• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    A method of demodulating a received signal resulting from the modulation of a basic chirp signal comprising a step of estimating (E46) a symbol carried by the received signal, implementing the following substeps: E43) of N decision components, from said received signal and a reference chirp signal obtained by modulating said base chirp signal with a reference symbol corresponding to a symbol of rank r, an index decision component I , denoted component D1, being a function of a term whose phase depends quadratically on I, with I an integer from 0 to N-1; • decision (E44) of the rank k of the symbol carried by said received signal, from the decision component, index k, denoted component Dk, having an extremum of value among said N decision components.
    公开号:FR3052615A1
    申请号:FR1655322
    申请日:2016-06-09
    公开日:2017-12-15
    发明作者:Stephane Paquelet；Patrick Savelli
    申请人:B Com SAS；
    IPC主号:
    专利说明:

    Method for demodulating a received signal, computer program product and corresponding device
    1 TECHNICAL FIELD
    The field of the invention is that of the transmission of data via a radiofrequency link based on the modulation of a so-called "chirp" waveform as used in the LoRa® technology.
    More specifically, the invention relates to a method for demodulating such a waveform, which has improved performance over existing techniques with comparable implementation complexity.
    Since the LoRa® technology is dedicated to low-power transmission by connected objects, the invention has applications in all areas of the personal and professional life in which the connected objects are present, in particular, but not exclusively, in the fields of health, sports, domestic applications (security, household appliances, etc.), followed by objects, etc.
    2 TECHNOLOGICAL BACKGROUND
    Presented as the "third revolution of the Internet", connected objects are gaining ground in all areas of everyday life and business. Most of these objects are intended to produce data through their built-in sensors to provide value-added services to their owner.
    By the intended applications, these connected objects are mostly nomadic. In particular, they must be able to transmit the data produced, regularly or on demand, to a remote user.
    To do this, the long-range radio transmission of cellular mobile radio type (2G / 3G / 4G ...) has been a technology of choice. This technology allowed for good network coverage in most countries.
    However, the nomadic aspect of these objects is often accompanied by a need for energy autonomy. However, even based on one of the most energy-efficient cellular mobile radio technologies, the current connected objects continue to present a prohibitive consumption to allow a large-scale deployment at a reasonable cost.
    Faced with the problem of the consumption of the radio link for such nomadic applications, new low-speed and low-speed radio technologies specifically dedicated to "Internet of Things" networks, ie radio technologies for so-called LPWAN networks. (for "Low-Power Wide-Area Networks" in English), are seeing the day.
    In this context, two kinds of technologies can be distinguished: on the one hand, there are proprietary technologies such as Sigfox® technology, LoRa® technology, or Qowisio® technology. In practice, these non-standardized technologies are all based on the use of the "Industrial, Scientific and Medical" frequency band, called ISM, and the regulations associated with its use. The advantage of these technologies is that they are already available and allow the rapid deployment of networks on the basis of limited investments. In addition, they allow the development of connected objects very energy efficient and low cost; on the other hand, there are several technologies promoted by standards bodies. For example, there are three technologies currently being standardized with 3GPP (for "3rd Generation Partnership Project"): NB-loT (for "Narrow Band - Internet of Things" in English), LTE MTC ( for "Long Term Evolution - Machine Type Communication" in English) and EC-GPRS (for "Extended Coverage - General Racket Radio Service"). However, such solutions are not yet fully specified and will also rely on licensed frequency bands.
    In this context, it appears that the proprietary technologies based on the use of the ISM band are presented as the solutions of choice in the short term and one or some of them can then be imposed de facto as the solution use.
    For example, EP 2 449 690 B1 describes an information transmission technique based on the modulation of a basic chirp signal on which LoRa® technology is based.
    However, some operators, such as Bouygues® or Orange® in France, have already focused on LoRa® technology to deploy their network dedicated to connected objects. However, the first feedback comes from unsatisfactory user experiences related to limited performance of the radio link in real conditions.
    There is therefore a need to improve the performance of a receiver implementing the LoRa® technology in real conditions, and in particular in the face of a mobile radio propagation channel presenting fading.
    There is also a need that such an improvement does not induce an overconsumption in energy of the receiver and therefore does not penalize the autonomy of the connected object carrying such a receiver.
    3 SUMMARY
    In one embodiment of the invention, there is provided a method for demodulating a received signal. This received signal results from the modulation of a basic chirp signal, whose instantaneous frequency varies linearly between a first instantaneous frequency f0 and a second instantaneous frequency f1 during a symbol time Ts, and the transmission of the modulated chirp signal in a channel of transmission. The modulation corresponds, for a symbol of rank s of a constellation of N symbols, s being an integer from 0 to N-1, to a circular permutation of the variation pattern of said instantaneous frequency on the symbol time Ts, obtained by a time shift of s times an elementary time duration Te, such that N * Tc = Ts.
    Such a method comprises a step of estimating a symbol carried by the received signal, implementing the following substeps: determination of N decision components, from the received signal and a reference chirp signal obtained in modulating the basic chirp signal by a reference symbol corresponding to a symbol of rank r in the constellation, a decision component of index I, denoted component Di, being a function of a term whose phase depends quadratically on I, with I an integer of 0 to N-1; decision of the rank k of the symbol carried by the signal received, from the decision component, index k, denoted component D ^, having an extremum value among the N decision components.
    Thus, the invention proposes a new and inventive solution to allow the estimation of a symbol carried by a received signal resulting from the modulation of a basic chirp signal having a linear variation of its instantaneous frequency, or in an equivalent manner, presenting a quadratic variation of its instantaneous phase.
    To do this, the claimed method proposes to take into account this quadratic variation of the instantaneous phase of the received signal in order to implement an optimal receiver to decide the rank of the received symbol.
    The reception performance is thus improved while maintaining a complexity comparable to that of receivers of the prior art.
    According to one embodiment, the step of estimating a symbol further comprises the following steps, for N samples of the received signal and for N samples of the reference chirp signal, taken at the same multiple instants of Te: conjugation of N samples of the reference chirp signal, respectively N samples of the received signal, delivering N samples of a conjugated chirp signal; term-by-term multiplication of the N samples of the conjugated chirp signal by the N samples of the received signal, respectively of the reference chirp signal, delivering N samples of a multiplied signal; direct or inverse Fourier transformation of the multiplied signal, delivering N samples Yi of a transformed signal, with I an integer ranging from 0 to N-1; and the component is furthermore function of a term proportional to an amplitude of the sample of index k, Y ^, among the N samples Yi of the transformed signal, as well as the phase of the sample 7¾.
    Thus, the claimed method proposes to take into account the complete information (ie amplitude and phase) contained in the samples of the signal at the Fourier transform output, direct or inverse, and not only on the basis of the module of these samples as done in the state of the art. The reception performance is thus improved while maintaining a comparable complexity.
    According to one embodiment, the component is furthermore function of a subset of N 'samples Y' among the N samples Yi of the transformed signal, with n different from ok, with N '<N, and with σ a parameter belonging to {-1,1}.
    Thus, the claimed method makes it possible to take into account the dispersion of the channel and the interference between symbols which results in deciding the rank of the received symbol, thereby improving the reception performance in the presence of a transmission channel having muitipies trips.
    In one embodiment, the method comprises a step of obtaining (E45) N channel coefficients and a sample of index n of the subset of samples Y "is weighted by a coupling coefficient proportional to the channel coefficient. / i (jk-n [N] depending on the difference between the indices ok and n, with σ a parameter belonging to {-1,1}, and a term whose argument depends quadratically on the index k. term proportional to an amplitude of the sample is a channel coefficient Hq independent of k.
    Thus, the terms weighting the samples Yn have a component that depends solely on the difference between the indices of these samples considered at the output of the Fourier transform. Indeed, the invariance in time of the impulse response of the channel leads to terms representative of the interference between symbols depending solely on the difference between the indices of the signal samples considered.
    However, the quadratic variation of the phase of the received signal requires that the coupling between samples is not invariant in time for a given difference between sample indices considered.
