channels are used.

However, it is not necessary for each channel to transmit information.

 On reception, for each useful transmission interval of duration T, the phase and the amplitude corresponding to each of the frequencies 31 to 3N are advantageously measured. Synchronous sampling is used to extract information from the signal.

 The amplitude representing the information is constant over the entire duration of the transmission interval of duration T or T + b T, the phase representing the information corresponds to the phase shift relative to a phase reference.

A transmitter dedicated to the emission of waves capable of being picked up by the receiver according to the present invention is described in a French patent application filed by the
Applicant, simultaneously with this Application for
Patent and bearing the number immediately below.

 To obtain a significant data rate, it is necessary to be able to distinguish close amplitudes and phases and therefore to have an amplitude and phase reference for each channel. This reference of the amplitude and of the phase is advantageously given by reference signals transmitted periodically by the transmitter to the receiver. The repetition frequency of reference signals depends on the stability of the propagation conditions and the local oscillators.

 Synchronization is maintained by the use of a stable short-term time base.

 In a first variant of reallsatiun of the device according to the present invention, amplitude and phase reference signals are periodically transmitted on all frequencies 31 to 3N of a time interval of duration T or T + AT. However, it should be noted that the frequent emission of calibration signals reduces the rate of useful information transmitted.

 In an advantageous variant of the device according to the present invention, only a few calibration signals are emitted, advantageously distributed regularly among the frequencies 31 to 3N. The coefficients of the other frequencies being determined by calculation by example by Interpolation.

 More generally, it is possible to distribute the calibration signals over time and / or on different channels.

 It is for example possible to periodically transmit test signals, each transmission being carried out on different channels. For example, a circular permutation of the channels assigned to the tests is carried out. The impulse response of the transmission medium is deduced for all the channels, for example by interpolation over time and / or over the frequencies. The matrix of amplitude and phase corrections to be applied to each channel is thus deduced.

 It is essential to compensate by calibrations the variations of the impulse response of the milleu of transmission, for example following a variation (even local) of the atmospheric conditions.

 The impulse response of the medium is determined, for example by calculating the Fourier Transform of the corrections to be made.

 In an exemplary embodiment, one channel in eight was used to calibrate the amplitude A and the phase of all the channels 31 to 3N. In such a type of device it is possible, either to carry out the calibration at each transmission interval of duration T, or as in the case of an embodiment described above, to devote only certain transmission intervals to the calibration . Synchronization is maintained by the use of a stable time base.

The number of channels and / or periods dedicated to the calibration depends on the error which one wants to be able to correct as well as disturbances which are likely to affect the transmission of information. The calibrations should for example be more frequent, to compensate for the displacement of the frequencies by Doppler effect in the case of displacements of
The transmitter relative to the receiver, for example in the case of radiotelephones or communications between aircraft.

 The first and the last channels risk being disturbed in particular by the filters of the transmitter and the receiver. Advantageously, as illustrated in FIG. 5, the first and the last channel are not used for the transmission of information. For example, we do not transmit anything on the first and last channel or we repeat the transmission of the second channel on the first and the penultimate on the last.

On figure 6, one can see various examples of chronograms of sequence of the useful Intervals of emission successive 8
In FIG. 6a, we can see useful transmission intervals 8. Between the useful transmission intervals 8 are found the transition intervals 81 do not show any reduction in transmitted power. In the guard intervals 81, for example, the signal transmitted during the end of the corresponding useful transmission interval is transmitted. The fact of not reducing the transmission power makes it possible to best use the amplifiers of the transmitters.

 In FIG. 6b, we can see successive useful emission intervals 8 which are not separated by transition intervals. This case corresponds to the maximum information rate. It has the disadvantage of poor transmission security in the event of disturbances. This variant is used for example for transmission F by cables.

 In FIG. Gc, one can see a succession of useful emission intervals 8 separated by transition intervals 81 during which the emission of modulated waves is stopped. This saves energy.

 The choice of the type and duration of transition intervals 81 depends on the equipment used, on the expected transmission and reception conditions. For example, if we expect large multiple echoes, it would be in our interest to use longer transition intervals. The length of the transition interval 81 will for example be determined from the worst conditions in which we want to be sure of obtaining a correct reception. For example, if we want to be able to get rid of multiple echoes coming from a maximum distance of 600 meters, a transition interval 81 corresponding to the propagation time of this echo, for example electromagnetic, and possibly time corresponding to the damping of the impulse response of this echo, for example 4 closed, will be used.

In FIG. 7, one can see an example of information coding capable of being implemented in a device according to the present invention. This type of coding has been described in Patent Applications FR 86 13937, FR 86 13938,
FR 86 13939, FR 86 13940, FR 86 13941, FR 86 18351, FR 86 18352. In this type of coding with each digital word is associated an amplitude and a phase in the complex plane. The torque (amplitude, phase) is equivalent to the real and imaginary part of the signal. In the example illustrated the couples (amplitude, phase) 14 are regularly distributed over concentric circles 150, 160, 170 and 180. In the example illustrated in FIG. 7, there are 32 different values which corresponds to a coding on five bits. It is understood that codings on a different number of bits such as for example 2,3,4 or 6 or more do not depart from the scope of the present invention. The size of the discs 13 centered on the points 14 corresponding to the same digital word makes it possible to tolerate a certain imprecision. The larger the diameter of the discs 13, the lower the error rate, but the less different the values can be. In the example shown in FIG. 7, the circles 150,160,170,180 have respective diameters P 1, P 2, P 3 and P 4 equal respectively to 2/2, 1, i and 2, the power of a transmitter being normalized to 1. In the example in figure 7, to reduce the risk of errors on reception, the disks 13 are kept as far away as possible.

Thus on each following circle, the points 14 are placed on the bisector of points 14 of the previous circle.

 It is understood that the arrangement of Figure 7 is given only by way of non-limiting example. For example, the distribution of points 14 on a rectangle or a spiral, for example logarithmic or Archimedean is not beyond the scope of the present invention. Likewise any other type of coding can be used; the type of coding depends on the bit rate and the nature of the information to be transmitted.

 The coding can be analog or digital depending on the desired application.

 In the device according to the present invention, it is possible to carry out the analysis of the impulse response of the transmission medium. Depending on the application, it is possible to use a real-time analysis or a delayed analysis.

 The analysis makes it possible to adapt the transmission standard to local conditions, for example on a local computer or telephone network or on radio-relay systems.

 For example, on a local network it is possible to perform the analysis each time the network is reconfigured. To eliminate the reflections in the cables, parts of the transition interval (of total duration h T) can be placed at the moment when these reflections are the most important.

 On radio-relay systems, for example, a computer is used to analyze the impulse response of the medium in real time and to adapt the transmissions so as to obtain the maximum bit rate allowed by medium disturbances. For example, the computer decreases when possible, without exceeding the allowed error rate, the duration T of the transmission interval. In an alternative embodiment, the computer chooses a transmission means from a plurality of available.

 In FIG. 8, one can see a diagram of an exemplary embodiment of a receiver on the present invention.

The embodiment illustrated in FIG. 8 comprises a reception antenna 40, an amplifier 603, a mixer 41, a bandpass filter 42, a variable gain amplifier 604, a mixer 4817, a low pass filter 4818, an analog-to-digital converter 4819, a reorthogonalization device 482, a demodulation device 48, a local oscillator 250, an automatic gain control device 605. a local oscillator 491, a servo device 49, an analysis circuit 601, a decision circuit 602, a processing device 45, an operating device 46.

 The antenna 40 is connected to the input of the amplifier 603. The output of the amplifier 603 is connected to a first input of the mixer 41. The output of the mixer 41 is connected to the input of the bandpass filter 42 .

The output of the bandpass filter 42 is connected to the input of the amplifier 604. The output of the amplifier 604 is connected on the one hand to a first input of the mixer 4817, and on the other hand, to the input of the automatic gain control circuit 605. The output of the automatic gain control 605 is connected to a gain control input of the amplifier 604. The output of the mixer 4817 is connected to the input of the low-pass filter 4818 The output of the low-pass filter 4818 is connected to the input of the analog-to-digital converter 4819.

The output of the analog digital converter 4819 is connected to the input of the reorthogonaltsation device 482.