    Thus, taking these two effects into account in the very structure of a component considered to estimate the received symbol makes it possible to implement an improved performance reception in the presence of a transmission channel having multiple paths, while at the same time making it possible to working in the frequency domain, ie by working on the Fourier transform output samples.
    According to various embodiments, the component is a function of a term proportional to: the real part of the sum Ση = ι® ^ ak-n [N] Sk; ° u of the conjugate complex of the sum, when the Fourier transformation is a direct Fourier transform and when the conjugated chirp signal corresponds to the conjugation of the reference chirp signal; or the real part of the sum
    or the sum conjugate complex, when the Fourier transform is an inverse Fourier transform and the conjugated chirp signal is the conjugation of the reference chirp signal; or the real part of the sum
    or complex
    conjugate of the sum, when the Fourier transform is a direct Fourier transform and when the conjugated chirp signal corresponds to the conjugation of the received signal; or the real part of the sum or conjugate complex of the sum, when the Fourier transform is an inverse Fourier transform and when the conjugated chirp signal corresponds to the conjugation of the received signal; with
    and with σ a parameter belonging to {-1,1}.
    Thus, taking into account in analytic form, ie in the structure of the estimator of the symbols received, the waveform of the signal considered, for example the quadratic variation of its instantaneous phase, allows a simple implementation. and efficient optimal receiver in the sense of the maximum likelihood in multipath transmission channel in the frequency domain, ie by working on the samples of the signal at the output of the Fourier transform, direct or inverse.
    Moreover, in a variant, only N 'channel coefficients are taken into account among the possible N, thereby simplifying the on-board processing in the receiver.
    According to one embodiment, the channel coefficients // σΚ-η [Ν] are harmful for n different from ok.
    Thus, the claimed method makes it possible to implement the optimal receiver in the sense of the maximum likelihood in the frequency domain, ie by working on the Fourier transform output samples in the presence of a channel which is reduced to an additive Gaussian white noise. , or channel AWGN (for "Additive white Gaussian noise" in English), which thus introduces no interference between symbols. The performances of the receiver are thus improved and present an optimality criterion in AWGN channel for a minimal additional computing cost.
    According to one embodiment, the obtaining step further comprises an estimation of the channel coefficients from the N samples Υ "of the transformed signal and at least one predetermined symbol k.
    Thus, the claimed method makes it possible to estimate the parameters necessary to take into account the transmission channel with a view to implementing an optimal receiver for estimating the symbols received while working in the frequency domain, ie by working on the samples at the Fourier transform output. Moreover, taking into account in analytic form the waveform of the signal considered, for example the quadratic variation of its instantaneous phase, makes it possible to have to estimate only the invariant part in time of the terms generating the interference between symbols, ie of the part depending only on the difference between the indices of the samples considered, thereby leading to an efficient implementation of the estimation step of the parameters representative of the impact of the transmission channel, and therefore the receiver.
    According to one embodiment, the estimated channel coefficients forming a vector
    the estimation of the coefficients being carried out on the basis of Ns received symbols, kj designating the rank of the i-th of said Ns symbols in the constellation of N symbols, η denoting the rank of a reference symbol used when receiving said i -th symbol, y ['' ^ denoting N samples of said transformed signal obtained during said reception of said i-th symbol, the estimated vector H_ of H_ =
    expresses itself as
    with
    when the Fourier transform corresponds to a direct Fourier transform and when the conjugated chirp signal corresponds to the conjugation of the reference chirp signal; or
    when the Fourier transform corresponds to an inverse Fourier transform and when the conjugated chirp signal corresponds to the conjugation of the reference chirp signal; or
    when the Fourier transform corresponds to a direct Fourier transform and when the conjugated chirp signal corresponds to the conjugation of said received signal; or
    when the Fourier transform corresponds to an inverse Fourier transform and when the conjugated chirp signal corresponds to the conjugation of said received signal; with
    and with σ a parameter belonging to {-1,1}.
    Thus, estimation of the parameters necessary to take into account the transmission channel corresponds to the minimum quadratic error between the emitted symbol and the received symbol, thereby reducing the estimation errors on the received symbol.
    Moreover, in a variant, only N 'channel coefficients are taken into account among the possible N, thereby simplifying the on-board processing in the receiver.
    According to one embodiment, the step of estimating the channel coefficients comprises the following sub-steps: calculation of parameters representative of the channel coefficient Hq and another of the channel coefficients; obtaining parameters representative of the remaining channel coefficients from the calculated parameters.
    Thus, the chirp waveform as well as the choice of the value of Te in effective systems such as the LoRa® (Sus) system, which remains large compared with the maximum temporal dispersion of the channel, lead to the fact that only two parameters remain to be estimated. (eg Wq and another term Hi with I not zero) for the determination of the set of terms H , thus leading to a great simplicity of implementation of the step of estimating the parameters representative of the impact of the channel transmission, and therefore the receiver in the end.
    According to one embodiment, the channel coefficient of non-zero index I is inversely proportional to
    Thus, the chirp waveform and the choice of the Te value in effective systems like the LoRa® system, which remains large compared to the maximum time dispersion of the channel, also lead to an exponential decay of the amplitude of the Hi terms. according to I. This shows that it can be envisaged to use only a restricted amount of the terms Hi to model the effect of the channel, eg the terms corresponding to an index I less than or equal to 10, thus reducing the computational complexity of the optimal receiver in the sense of maximum likelihood.
    According to one embodiment, the predetermined symbol is a symbol of a learning sequence or a received symbol whose rank k has been decided during a previous execution of said symbol estimation step.
    Thus, the estimation of the parameters necessary to take into account the transmission channel can be done on the basis of known symbols, eg learning or synchronization sequences, thereby allowing a robust estimation of these parameters, or the database of previously received data symbols, thereby allowing this estimate to be refined during reception. The invention also relates to a computer program, comprising program code instructions for implementing a method for demodulating a received signal resulting from the modulation of a basic chirp signal as described above, according to any one of its various embodiments, when said program is executed by a processor.
    In another embodiment of the invention, there is provided a device for demodulating a received signal, the received signal resulting from the modulation of a basic chirp signal as described above.
    Such a demodulation device comprises a reprogrammable calculation machine or a dedicated calculation machine, adapted to and configured to: determine N decision components, from the received signal and a reference chirp signal obtained by modulating the chirp signal by a reference symbol corresponding to a symbol of rank r in the constellation, a decision component of index I, denoted component D], being a function of a term whose phase depends quadratically on I, with I an integer from 0 to N-1; deciding on the rank k of the symbol carried by the signal received, from the decision component, index k, denoted component D ^, having an extremum of value among the N decision components.
    Such a demodulation device is particularly suitable for implementing the method of demodulating a received signal resulting from the modulation of a basic chirp signal according to the invention (according to any one of the various embodiments mentioned above).
    Thus, the characteristics and advantages of this device are the same as those of the demodulation method described above. Therefore, they are not detailed further.
    4 LIST OF FIGURES Other features and advantages of the invention will become apparent on reading the following description, given by way of indicative and nonlimiting example, and the appended drawings, in which: FIG. 1 illustrates the characteristics of an unmodulated chirp used in LoRa® technology; FIG. 2 illustrates the instantaneous frequencies and the instantaneous phases of different chirps modulated according to the LoRa® technology; Figures 3a and 3b illustrate receiving structures according to different embodiments of the invention; FIG. 4 illustrates the steps of a demodulation method according to various embodiments of the invention; FIG. 5 illustrates the decrease of the coupling terms between samples according to different embodiments of the invention; FIG. 6 illustrates the performances obtained compared with those obtained by the technique of the prior art in a particular embodiment of the invention; FIGS. 7a and 7b show examples of demodulation device structures according to different embodiments of the invention.
    DETAILED DESCRIPTION OF THE INVENTION
    In all the figures of this document, the elements and identical steps are designated by the same reference.
    The general principle of the invention is based on the estimation of a symbol of a received signal, corresponding to a modulated chirp signal transmitted in a transmission channel, from N decision components representative of the symbol, in a constellation of N symbols.
    To do this, the 1-th component among the N decision components is a function of I via a complex term whose argument varies quadratically as a function of I. The index k representative of the symbol received in the constellation of N symbols is then determined as a function of the index k of the decision component that has an extremum of value among the N decision components.
    The proposed solution makes it possible in particular to demodulate a signal generated using the technique described in the aforementioned patent EP 2,449,690 B1.
    As already indicated, this patent EP 2 449 690 B1 describes an information transmission technique based on the modulation of a basic chirp signal. As shown in FIG. 1, the instantaneous frequency 102 of the basic chirp signal varies linearly between a first instantaneous frequency f0 and a second instantaneous frequency f1 during the duration Ts of a symbol. Such an instantaneous frequency is here representative of the speed of rotation in the complex plane of the vector whose coordinates are given by the signals in phase 100 and in quadrature 101 representing the modulating signal intended to modulate the radio carrier so as to transpose the chirp signal. base on carrier frequencies and thus generate a radiofrequency signal.
    Since the chirp signal has a constant envelope, the signal in phase 100, respectively in quadrature 101, oscillates between two extreme values, 10 and 11, respectively Q 0 and Q 1, its frequency varying linearly over time just like the instantaneous frequency 102 of the chirp signal. resulting basic By the linear variation of the instantaneous frequency 102, the basic chirp signal thus defined has an instantaneous phase 103 that varies quadratically between two values φο and Φι during the duration Ts, the instantaneous frequency being the derivative of the instantaneous phase.
    The modulated chirp signals are then obtained by circular permutation of the variation pattern of the instantaneous frequency of the base chirp, over a duration Ts, obtained following a time shift of k times an elementary time duration, called "chip" duration, Te. The index k is then representative of the rank of a symbol in a constellation of N symbols and we then have N * Tc = Ts. By way of illustration, FIG. 2 represents the instantaneous frequency 102, 102 ', 102 ", 102'" and the instantaneous phase 103, 103 ', 103 ", 103"' of different modulated chirps respectively corresponding to k = 0, k = l, k = 2 and k = 3, ie allowing the transmission of information on the basis of a constellation of four symbols. The base chirp, corresponding to k = 0, is then interpreted in this case as bearing the rank zero symbol in the constellation.
    The inventors have found that, according to this technology, the determination of the value of a symbol received via such a signal, ie the determination of its rank k in the constellation of N symbols, is equivalent to the determination of the index k having used as a basis for calculating the time offset used to generate the instantaneous phase and frequency pattern of the modulated chirp signal in question.
    It appears that the basic chirp signal can be expressed in the time domain, and over the duration of a symbol period, i.e. for t ranging from 0 to Ts as
    with 00 the initial value of the phase.
    In practice, the signal LoRa® is such that the bandwidth of the signal chirp, i.e. | / i - fo , is adjusted in inverse of the chip duration Te, and fl is chosen such that A = - / o. Knowing that Ts = N * Tc, the expression of the instantaneous phase of the chirp signal can then be rewritten as
    with σ a parameter belonging to {-1,1} making it possible to model both rising chirps (i.e. with an increasing instantaneous frequency), and descendants (i.e. with decreasing instantaneous frequency). The analytic expression, Sfe (t), of a chirp modulated by a symbol of rank k in the constellation of N symbols (k being thus between 0 and N-1), and thus corresponding to a circular permutation of the pattern of basic chirp as described above, then expresses itself as
    (Eq-1) where [. ] designates the modulo function.
    This equation can then be reformulated as follows, for t between 0 and
    Ts = N * Tc: (Eq-2a) with: (Eq-2b) (Eq-2c)
    FIGS. 3a and 3b will now describe two reception structures for estimating a symbol carried by a received signal, corresponding to a basic chirp signal modulated according to the technique described above, ie making it possible to decide on the index k used to generate the pattern of variation of the instantaneous frequency and phase of this signal, according to various embodiments of the invention.
    More particularly, these figures illustrate the structures used to perform the processing on the phase, I, and quadrature signals, Q, representing the modulating signal obtained after radio frequency demodulation, or RF demodulation, of the received radio frequency signal (in the following patent term, the term RF demodulation means the baseband transposition of the received signal, this transposition delivering analog I and Q signals representing the signal modulating the received RF carrier, and the term demodulation denotes the processes carried out on the I signals and Q, often after sampling and quantization, leading to the determination of the information contained in the modulating signal). During this RF demodulation, it is always possible to choose a carrier frequency so that fi = -fo.
    In practice, such I and Q signals are obtained via the use of an RF receiver known to those skilled in the art (for example a direct conversion receiver, superheterodyne or any equivalent architecture), implementing a quadrature RF demodulator and delivering two analog I and Q channels
    The I and Q signals are then sampled by an ADC digital analog converter 301 (for example a flash converter, or based on sigma-delta modulator, or the SAR type for "Successive Approximation Register" in English, or any other equivalent) present. on the corresponding reception channel. In a conventional reception chain, such a converter operating at a sampling frequency often high compared to the bandwidth of the wanted signal, the signal delivered by the ADC is decimated by a decimation stage 302 (for example a filter with a CIC type linear phase for "Cascaded Integrator-Comb in English" or any other equivalent) present on each of the I and Q channels so as to deliver each N samples, which can be interpreted as the real and imaginary parts of N complex samples.
    The N complex samples are then delivered to a demodulation device 300, 300 'comprising different modules.
    According to the embodiment illustrated in FIG. 3a, the N complex samples are directly delivered to a complex multiplier 303. The complex multiplier 303 then multiplies term by term the N complex samples with N complex samples representative of a conjugated reference chirp signal. delivered by a generation module 307, here a correspondence table, or LUT (for "LookUp Table" in English), storing the corresponding precomputed samples.
    Such a conjugated chirp signal is here defined as a chirp signal whose instantaneous frequency varies inversely from that of the chirp signal in question. For example, reconsidering the case of a basic chirp signal as discussed above in relation to FIG. 1, ie whose instantaneous frequency varies linearly from f0 to fl over a duration Ts, the base conjugated chirp signal will then have a instantaneous frequency that varies linearly from fl to fO over the same duration Ts. In this way, the multiplication of a chirp signal by its conjugate cancels the linear variation of its instantaneous frequency. The result then has a constant instantaneous frequency.
    In another embodiment illustrated in FIG. 3b, the sign of the imaginary part of the N complex samples corresponding to the received signal is inverted by an inversion module 310. Thus, the inversion module 310 delivers signals corresponding to the signals base band I and Q representative of the conjugated chirp signal of the chirp signal actually received.
    The N complex samples thus obtained are then delivered to the complex multiplier 303 which multiplies them term by term with N complex samples representative of the reference chirp signal delivered by the generation module 307.
    The N complex samples delivered by the complex multiplier 303 are therefore, in this second embodiment, the conjugate complexes of those obtained in the embodiment described above in relation to FIG. 3a.
    The N complex samples delivered by the complex multiplier 303 are then delivered to a discrete Fourier transform module 304.
    In one embodiment, the implemented discrete Fourier transform is a direct discrete Fourier transform. In another embodiment, the implemented discrete Fourier transform is an inverse discrete Fourier transform.
    Thus, four embodiments appear here: in a first embodiment, the conjugation is applied to the reference chirp signal (case of FIG. 3a), and the implemented discrete Fourier transform is a direct discrete Fourier transform; in a second embodiment, the conjugation is applied to the reference chirp signal (case of FIG. 3a), and the implemented discrete Fourier transform is an inverse discrete Fourier transform; in a third embodiment, the conjugation is applied to the received chirp signal (the case of FIG. 3b), and the implemented discrete Fourier transform is a direct discrete Fourier transform; in a fourth embodiment, the conjugation is applied to the received chirp signal (the case of FIG. 3b), and the implemented discrete Fourier transform is an inverse discrete Fourier transform.
    In variants, N expresses itself as a power of two and the discrete Fourier transform in question is implemented as a fast Fourier transform.