The output of the reorthogonalisation device 482 is connected to the input of the demodulation device 48. The output of the demodulation device 48 is connected on the one hand, to the input of the analysis link 601, and on the other hand, at the input of the servo device 49. The output of the analysis circuit 601 is connected to the input of the decision device 602. The output of the decision device 602 is connected to the input of the processing device 45. The output of the information processing device 45 is connected to the input of the operating device 46. A first output of the servo device 49 is connected to the analog digital converter 4819, to the reorthogonallsation device 482 , the demodulation device 48, the analysis circuit 601 and the decision device 602. A second output of the servo device 49 is connected to the local oscillator 491. A third output of the device the servo 49 is connected to the local oscillator 250.

 The antenna 40 receives the high frequency signal from the transmitter.

 The amplifier 603 amplifies the signal picked up by the antenna 40. By beating with a high frequency signal supplied by the local oscillator 250, the mixer 41 lowers the frequency of the received signal.

 The signal is filtered by a filter 42. The filter 42 makes it possible to eliminate signals foreign to the signals which it is desired to be able to receive. The filter 42 is advantageously a surface acoustic wave filter (SAW in English terminology).

 The amplifier 604 performs, under the control of the automatic gain circuit 605, the amplification of the intermediate frequency signal. The automatic gain circuit picks up the signal at the output of the amplifier 604. Integration over a sufficiently long period of time provides the average value of the amplitude of the signal for the production of an amplifier control signal. 604 to optimize reception.

 The mixer 4817 performs beats between a signal supplied by the local oscillator 491, and the signals amplified by the amplifier 604. The mixer 4817 delivers a signal on low carriers. The filter 4818 selects the desired part of the spectrum.

 The 4819 analog-to-digital converter performs digital signal sampling.

 To be able to obtain significant information rates, it is essential to carry out a total separation of the signals belonging to various channels. The reorthogonalisation circuit 482 advantageously makes it possible to eliminate crosstalk between channels. Crosstalk could for example come from multiple echoes which delay part of the signal.

Such signals arrive at the receiver notably during the reception of the following pattern. The reorthogonalisatlon device 482 comprises a motlf change detection circuit.

For example, it includes means for subtracting the signal with a signal delayed by a duration T. As long as the two samples are taken in the same transmission interval of duration T + AT their difference is almost constant. This is true, for each transmission interval for a duration, T minus the time of arrival of the most distant multiple echo.

On the other hand, the rapid fluctuation of this difference indicates that the two samples no longer belong to the same emission interval. Thus from the difference of two samples, the instant of the transmission interval change is determined and consequently a synchronization of the transmission intervals (also called packet synchronization). The signal from multiple echoes that could cause crosstalk is either eliminated in the case illustrated in FIG. 6a, or added coherently to the previous pattern in the case illustrated in FIG. 6c. In the first case, the duration T of the transition interval is advantageously greater than the duration of propagation of the multiple echoes which it is desired to be able to eliminate. The elimination of multiple echoes is carried out for example by ignoring the signals received during the transition intervals 81 of duration AT.

 In the second case, the signals arrived during the transition interval are taken and added to the start of the corresponding useful transmission interval. This latter embodiment requires delay means making it possible to store previous patterns before their processing by the demodulation device 48.

 The demodulation device 48 performs the separation of the signals belonging to the various channels. In the example illustrated in the figure, the processing is digital.

For example, we use a device to calculate the
Discrete Fourier transform. Advantageously, a device for calculating the fast Fourier transform (FFT in English terminology) is used. However, analog separation, for example by using frequency mixer banks, separated by 1 / T, does not depart from the scope of the present invention.

 The demodulated signals are supplied on the one hand, to an analysis circuit 601, and on the other hand, to a servo device 49.

 The analysis circuit 601 performs the analysis of the received signals, the equalization and the calibration of the signals, on the basis of calibration or test signals received from the transmitter.

 The servo device 49 performs synchronization between the various stages of the receiver and between the receiver and the transmitter. In particular, it provides the local oscillators 250 and 491 with synchronization signals allowing their stable operation over time. It also supplies a sampling frequency to the analog-digital converter 4819, to the reorthogonalisation device 482, to the demodulation device 48, to the analysis circuit 601 and to the decision device 602.

 The signals normalized by the analysis circuits 601 are supplied to the decision device 602.

 The decision circuit 602 determines in the presence of which point 14 of FIG. 7 we are in and therefore in the presence of what torque (amplitude, phase) or (real part of the signal, imaginary part of the signal) we are in. The decision device 602 advantageously associates a digital word with each pair.