    The N transformed complex samples delivered by the discrete Fourier transform module 304 are then supplied to a generation module 305 of N decision components representative of the rank k, in the constellation of N symbols, of the symbol carried by the received signal.
    The N components are then delivered to a decision module 306 which decides the rank k of the received symbol as a function of the index of the component which has a value extremum among the N components.
    In a variant, the N components representative of the rank k of the symbol modulating the basic chirp signal take into account the effect of the propagation channel. A channel estimator 308 then estimates channel coefficients based on samples provided by the discrete Fourier transform module 304 and the rank of the corresponding received symbol decided by the decision module 306.
    A method of demodulating a received signal, in particular for estimating a symbol carried by the received signal according to various embodiments of the invention, will now be described with reference to FIG.
    During a step E40, a conjugated chirp signal is obtained. As described above in relation to FIGS. 3a and 3b, this conjugated chirp signal may correspond to either the signal resulting from the conjugation of the baseband signal Sj- (t) representing the reference chirp of a duration Ts delivered by the generation module 307 (first and second embodiments mentioned above), or to the signal resulting from the conjugation of the baseband signal y (t) representing the received chirp signal (third and fourth embodiments mentioned above), also of a duration ts.
    In general, the reference chirp signal corresponds to a basic chirp signal modulated by a reference symbol of rank r in the symbol constellation. In a variant, r is equal to 0 and the reference chirp signal is the basic chirp signal.
    During a step E41, the complex multiplier 303 delivers the multiplied signal to the discrete Fourier transform module 304.
    In the aforementioned first and second embodiments, this multiplied signal is thus expressed as yCOs ^ CO, and in the third and fourth embodiments mentioned above, this multiplied signal is thus expressed as y * (t) Sj. ), ie as the conjugate complex of the signal delivered by the complex multiplier 303 in the first and second embodiments.
    An analytical expression of the product y (t) Sr (0 is first derived hereinafter.
    In general, the received chirp signal has been propagated via a radio propagation channel whose impulse response h (t) can be expressed conventionally as a sum of P time-shifted paths, each path then being able to be modeled by an amplitude. Ap complex and a real delay Τρ so that (Eq-3) with 5 (t) the distribution of Dirac.
    Moreover, the received signal is also tainted with an additive noise w (t) assumed Gaussian and centered so that one can write in a general way that: y (t) = (h * Sfc) (t) + w (t) with t E [0, Ts -I- Τγηαχ] 6t Tyfidx = the support of the impulse response / i (t) being [Ο, τ ^ αχ] ·
    Once the receiver is synchronized in time, it can then be written, assuming that the received signal corresponds to a basic chirp signal modulated by a symbol of rank k in the constellation of symbols, that
    Thus, at the output of the complex multiplier 303 and in the first and second embodiments mentioned above, it appears that
    In a step E42, a Fourier transform is applied by the discrete Fourier transform module 304 to output a transformed signal.
    In order to simplify the writing, the continuation of the computation is presented for the particular case where the reference symbol corresponds to the basic chirp, i.e. for r = 0, even if the results will be given for the general case.
    Noting
    and
    and defining e as
    we can then use the expression of Sfc (t) given by (Eq-2a) to express Sfc (t - r) s * (t) as
    By applying a direct discrete Fourier transform (DFT) on the sampled signal ΐί ^ "(τ) = 5] ^ (ηΤ ^ - τ) 3 * (ηΤ ^), it appears that
    Noting q the term
    it seems that
    So
    This equation can be reformulated so as to reveal the terms that depend on the propagation channel and those related to the waveform used. So
    We can then finally express the samples of the transformed signal as
    or in another form
    with I and k from 0 to N-1 and
    (Eq-4a) (Eq-4b) (Eq-4c) (Eq-4d)
    In the general case where the reference chirp signal corresponds to a basic chirp signal modulated by a reference symbol of rank r in the symbol constellation, the calculation gives for the N samples of the transformed signal Yi obtained at the output of the transform module Fourier 304: in the first aforementioned embodiment (corresponding to the application of a direct Fourier transform on y (nTc) Sr (nTi.) and on winTc) Sy {nTc)):
    in the aforementioned second embodiment (corresponding to the application of an inverse Fourier transform on y (nTc) s ^ (nTi.) and on w (nTc) Sr (nTc)):
    (Eq-5b) in the third aforementioned embodiment (corresponding to the application of a direct Fourier transform on y * (n7 £) s ^ (n7'c) and on w * (nTc) Sr (nTc )):
    (Eq-5c) in the fourth embodiment mentioned above (corresponding to the application of an inverse Fourier transform on y * (nTc) Sy (nTi.) And on w * (nTc) Sr (nTc)):
    (Eq-5d)
    Moreover, in order to simplify the reading, the same notations Yi, Hi and Wi are used to designate the corresponding samples obtained at the output of the Fourier transform module 304 regardless of the aforementioned embodiment.
    During a step E43, N decision components Dj, I integer from 0 to N-1, which can be interpreted as representing the N components of a decision vector {Dq, D ^, ... and representative of the rank of the symbol carried by the received signal are determined by a generation module 305.
    To do this, it is proposed in one embodiment to apply a maximum likelihood criterion on the N samples Yi delivered by the discrete Fourier transform module 304. Indeed, the Gaussian hypothesis for the additive noise νν ( ηΤ ^) remains true for the Wi samples obtained at the output of the discrete Fourier transform module 304, the Fourier transform of a Gaussian distribution giving another Gaussian distribution.
    By reconsidering, for example, the aforementioned first embodiment (corresponding to the application of a direct Fourier transform on y (nTc) and reconsidering the particular case where the reference symbol corresponds to the basic chirp signal, ie for r = 0, for clarity in the writes, Wi samples can be expressed as follows on the basis of Equation (Eq-5a):
    Thus, applying a maximum likelihood criterion, the rank of the symbol modulating the basic chirp signal and corresponding to the received signal, corresponds to the index k maximizing the probability density of the symbol observed in reception, or, in the case of a Gaussian density, with the index k minimizing the argument of the Gaussian function, ie the quantity
    Equivalently, after development of the squared module and change of variable from n to N-n, it appears that the rank of the symbol corresponding to the received signal is expressed as a function of the index k maximizing the quantity
    where 3i (.) denotes the real part. Equivalently, the conjugate complex of the argument of the real part above could be taken.
    In other words, N decision components D ;, I from 0 to N-1, allowing the estimation of the rank of the symbol carried by the received signal can be determined on the basis of this expression taken for the various rank hypotheses. possible symbols (ie the N hypotheses corresponding to k ranging from 0 to N-1 in the expression above). Each of the N decision components Di then corresponds to the quantity above taken for the corresponding symbol rank assumption, and the estimate k of the rank of the symbol carried by the received signal is then expressed as a function of the component of decision, of index k, denoted component D ^, having an extremum of value among the N decision components Di thus determined.
    In the general case where the reference chirp signal corresponds to a basic chirp signal modulated by a reference symbol of rank r in the symbol constellation, an equivalent calculation makes it possible to define the N decision components D; obtained at the output of the generation module 305, the decision component of index k, Dfc, expressing itself as: in the first embodiment mentioned above (corresponding to the application of a direct Fourier transform sury (nTi.) Sr (nTi.) And on w (nTc) Sr (nTc)):
    (Eq-6a) in the aforementioned second embodiment (corresponding to the application of an inverse Fourier transform on y (nT ^) Sr (nTr) and on w (nTr) Sr (nTc)):
    (Eq-6b) in the third aforementioned embodiment (corresponding to the application of a direct Fourier transform on y * (nTr) Sr (nTr) and on w * (nTc) Sy (nTc ')):
    (Eq-6c) in the fourth aforementioned embodiment (corresponding to the application of an inverse Fourier transform on y * (nTr) s ^ nTr) and on w * (nTc) Sy (nTc)):
    (Eq-6d)
    As discussed above, in variants, it is the conjugate complex of the real-part argument defining 0¾. which is taken from the equations (Eq-6a) to (Eq-6d).
    Alternatively, the radio propagation channel is reduced to a single path (e.g., in the case of a point-to-point link in direct view). In this case, the impulse response given by equation (Eq-3) is reduced to a single term of amplitude Aq. Similarly, assuming a perfect synchronization of the receiver, we have Tq = 0. It then appears from the equations (Eq-4a) and (Eq-4b) that all the terms H (are harm for I going from 1 to N-1, and that only Hg is non-zero.
    Thus, in this particular case where the propagation channel is reduced to an AWGN channel (for "Additive White Gaussian Noise" in English), the N decision components D; obtained at the output of the generation module 305 and given in the general case by the equations (Eq-6a) to (Eq-6d) are simplified and the decision component of index k, D ^., is expressed as: in the first aforementioned embodiment (corresponding to the application of a direct Fourier transform on γ (ηΤ ^) 5 ^ (ηΤι.) and on w (nTi.) Sr (nTi.)):
    (Eq-7a) in the above-mentioned second embodiment (corresponding to the application of an inverse Fourier transform on y (nTc) Sr (nTi.) And on w (nTi.) Sr (nTi.)):
    (Eq-7b) in the third aforementioned embodiment (corresponding to the application of a direct Fourier transform on y * (ηΤ ^) 3 ^ (ηΤι ·) and on w * (ηΓ ^)):
    (Eq-7c) in the fourth aforementioned embodiment (corresponding to the application of an inverse Fourier transform on y * (ηΤ ^) 3 ^ (ηΤι ·) and on ιν * (ηΓ ^) 5 ^. ηΓ ^)):
    (Eq-7d)
    As discussed above, in variants, it is the conjugate complex of the real-defining part argument that is taken from the equations (Eq-7a) to (Eq-7d).
    It thus appears on the equations (Eq-7a) to (Eq-7d) that the optimal receiver in AWGN channel in the sense of the maximum likelihood applied to the samples taken at the output of Fourier transform, direct or inverse, involves a term 5¾ (whose expression is given by the equation (Eq-4c)) whose phase varies quadratically as a function of the index of the sample considered in the decision components 0¾ making it possible to estimate the symbol received.
    This quadratic variation is directly related to the quadratic variation of the instantaneous phase of the received signal. Taking into account the particular variation law of this instantaneous phase thus makes it possible to implement the optimum receiver in the sense of maximum likelihood at an analytical cost comparable to that related to the receiver of the state of the art which bases the decision only. on the sample module at the Fourier transform output as described in patent document EP 2 449 690 B1.
    It also appears in this case that the only coefficient related to the propagation channel present in the equations (Eq-7a) to (Eq-7d), i.e. the coefficient Hq, is reduced to a normalization constant independent of the index k. However, it appears that the phase of this term Hq (phase related to the propagation time experienced by the signal received since its transmission) is summed with the phase of other terms depending on k in the argument of the real part function appearing in the equations (Eq-7a) to (Eq-7d). Thus, although independent of k, the term Hq still impacts the index k corresponding to the decision component 0¾. having an extremum of value among the N decision components.
    Moreover, reconsidering the equations (Eq-6a) to (Eq-6d), it now appears for a channel with multiple paths that the coupling terms of Dj ^ weighting the samples for n different from k, are proportional to a coefficient of channel / ffffe-îz [w] depending solely on the difference between the indices of the signal samples considered at the output of the Fourier transform, direct or inverse. Indeed, the invariance in time of the impulse response of the channel leads to terms representative of the interference between symbols depending solely on the difference between the indices of the signal samples considered.
    However, the quadratic variation of the phase of the received signal requires that the coupling between samples is not invariant in time for a given difference between sample indices considered. More particularly, the term 5¾. whose phase varies quadratically as a function of the index of the sample considered, and which is intrinsically linked to the very structure of the waveform used, is here also present.
    Thus, taking these two effects into account in the very structure of the N decision components used to estimate the received symbol makes it possible to implement a receiver in the sense of maximum likelihood in the presence of a propagation channel having multiple paths, while allowing to work in the frequency domain, ie by working on the samples at the output of Fourier transform, direct or inverse.
    During a step E44, an estimate Ic of the rank k of the symbol carried by the received signal is decided from the index of the decision component 0¾ which has an extremum of value among the N components determined during the step E43. In particular, the estimate k corresponds to
    The combination of steps E43 and E44 then makes it possible to implement a step E46 for estimating the received symbol.
    It appears in the light of the expressions of the decision components given by (Eq-6a) to (Eq-6d) or (Eq-7a) to (Eq-7d), that in some embodiments, the channel coefficients H (, I ranging from 0 to N-1, must be known for the implementation of the decision step E44.
    In one embodiment, the channel coefficients H (are initialized to a default value, eg Hq is set to 1 and the channel coefficients H (, I ranging from 1 to N-1, are set to 0 to allow In this way, the reception of first symbols can take place and a obtaining of the channel coefficients H (, I ranging from 0 to N-1 can then be performed, as described below in connection with FIG. step E45, for a subsequent implementation of the decision step E44.
    During a step E45, the N channel coefficients I ranging from 0 to N-1 are thus obtained.
    In one embodiment, the characteristics of the propagation channel are known (eg in a static configuration) and the N channel coefficients obtained then correspond to N predetermined channel coefficients which can be directly loaded at initialization in the decision module. 306.
    In another embodiment, the characteristics of the propagation channel are unknown in advance (eg in the case of receiver and / or transmitter mobility) and the N channel coefficients obtained correspond to N estimated channel coefficients Hi during a substep E451.
    More particularly, the method described bases this estimate on the samples delivered by the discrete Fourier transform module 304 during a prior implementation of the steps E40 to E42 as well as on the rank of at least one corresponding predetermined symbol.
    Alternatively, the predetermined symbols in question are symbols of a learning sequence (e.g., a preamble or a training sequence of a radio frame), thereby allowing a robust estimation of the channel coefficients. In the case of a LoRa® transmission, it is then a plurality of basic chirp signals, ie corresponding to a symbol of rank 0 in the constellation, of positive or negative slope (ie the value of σ varies between +1 and -1 from one chirp to another).
    In another variant, the predetermined symbols in question are data symbols whose rank was previously determined during the execution of a previous step E44, thereby making it possible to refine the estimate of the channel coefficients at during the reception.
    In one embodiment, this estimation is performed on a single received symbol in order to simplify this estimation step and to reduce the overall consumption of the connected object embodying the described technique.
    In another embodiment, this estimate is made on the basis of a plurality of received symbols, thereby allowing the estimate to be averaged to reduce its variance.
    In general, if we consider Ns symbols for estimating the N channel coefficients Hj, I ranging from 0 to N-1, noting kj the rank of the i-th of these Ns symbols in the constellation of N symbols, and η the rank of the reference symbol used when receiving this i-th symbol, the equations (Eq-5a) to (Eq-5d) give us the expression of the N samples of the transformed signal I ranging from 0 to N- 1, obtained at the output of the Fourier transform module 304 in the four aforementioned embodiments upon receipt of this i-th symbol.
    By algebraic manipulation, it is possible to isolate the N channel coefficients Hj in these equations. Thus, adopting a vector notation for greater clarity and noting W the vector whose components are the N coefficients of the channel Hj, we can write from the equations (Eq-5a) to (Eq-5d) that
    (Eq-8) with and with the vector components given by:
    In the first embodiment mentioned above (corresponding to the application of a direct Fourier transform on y (nTc) s ^ (nTc) and on w (nTc) s * (nTc)) by:
    (Eq-9a)
    In the second embodiment mentioned above (corresponding to the application of an inverse Fourier transform on y (nTc) SrinTc ') and on w (nTc) Sy (nTc)) by:
    (Eq-9b)
    In the third embodiment mentioned above (corresponding to the application of a direct Fourier transform on y * (nTr) Sr (nTr) and on M / CnT ^ Sr-Cnr ^)) by:
    (Eq-9c)
    In the fourth embodiment mentioned above (corresponding to the application of an inverse Fourier transform sury * (nTr) Sr (nTr) and on w * inTr) Sr-inTc ')) by:
    (Eq-9d) and with ^ '^' ^ a vector whose l-th components is proportional to the sample Wi obtained at the output of the Fourier transform module 304 when receiving the i-th symbol used for the channel estimation. It thus appears that is a white and centered Gaussian vector.