 The receiver according to the present invention includes other devices such as for example a processing device 45. The processing device 45 performs the desired processing on the I signed. For example, in a television receiver, the processing circuit 45 reconstructs the image and the sound from digital signals.

Advantageously, the processing device 45 uses image decompression algorithms insofar as image compression algorithms have been used on transmission.

 The information processing device 45 is connected to the operating device 46. The operating device 46 makes it possible to process the signals received. The type of operating device mainly depends on the type of receiver used. For example, for the television signal transmission, a cathode ray tube or a flat screen and a loudspeaker will be used in particular. For the transmission of telephone data, the operating device is for example a telephone exchange or a telephone. For the transmission of data, the operating device 46 may for example be a computer receiving the data to be processed or stored.

 In FIG. 9, an exemplary embodiment of the receiver according to the present invention can be seen comprising for the low carriers a processing chain for the real part and a processing chain for the imaginary part of the signal in phase quadrature.

 The device in FIG. 9 comprises an antenna 40, an amplifier 603, a mixer 41, a bandpass filter 42, an amplifier 604 with variable gain, an automatic gain control circuit 605, a mixer 4817, a mixer 4814, a low-pass filter 4818, a low-pass filter 4815, an analog-to-digital converter 4819, an analog-to-digital converter 4816, a reorthogonalisation device 4821, a reorthogonalisation device 4822, a demodulation device 48, an analysis circuit 601 , a decision circuit 602, an information processing device 45, n display device 462, a sound device 461, a servo device 49 and a n / 2 phase shifter 4813.

 The antenna 40 is connected to the input of the amplifier 603. The output of the amplifier 603 is connected to a first input of the mixer 41. The output of the mixer 41 is connected to a bandpass filter 42. The output of the bandpass filter 42 is connected to the input of an amplifier 604. The output of the amplifier 604 is connected to the input of an automatic gain control device 605, to a first input of a mixer 4817 and to a first input of the mixer 4814. The output of the automatic gain control device 605 is connected to a gain control input of an amplifier 604. The output of the mixer 4817 is connected to the input of the filter 4818. The output of mixer 4814 is connected to the input of filter 4815. The output of filter 4818 is connected to the input of analog digital converter 4819. The output of b-pass filter 4815 is connected to the input of analog converter- digital 4816. Ls out e of the analog-digital converter 4819 is connected to the input of the reorthogonization device 4821. The output of the analog-digital converter 4816 is connected to the input of the reorthogonalisation device 4822. The outputs of the reorthogonalisation device 4821, 4822 are connected at the inputs of the demodulation device 48. The output of the demodulation device 48 is connected to the input of the analysis circuit 601 and of the servo device 49. The output of the analysis circuit 601 is connected to the input of the decision device 602. The output of the decision device 602 is connected to the input of the information processing device 45. The output of the information processing device 45 is connected to the operating device such as for example the display 462 and sound device 461. A first output of the servo device 49 is connected to the analog-digital converters 48 19 and 4816, to the reorthogonalisatlon devices 4821 and 4822, to the demodulation device 48, to the analysis circuit 601 and to the decision device 602. This output provides the sampling frequency. In the exemplary embodiment illustrated in FIG. 9, the beat frequency is directly supplied by outputs of the servo device 49. A high frequency output is connected to the second input of the mixer 41. A medium frequency output is connected to the input of the phase shifter of n / 2 4813 and to the second input of the mixer 4817. The output of the phase shifter 4813 is connected to the second input of the mixer 4814. In the device illustrated in FIG. 9 we work on the real and imaginary part of the signal, in phase quadrature. Thus it is possible to lower the frequency, without losing information.

The demodulation device 48 advantageously comprises a device for calculating the transform of
Fourier.

Advantageously, the device for calculating the
Fourier transform is a device for calculating the
Discrete Fourler transform.

Advantageously, the device for calculating the
Fourier transform is a device for calculating the
Fast Fourier transform (FFT in English terminology). The use of Transformation algorithm
Fourler rapid requires performing calculations on a number of samples equal to a power of two. During television transmissions, for example 256,512,1024 or 2,048 channels are used. However, it is not necessary for each channel to transmit information. The use of a fast Fourier Transform calculation device to demodulate the received signal makes it possible to use standard circuits or a combination of standard circuits. of fast Fourier Transforms. It is understood that other alternative embodiments, such as for example homodyne demodulation do not depart from the scope of the present invention.

 In FIG. 10, one can see an example of re-assembly of the analysis device 601. The device in FIG. 10 comprises a switching device 586, equalization device 587, a device for analyzing the test signals 588 and a sequencer 585.

 The routing device 586 receives the signals to be processed. The outputs of the switch device 586 are connected on the one hand to the equalization device 587 and on the other hand to the analysis and test device 588. The output of the analysis and test device 588 is connected to on the one hand to the equalization device 587 and on the other hand to the synchronization device 490.

 The routing device 586 separates the test signals that it sends to the analysis and testing device 588 from the information signals that it sends to the equalization device 587. The detection of the testing signals can be made, for example, according to a predetermined transmission standard. For example, the routing device 586 "makes it so that one channel out of 8 is reserved for the test signals on each transmission interval. In another transmission standard, the test signals can correspond to all the channels of a transmission interval on for example 100.

These two types of test signals used for the calibration of the amplitude and / or of the received phase can be mixed to give for example a test channel on 16 every 64 intervals.

 In a first embodiment, the receiver according to the present invention is designed to be able to follow a single standard. In such a case it is necessary to perform a first synchronization or else receive a synchronization from another device of the receiver.

 In a second variant embodiment of the receiver according to the present invention, the receiver can receive several transmission standards. In this case it is necessary to detect to which transmission standard the received signals belong. Insofar as transmissions on separate channels allow transmission by multiplexing several channels and / or by time multiplexing to send information of different natures, it is possible to reserve, for example, part of the information rate to service information. The service information can, for example, periodically contain information concerning the type of transmission carried out.

 The transmission standard can also be chosen by switching by the user choosing the desired program.

 This one chooses for example to pass from a television program to a radio program. The information on the transmission standards is for example stored in a permanent memory (not shown).

 The switching device 586 comprises for example multiplexers and a wired logic carrying out the orders supplied by the sequencer 585.

 The values of the test signals must be known to the receiver. For example, the test signals are pseudo-random signals. They are generated in the transmitter and the receiver according to the same algorithm which makes it possible to compare the received signal with a signal identical to the signal which was emitted.

 The analysis and test device 588 detects the level received in each of the test channels. It determines the phase shift and the attenuation received in the test channels. From the attenuation and the phase shift the analysis and test device 588 determines the attenuations and the phase shifts in the intermediate channels between the test channels, using for example the interpolation method. The interpolation can for example be a linear interpolation.

 To obtain an important data rate, it is necessary to be able to distinguish close amplitudes and phases and therefore to have an amplitude and phase reference for each channel. This amplitude and phase reference is advantageously given by reference signals transmitted periodically by the transmitter to the receiver. The repetition frequency of reference signals depends on the stability of the propagation conditions and the local oscillators.

 In a first variant embodiment of the device according to the present invention, amplitude and phase reference signals are periodically transmitted on all frequencies 31 to 3N of a time interval of duration T or T + T. However, it it is necessary to note that the frequent emission of calibration signals reduces the rate of useful information transmitted.

 In a significant variant of the device according to the present invention, only a few calibration signals are emitted, advantageously distributed regularly among the frequencies 31 to 3 N, the coefficients of the other frequencies being determined by calculation, for example by interpolation.

 More generally, it is possible to distribute the calibration signals over time and / or on different channels.

 It is for example possible to periodically transmit test signals, each transmission being carried out on different channels. For example, a circular permutation of the channels assigned to the tests is carried out. The impulse response of the transmission medium is deduced for all the channels, for example by interpolation over time and / or over the frequencies. This deduces the matrix of amplitude and phase corrections to be applied to each channel.

It is essential to compensate by calibrations for variations in the impulse response of the transmission medium, for example as a result of a variation (even
local) atmospheric conditions.

We determine the impulse response of the medium, by
example by calculating the Fourier Transform of the corrections to
to bring.

In an example of realization one channel on 8 was used
to the calibration of the amplitude A and the phase of all
channels 31 to 3N. In such a device it is possible,
either to carry out the calibration at each useful interval
of duration T, either as in the case of a realization
previously described from. unlawfully devote some
emission intervals at calibration. Synchronization is
maintained by the use of a stable time base.