    The vector H can then be estimated on the basis of a maximum likelihood criterion. Since the probability density of the vector is Gaussian, the vector
    estimated H maximizing the probability density of the symbol observed in reception, knowing that a symbol of rank k has been transmitted, corresponds to the vector H_ minimizing the Gaussian function argument, i.e. the quantity
    where | It refers to the Hermitian norm.
    After developing the square of this norm, it appears that H_ expresses itself as the average over the Ns considered symbols of i.e.
    (Eq-9e) with the vector given by the equations (Eq-9a) to (Eq-9d) according to the aforementioned embodiment.
    A simplification in the estimation of the channel parameters according to one embodiment of the invention will now be described with reference to FIG.
    More particularly, reconsidering the equations (Eq-4a) and (Eq-4b), it appears that the variations of the argument of the function 0jvC); ί · β
    > remain weak around I integer. Indeed, in the LoRa® technology, Te is chosen equal to 8μs, which remains low compared to the dispersion observed in most known radio propagation channels (ie before the differences between the Τρ, p nonzero delays associated with each journey beyond the main journey time, which is often the direct route, and the delay of that main journey). For example, the models of urban propagation channels given in the standardization document "3GPP TS 45.005 V8.8.0: 3rd Generation Partnership Project; Technical Specification Group GSM / EDGE Radio Access Network; Radio Transmission and Receiving ", published by ETSI in April 2010, give time, i.e. deviations corresponding to Τρ - Tq, less than 5μs.
    In this way, assuming a perfect synchronization of the receiver, which amounts to considering Tq = 0 in the preceding equations, a limited development of) around the values of its integer multiple argument leads to being able to express the channel coefficients H; for l Ψ 0 as
    (Eq-10) with Iq a parameter speaking according to the parameters of the propagation channel as
    It thus appears that the set of parameters H i, I ranging from 0 to N-1 can be determined on the basis of only two parameters, which drastically simplifies the channel estimation step.
    In a variant, the two parameters in question are Hq and another one with I different from zero. Indeed, the equation (Eq-10) shows us that the parameters Hj with I different from zero can be deduced from one of them. In this variant, the parameter Hq and the parameter Hj considered can be estimated from the equations (Eq-9e) and (Eq-9a) to (Eq-9d) according to the aforementioned embodiment. Indeed, these parameters Hq and are respectively the first and the 1st component of the vector W defined above and can thus be estimated according to the technique described to estimate this vector.
    In another variant, the two parameters in question are Hq and the parameter Iq introduced into the equation (Eq-10). The parameter Iq can thus be alternatively estimated by injecting the equation (Eq-10) into the equation (Eq-8), leading to
    Then taking a maximum likelihood criterion applied to this equation to determine the Ig parameter, a calculation similar to that described above in relation to obtaining the equation (Eq-9e) gives
    with the vector given by one of the equations (Eq-9a) to (Eq-9d) according to the aforementioned embodiment, with C * the transposed vector of the vector C *, itself being obtained by conjugating each component of C. / g represents in this formula the estimate of 7g.
    In the embodiment where channel estimation is performed on a single received symbol, the above equations remain valid considering Ns = 1.
    Moreover, in the variant where the set of parameters H (, I ranging from 0 to N-1, is determined on the basis of the two parameters Hg and Ig, the expressions of the N decision components Di obtained at the output of the module of generation 305 and allowing the estimation of the emitted symbol, given in the general case by the equations (Eq-6a) to (Eq-6d), are simplified on the basis of the equation (Eq-10), and the component of decision of index k, D ^, is expressed as: in the first aforementioned embodiment (corresponding to the application of a direct Fourier transform on y (nT (.) Sr (nTi.) and on w ( nTc) Sr (nTc)):
    (Eq-11a) in the above-mentioned second embodiment (corresponding to the application of an inverse Fourier transform on y (nTc) Sr (nTi.) And on w (nTc) Sr (nTc)):
    (Eq-llb) in the third aforementioned embodiment (corresponding to the application of a direct Fourier transform on y * (nTc) s ^ (nTc) and on w * (nTc) Sy (nTc)):
    (Eq-llc) in the fourth aforementioned embodiment (corresponding to the application of an inverse Fourier transform on y * (nTc) s ^ (nTc) and on w * (nTc) Sy (nTc)):
    (Eq-lld)
    As indicated previously, in variants, it is the conjugate complex of the argument of the real part defining 0¾. which is taken from the equations (Eq-lla) to (Eq-lld).
    Otherwise. It appears in the light of the equation (Eq-10) (and thus equations (Eq-lla) to (Eq-lld) derived from this equation (Eq-10)) that the approximation of the function 0jvC) ( approximation allowed by the choice of a chip time value Te large in front of the deviations of die associated with each path beyond the principal diai, eg as in the LoRa® technology) in the expression of the coefficients of canai for a non integer null, shows a variation of the amplitude of these terms as the function
    represented in FIG. 5. This shows an exponential decay of the amplitude of the coefficients H; as a function of the index I, the amplitude of the coefficient H ^ o being divided by 10 relative to that of Ηχ.
    Consequently, the effect of the channel can be correctly modeled taking into account only a small number of parameters H i, for example the N 'first channel coefficients of index I ranging from 0 to N'-1, simplifying thus the on-board processing in the receiver for the decision of symbols received in the presence of a propagation channel having multipaths.
    In a variant, the N 'channel coefficients (N' <N) are obtained by applying the general method described above in relation to the equations (Eq-8) and (Eq-9a) to (Eq-9e) applied to the vector
    The vectors to be considered for the implementation of this method, according to the mode of realization among the four modes of realization aforesaid considered, are then those given by the equations (Eq-9a) to (Eq-9d), but restricted to them. No first terms.
    In another variant, the N 'canai coefficients considered are determined from only two parameters as described above in reciation with equations (Eq-10) and following ones (eg Hg and another one of H (with i different from zero, or Hg and Ig.) Again, the vectors under consideration must be restricted to the first N terms.
    In yet another variant, only four potential coefficients are taken into account among the N possibies, but they are not the first prime coefficients, i.e. the canai coefficients of index i less than N '. In this case, the general method described above in relation to equations (Eq-8) and (Eq-9a) to (Eq-9e) is related, but NN 'corresponding channel coefficients are assumed to be harmful, ii The same is true when the N coefficients considered are determined from only two parameters as described above in relation to equations (Eq-10) and following. This makes it possible to simplify the structure of the estimation element of the received symbol when a characteristic of the propagation channel can be presupposed.
    With reference to FIG. 6, the performances obtained when the described technique is used in the case of an AWGN type propagation channel are compared with those obtained when the technique of the prior art is used.
    In this situation, obtaining the channel coefficients according to the described technique performed in step E45 amounts to obtaining a single parameter Hq, the other terms H (being impaired for I ranging from 1 to N-1 as described above in relation with FIG. 4. Moreover, the decision of the rank of the symbols received during step E44 is based in this case on the use of decision components given by the equation among the equations ( Eq-7a) to (Eq-7d) corresponding to the aforementioned embodiment, and determined in step E43.
    According to the technique of the prior art described in patent document EP 2,449,690 B1, the rank of the received symbol is determined solely on the basis of the Fourier transform output sample having the maximum amplitude, independently of any phase information.
    It then appears that the use of the technique described (curve 600b), allows a gain of the order of 1 decibel on the ratio Eb / NO (ie on the ratio of energy per received bit reduced to the spectral density of noise power) necessary to obtain a bit error rate, or BER, given with respect to the known technique (curve 600a). At a given BER, such a gain on the Eb / NO ratio translates directly to the signal-to-noise ratio required at the receiver input. This results in a corresponding gain on the scope of the overall system and therefore on the coverage of the cells of the network in question. In practice, 1 decibel of gain on the signal-to-noise ratio at the input of the receiver corresponds to a consequent increase of the range of 12%.