The analysis and test device includes
example the memory device and microprocessors of
fast signal processing. The phase shift values and
attenuations for each channel are transmitted to the device
equalization 587.

The equalization device 587 applies to each
channel an amplification and a phase shift opposite to those induced
by transmission. So the amplitudes of all the channels at
reception after equalization by circuit 587 are
proportional to smplitudes during transmission. Likewise, the
relative phase difference between channels on reception, after
treatment by the equalizing device 587 is equal to
relative phase shift between the channels on transmission.

In an analog embodiment variant the
equalization device 587 has variable phase shifters
and variable amplifiers. Analog phase shifters can have digital control, for example,
use charge transfer devices (CCD in
Anglo-Saxon terminology) with a single entry and a
plurality of outputs. Each output corresponds to a phase shift
different.

In a digital variant of the
equalization device 587 multiplications and additions are used to make amplitude and phase corrections. Cable logic and / or microprogrammed or programmed logic are used.

 In the figure there can be seen an architecture of known type capable of being implemented in the device according to the present invention. The architecture of Figure II is likely to be used in the reorthogonalisation device. The storage device 4841 is for example a dual port device. The data to be stored arrives through the input port. This rearranged data leaves from the output port. The 4842 sequencer provides the addresses for recording and re-reading the data. Depending on the type of data reorganization desired it is possible to read whole words or only parts of words or individual bits. The storage device 4841 comprises for example integrated circuits of random access memory (RAM in English terminology).

 The sequencer 4842 includes for example a wired logic and counters. In an alternative embodiment, to use standard circuits it is possible to replace the sequencer 4842, for example with a microprocessor.

Advantageously, the microprocessor is of the signal processing type.

 In Figure 12, we can see a second embodiment of a reorthogonalisatlon device. In an exemplary embodiment illustrated in FIG. 12, the device 482 comprises a storage device 4825, an arithmetic and logic unit 4826, a multiplexer 4823 as well as a sequencer 4824. The output of the storage device 4825 is connected to the input of the arithmetic and logic unit 4826 and to a first input of the multiplexer 4823. The output of the arithmetic and logic unit 4825 is connected to a second input of the multiplexer 4823. The sequencer 4824 receives the general synchronization signals from the receiver for example from of a servo device 49. The sequencer 4824 sends control and synchronization signals to the multiplexer 4823. The sequencer 4824 sends address and synchronization signals to the storage device 4825. In the device illustrated in FIG. 12 Addressing the memory 4825 by the sequencer 4824 makes it possible to rearrange the digital words. The arithmetic and logic unit is responsible for carrying out the desired summations of the signals. The switching of the multiplexer 4823 makes it possible to choose between the two rearrangement modes according to the desired transmission standard and the reception phase in progress.

On the other hand it is possible to use to exploit the signal to use an amplitude / phase demodulator such as that illustrated in FIG. 9 of Patent FR 86 13937 filed on 7
October 1986 by the Applicant.

 The transmitter according to the present invention sends coding signals allowing upon reception the precise synchronization of a time base of the receiver with a time base of the transmitter. Thus it is possible to have a good temporal and / or phase resolution.

 In a first embodiment of the device according to the present invention, digital synchronization is used.

 In an exemplary embodiment illustrated in FIG. 14, an analog synchronization is used.

 In the example illustrated in FIG. 13, a set 3000 of signals modulated over N channels is transmitted.

 The spectrum is substantially rectangular having a frequency width f 1 equal to B, the pass band and a height Am corresponding to the average amplitude A 2 of the signal.

Within the band B, two frequencies A are emitted and with an amplitude AM significantly greater than Am. For example
AM is 12 dB higher than Am. So at reception, knowing the frequencies A and fB f we will be able to separate A and fB. By knowing frequency A and fB on the one hand and their difference at reception on the other hand, we obtain a reference of the frequency from which we can deduce a time reference. On reception, the difference A - fB is obtained, for example, by beating the frequencies A and fB in a mixer.

 In an exemplary embodiment of the device - according to the present invention, B is equal to 8 MHz and fA is distant from 5 MHz.

 In FIG. 14 is shown an example of an analog embodiment of the control device 49 of FIGS. 8 and 9.

The device of FIG. 14 is intended to operate with a signal emitted by the transmitter as illustrated in FIG. 13. The control device 49 comprises a bandpass filter 701 and a bandpass filter 702, a mixer 703, a phase locked loop 704 (PLL in English terminology), a frequency division phase locked loop 709 (PLL), a frequency division phase locked loop 710 (PLL) and a 711 frequency division phase locked loop (PLL). Phase-locked loops include, for example, a mixer, a low-pass filter, a voltage-controlled oscillator. In FIG. 14, the loop 704 comprises a mixer 705, a low-pass filter 706, an oscillator controlled by a voltage 707 (VCO in English terminology).

 The input of the device 49 is connected to the inputs of the filters 701 and 702. The output of the filter 701 is connected to a first input of the mixer 703. The output of the filter 702 is connected to a second input of the mixer 703. The output of the mixer 703 is connected to a first input of the mixer 705. The output of the mixer 705 is connected to the input of the low-pass filter 706. The output of the oscillator 707 is connected to the input of the phase lock loop 709 , at the input of the phase lock loop 710 at the input of the phase lock loop 711. The output of the low pass filter 706 is connected to the input of the oscillator 707. The output of the The oscillator 707 is connected to the second input of the mixer 705. The outputs of the phase lock loops 709, 710 and 711 constitute the outputs of the servo device 49 providing the desired frequencies.

The filter 701 selects the frequency f A 'the filter 702 selects the frequency fB. The mixer 703 performs the beat between the frequency A and
The phase locked loop 704 provides the value of the differences between frequencies f A and fB. The difference between the frequencies fg, A at the emission determined by the emission standard is known. The reception comparison provides a frequency and phase reference.

 The phase locked loops 709, 710 and 711 make it possible to provide frequency and phase references sufficiently stable for the operation of the device according to the present invention. For example, the loop 709,710,711 makes it possible to supply a frequency reference respectively to the local oscillator 250 and to the local oscillator 491 of FIG. 8 and a sampling clock signal to the digital devices of FIGS. 8 or 9. The frequencies d oscillation depend on the oscillator settings.

 The invention applies to devices for receiving analog and / or digital information, communications between computers, telephone communications between exchanges, telephone communications between radiotelephones and communication stations, radio communications between earth and satellite stations, communications between satellites, acoustic communications in air and / or water, the creation of local computer networks and the reception of radio and television broadcasts.

 The device according to the present invention also applies to sonars and radars.

 The invention is particularly well suited to the production of high fidelity radio receivers as well as to the production of high definition television receivers (HDTV in English terminology) and / or digital television.

Claims (12)

 1. Receiver comprising sampling means synchronous with the signal, characterized in that it comprises means for demodulating a transmission of modulated waves using symbols transmitted during a duration T 4, T on a plurality of frequencies, two emission frequencies being 1 / T apart, T - being the useful emission interval and AT being the transition interval and that it comprises a servo device (49) ensuring synchronization of the receiver with the signal received.  2. Receiver according to claim 1, characterized in that it comprises an automatic gain control device (AGC) controlled by a device for detecting the average power of at least part of the signal.  3. Receiver according to claim 1 or 2, characterized in that it comprises decoding means (45) of torque (amplitude, phase) or (real part, imaginary part) to convert them into digital words.  4. Receiver according to claim 1, 2 or 3, characterized in that it comprises at least one device (483) for calculating the fast Fourier transform (FFT).  5. Receiver according to claim 1, 2, 3 or 4, characterized in that it comprises a test device (601) capable of providing amplitudes and / or phases. reference from calibration signals.  6. Receiver according to claim 5, characterized in that it comprises an equalization device (587) compensating for the disturbances of the signal from the transmission.  7. Receiver according to any one of the preceding claims, characterized in that it comprises reorthogonalization means (482,4821,4822) using the transition interval a T to make a plurality of channels orthogonal.  8. Receiver according to any one of the preceding claims, characterized by. the fact that said receiver is a receiver of radio transmissions.  9. Receiver according to claim 1, 2, 3, 4, 5, 6, or 7, characterized in that said receiver is a receiver of television broadcasts.  10. Method for receiving modulated waves characterized in that the modulated waves are electromagnetic waves.  11. Method according to claim 10, characterized in that the step of separating N orthogonal channels comprises a step of calculating the fast Fourier transform (FFT) of the signal.  12. Method according to claim 11, characterized in that it comprises a step of reconstituting a television signal from the signals received in the N channels.