    The gains expected when the propagation channel has fading are even higher, the technique described allowing indeed to correct the interference between symbols resulting from the muiti paths and thus to improve the discrimination between the emitted symbol and its adjacent.
    FIGS. 7a and 7b show examples of demodulation device structures 300, 300 'of the received symbols making it possible to implement a demodulation method described with reference to FIG. 4 according to various embodiments of the invention.
    The demodulation device 300, 300 'comprises a random access memory 703, 713 (for example a RAM memory), a processing unit 702, 712 equipped for example with a processor, and controlled by a computer program stored in a memory dead 701, 711 (for example a ROM or a hard disk). At initialization, the code instructions of the computer program are for example loaded into the RAM 703, 713 before being executed by the processor of the processing unit 702, 712.
    These FIGS. 7a and 7b only illustrate one of a number of possible ways of realizing the demodulation device 300, 300 'so that it performs certain steps of the method detailed above in connection with FIG. 4 (in any one of different embodiments). Indeed, these steps can be performed indifferently on a reprogrammable calculation machine (a PC computer, a DSP processor or a microcontroller) executing a program comprising a sequence of instructions, or on a dedicated computing machine (for example a set of logical gates such as an FPGA or an ASIC, or any other hardware module).
    In the case where the demodulation device 300, 300 'is realized with a reprogrammable calculation machine, the corresponding program (that is to say the instruction sequence) can be stored in a removable storage medium (such as for example a floppy disk, a CD-ROM or a DVD-ROM) or not, this storage medium being readable partially or totally by a computer or a processor.
    权利要求:
    Claims (13)
    [1" id="c-fr-0001]
    A method of demodulating a received signal, said received signal resulting from the modulation of a basic chirp signal whose instantaneous frequency (102, 102 ', 102 ", 102'") varies linearly between a first instantaneous frequency f0 and a second instantaneous frequency f1 during a symbol time Ts, said modulation corresponding, for a symbol of rank s of a constellation of N symbols, s being an integer from 0 to N-1, to a circular permutation of the variation pattern of said instantaneous frequency on said symbol time Ts, obtained by a time shift of s times an elementary time duration Te, such that N * Tc = Ts, and the transmission of the modulated chirp signal in a transmission channel, characterized in that it comprises a step of estimating (E46) a symbol carried by said received signal, implementing the following substeps: determination (E43) of N decision components, from said received signal and a chirp signal of reference obtained by modulating said basic chirp signal by a reference symbol corresponding to a symbol of rank r in said constellation, a decision component of index I, denoted component Dj, being a function of a term whose phase depends quadratically on I, with I an integer of 0 to N-1; decision (E44) of the rank k of the symbol carried by said received signal, from the decision component, of index k, denoted component having a value extremum among said N decision components.
    [2" id="c-fr-0002]
    2. Method according to claim 1, characterized in that said step of estimating a symbol further comprises the following steps, for N samples of said received signal and for N samples of said reference chirp signal, taken at the same multiple times of Te: conjugation (E40) of said N samples of said reference chirp signal, respectively said N samples of said received signal, delivering N samples of a conjugated chirp signal; multiplying (E41) forward term said N samples of said chirp signal conjugated by said N samples of said received signal, respectively said reference chirp signal, delivering N samples of a multiplied signal; Fourier transform (E42) direct or inverse of said multiplied signal, delivering N samples Yi of a transformed signal, with I an integer ranging from 0 to N-1; and in that said component 0¾ is furthermore a function of a proportion proportional to an amplitude of the sample of said index k, Y1, among said N samples Yi of said transformed signal, as well as the phase of said sample Y / ^.
    [3" id="c-fr-0003]
    3. Method according to claim 2, characterized in that said component 0¾ is furthermore function of a subset of N 'samples among said N samples Yi of said transformed signal, with n different from ok, with N' <N, and with σ a parameter belonging to {-1,1}.
    [4" id="c-fr-0004]
    4. Method according to claim 3, characterized in that the method comprises a step of obtaining (E45) N channel coefficients, and in that a sample of index n of said subset of samples Y ^ is weighted by a coupling coefficient proportional to the channel coefficient / ίσ ^ _ "[Λί] dependent on the difference between the indices ok and n, and to a term whose argument depends quadratically on said index k, and in that said term proportional to an amplitude of the sample is a channel coefficient Hq independent of k.
    [5" id="c-fr-0005]
    5. Method according to claim 4, characterized in that said component is a function of a term proportional to: the real part of the sum

    or the conjugate complex of said sum, when said Fourier transform is a direct Fourier transform and when said conjugate chirp signal corresponds to the conjugation of said reference chirp signal: or the real part of the sum

    or the conjugate complex of said sum, when said Fourier transform is an inverse Fourier transform and when said conjugate chirp signal corresponds to the conjugation of said reference chirp signal; or the real part of the sum

    or the conjugate complex of said sum, when said Fourier transform is a direct Fourier transform and when said conjugate chirp signal corresponds to the conjugation of said received signal; or the real part of the sum

    or the conjugate complex of said sum, when said Fourier transform is an inverse Fourier transform and when said conjugate chirp signal corresponds to the conjugation of said received signal; with

    and with σ a parameter belonging to {-1,1}.
    [6" id="c-fr-0006]
    6. Method according to claim 5, characterized in that said channel coefficients are harmful for n different from ok.
    [7" id="c-fr-0007]
    7. Method according to any one of claims 4 to 6, characterized in that the obtaining step (E45) further comprises an estimate (E451) of said channel coefficients from said N samples of said transformed signal and of at least one symbol k, predetermined.
    [8" id="c-fr-0008]
    The method of claim 7, said estimated channel coefficients forming a vector

    said estimation of said coefficients being carried out on the basis of Ns received symbols, kj designating the rank of the i-th of said Ns symbols in the constellation of N symbols, η denoting the rank of a reference symbol used when receiving said i-th th symbol, designating N samples of said transformed signal obtained during said reception of said i-th symbol, characterized in that said vector H is expressed as

    with

    when said Fourier transform corresponds to a direct Fourier transform and when said conjugated chirp signal corresponds to the conjugation of said reference chirp signal; or

    when said Fourier transform corresponds to an inverse Fourier transform and when said conjugated chirp signal corresponds to the conjugation of said reference chirp signal; or

    when said Fourier transform corresponds to a direct Fourier transform and when said conjugated chirp signal corresponds to the conjugation of said received signal; or

    when said Fourier transform corresponds to an inverse Fourier transform and when said conjugated chirp signal corresponds to the conjugation of said received signal; with

    and with σ a parameter belonging to {-1,1} ·
    [9" id="c-fr-0009]
    9. Method according to any one of claims 7 or 8, characterized in that said step of estimating said channel coefficients comprises the following sub-steps: calculating parameters representative of said channel coefficient Wq and another of said coefficients channel; obtaining parameters representative of the remaining channel coefficients from said calculated parameters.
    [10" id="c-fr-0010]
    10. Process according to any one of claims 4 to 9, characterized in that said channel coefficient of non-zero index I is inversely proportional to


    [11" id="c-fr-0011]
    11. Method according to any one of claims 7 to 10, characterized in that said predetermined symbol is a symbol of a learning sequence or a received symbol whose rank k was decided during a previous execution of said step symbol estimation.
    [12" id="c-fr-0012]
    A computer program product, comprising program code instructions for implementing a method according to any one of claims 1 to 11, when said program is run on a computer.
    [13" id="c-fr-0013]
    13. Device for demodulating (300, 300 ') a received signal, said received signal resulting from the modulation of a basic chirp signal whose instantaneous frequency (102, 102', 102 ", 102 '") varies linearly between a first instantaneous frequency f0 and a second instantaneous frequency f1 during a symbol time Ts, said modulation corresponding, for a symbol of rank s of a constellation of N symbols, s being an integer from 0 to N-1, to a permutation circular pattern of variation of said instantaneous frequency on said symbol time Ts, obtained by a time shift of s times an elementary time duration Te, such that N * Tc = Ts, and the transmission of modulated chirp signal in a transmission channel characterized in that it comprises a reprogrammable calculating machine (702, 712) or a dedicated computing machine (702, 712) adapted and configured to: determine N decision components from said received signal and 'a signal reference chirp obtained by modulating said basic chirp signal by a reference symbol corresponding to a symbol of rank r in said constellation, a decision component of index 1, denoted component Dj, being a function of a term whose phase depends quadratically of 1, with I an integer from 0 to N-1; deciding on the rank k of the symbol carried by said received signal, from the decision component, index k, denoted component D ^, having a value extremum among said N decision components.
    类似技术:
    公开号 | 公开日 | 专利标题
    EP2427992B1|2014-10-29|Method for pre-equalizing a data signal by fdd time reversal
    EP3469719B1|2020-04-29|Method for demodulating a received signal, and corresponding computer program product and device
    EP1941547B1|2009-09-30|Method for transmitting a multi-carrier signal designed for limiting interference, signal, emitting device, receiving method and device, and corresponding computer programs
    EP1482648B1|2006-01-25|Ultra wide band receiver and corresponding reception method
    EP1595373A1|2005-11-16|Wireless data transmission method, and corresponding signal, system, transmitter and receiver
    EP2732592B1|2018-10-31|Method and module for frequency bias estimation in a system for digital telecommunications
    EP1260071B1|2008-01-23|Method and device for estimating channel propagation
    EP3202077B1|2020-09-16|Method of sending a multicarrier signal, method of reception, devices, and computer programs associated therewith implementing an oqam type modulation
    WO2004095729A2|2004-11-04|Multi-carrier signal designed to reduce cell interferences in a radiocommunication network, corresponding receiver, transmitter, construction method and reception method
    EP3469720B1|2020-05-13|Method for generating a pulse-position-modulated signal, demodulation method, and corresponding computer program product and devices
    WO2020260177A1|2020-12-30|Method for generating a signal comprising a succession of chirps over time, method for estimating vehicle symbols using such a signal, computer program products and corresponding devices
    FR2942576A1|2010-08-27|METHOD OF ESTIMATING A CARRIER FREQUENCY OFFSET IN A TELECOMMUNICATION SIGNAL RECEIVER, IN PARTICULAR A MOBILE DEVICE
    EP1296460B1|2006-04-19|Method and apparatus for determining the fading coefficients of a multipath transmission channel connecting in particular a base station and a mobile telephone
    EP1288675A1|2003-03-05|Method and device for estimating the velocity of a mobile terminal, in particular, of a cellular telephone
    WO2015193491A1|2015-12-23|Method and system for acoustic communication
    FR3109681A1|2021-10-29|A method of estimating at least one signal timing information comprising a plurality of chirps, corresponding computer program and device product.
    FR3108815A1|2021-10-01|A method of estimating symbols conveyed by a signal comprising a plurality of chirps, corresponding computer program and device product.
    CA2781137C|2018-03-20|Emission-reception process and system for a digital signal over radio
    FR2802043A1|2001-06-08|RADIO CHANNEL ESTIMATION AND DEMODULATION DEVICES
    EP3158700B1|2018-08-29|Method for estimating a radioelectric propagation channel
    FR3102320A1|2021-04-23|Selection method for optimizing ambient backscatter communications, and associated system
    FR2938988A1|2010-05-28|METHOD FOR CONTINUOUS PHASE MULTISTATE MODULATION AND TRANSMITTER USING THE METHOD.
    同族专利:
    公开号 | 公开日
    JP2019518383A|2019-06-27|
    AU2017278823A1|2018-12-13|
    US10686488B2|2020-06-16|
    WO2017211552A1|2017-12-14|
    EP3469719B1|2020-04-29|
    CN109314539B|2021-01-19|
    EP3469719A1|2019-04-17|
    JP6974359B2|2021-12-01|
    FR3052615B1|2019-11-01|
    CA3024672A1|2017-12-14|
    US20190149187A1|2019-05-16|
    KR20190015291A|2019-02-13|
    CN109314539A|2019-02-05|
    KR102303021B1|2021-09-16|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    EP2449690A1|2009-07-02|2012-05-09|Nanoscale Labs|Communications system|
    EP2763321A1|2013-02-05|2014-08-06|Semtech Corporation|Low power long range transmitter|
    EP2975814A1|2014-07-18|2016-01-20|Semtech Corporation|Chirp Signal Processor|
    US7245930B1|1998-11-24|2007-07-17|Hughes Network Systems, Llc|Acquisition mechanism for a mobile satellite system|
    US20070041457A1|2005-08-22|2007-02-22|Tamer Kadous|Method and apparatus for providing antenna diversity in a wireless communication system|
    FR2894413A1|2005-12-07|2007-06-08|France Telecom|METHOD AND DEVICE FOR DYNAMIC INTERLACING|
    US7974360B2|2006-05-24|2011-07-05|Qualcomm Incorporated|Multi input multi output orthogonal frequency division multiple access communication system|
    US7872961B2|2006-06-29|2011-01-18|Motorola Mobility, Inc.|Orthogonal frequency division multiple access message processing method and apparatus|
    KR100992477B1|2008-06-27|2010-11-08|성균관대학교산학협력단|Frequency offset estimation scheme for system having a plurality of sub-carriers|
    GB2524944A|2014-02-24|2015-10-14|Neul Ltd|Low complexity GMSK receiver for fast varying channels and narrow bandwidth channels|EP3273607A1|2016-07-20|2018-01-24|Semtech Corporation|Method and system of timing and localizing a radio signal|
    FR3098066A1|2019-06-25|2021-01-01|Université De Bordeaux|A method of generating a signal comprising a temporal succession of chirps, a method of estimating symbols conveyed by such a signal, computer program products and corresponding devices.|
    FR3108815A1|2020-03-24|2021-10-01|Universite de Bordeaux|A method of estimating symbols conveyed by a signal comprising a plurality of chirps, corresponding computer program and device product.|
    法律状态:
    2017-06-23| PLFP| Fee payment|Year of fee payment: 2 |
    2017-12-15| PLSC| Publication of the preliminary search report|Effective date: 20171215 |
    2018-06-26| PLFP| Fee payment|Year of fee payment: 3 |
    2019-06-27| PLFP| Fee payment|Year of fee payment: 4 |
    2020-06-29| PLFP| Fee payment|Year of fee payment: 5 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1655322A|FR3052615B1|2016-06-09|2016-06-09|METHOD OF DEMODULATING A RECEIVED SIGNAL, COMPUTER PROGRAM PRODUCT AND CORRESPONDING DEVICE|
    FR1655322|2016-06-09|FR1655322A| FR3052615B1|2016-06-09|2016-06-09|METHOD OF DEMODULATING A RECEIVED SIGNAL, COMPUTER PROGRAM PRODUCT AND CORRESPONDING DEVICE|
    EP17722839.2A| EP3469719B1|2016-06-09|2017-05-16|Method for demodulating a received signal, and corresponding computer program product and device|
    KR1020187035537A| KR102303021B1|2016-06-09|2017-05-16|Received signal demodulation method, corresponding computer program and apparatus|
    CA3024672A| CA3024672A1|2016-06-09|2017-05-16|Procede de demodulation d'un signal recu, produit programme d'ordinateur et dispositif correspondants|
    PCT/EP2017/061758| WO2017211552A1|2016-06-09|2017-05-16|Method for demodulating a received signal, and corresponding computer program product and device|
    US16/307,059| US10686488B2|2016-06-09|2017-05-16|Method for demodulating a received signal, corresponding computer program and device|
    JP2018564734A| JP6974359B2|2016-06-09|2017-05-16|How to demodulate received signals, corresponding computer program products and devices|
    CN201780035749.1A| CN109314539B|2016-06-09|2017-05-16|Method, computer-readable storage medium, and apparatus for demodulating received signal|
    AU2017278823A| AU2017278823A1|2016-06-09|2017-05-16|Method for demodulating a received signal, and corresponding computer program product and device|
    [返回顶部]