Method and device for multi-service transmission with FC-OFDM modulation and corresponding receiver

Abstract

A method implementing the same frequency-time transform of size M irrespective of the service. The method adds, during a frame setup, a cyclic extension of L=L.sub.1+L.sub.2 samples in order to obtain a sequence of M+L samples. The method carries out a time-domain filtering according to a function (n) of the samples n of the sequence of M+L samples.

Claims

1. A method comprising: transmitting data corresponding to a service from amongst several communications services implementing a same frequency-time transform (IFFT) of size M irrespective of the service in order to generate orthogonal multi-carrier symbols of M samples starting from data symbols including: during a frame setup, adding a cyclic extension of L=L.sub.1+L.sub.2 samples in order to obtain a sequence of M+L samples, and setting up the frame according to a configuration depending on the service, which allows a selection between no, one or several processing operations from amongst a padding, a discrete Fourier transform of size N and a spread for adapting the data prior to mapping at an input of the frequency-time transform, and the method comprises time-domain filtering according to a function (n) of the samples n of the sequence, L and M being non-zero natural numbers, L<M: $f\left(n\right)=\{\begin{array}{cc}a\left(n\right)& n\left[0,L-1\right]\\ 1& n\left[L,M-1\right]\\ b\left(n\right)& n\left[M,M+L-1\right]\end{array}$ with a(n) and b(n) monotonic functions and with a.sup.2 (n)+b.sup.2 (n+M)=1 for n[0, L1], L.sub.10 and L.sub.20.

2. The method as claimed in claim 1, in which the function (n) is non-symmetrical: (n)(M+L1n) for n[0, L1].

3. The method as claimed in claim 1, in which the cyclic extension comprises a cyclic prefix of L.sub.1 samples and a cyclic suffix of L.sub.2 samples, L.sub.10 and L.sub.20.

4. The method as claimed in claim 1, in which the transmitting complies with a time-frequency frame with preambles multiplexed within the frame according to a configuration depending on the service under a constraint of a maximum number of preambles, the configuration depending on the service being able to allow a time-domain multiplexing of the preambles or a time-domain and frequency-domain multiplexing of the preambles to be selected.

5. The method as claimed in claim 1 in which the spread has a spread factor P and each spread data value is multiplied by a weighting w.sub.p, p=1, . . . , P.

6. The method as claimed in claim 1 in which $a\left(n\right)=\cos \left(\frac{}{2L}\left(L-1-n\right)\right)\mathrm{and}b\left(n\right)=\sin \left(\frac{}{2L}\left(M+L-1-n\right)\right).$

7. The method as claimed in claim 1 in which $a\left(n\right)=\cos \left(\frac{}{2L^2}{\left(L-1-n\right)}^2\right)\mathrm{and}b\left(n\right)=\sin \left(\frac{}{2L^2}{\left(M+L-1-n\right)}^2\right).$

8. The method as claimed in claim 1 in which $b\left(n\right)=e^{-\left({\left(\frac{n-M}{L}\right)}^{}\right)}$ and a(n)={square root over (1b.sup.2 (n+M))} for n[0,L1], and being parameters having a real value strictly greater than zero.

9. The method as claimed in claim 1 in which, during the frame setup, two successive symbols overlap in time by a value D, D being an integer greater than or equal to zero.

10. The method as claimed in claim 1, comprising transmission of a signaling message coding the configuration.

Description

LIST OF THE FIGURES

(1) Other features and advantages of the invention will become apparent during the description that follows presented with regard to the appended figures given by way of non-limiting example.

(2) FIG. 1 is a diagram of the transmission chain of the waveform DFT-s-OFDM according to the prior art.

(3) FIG. 2 is a diagram of the structure of a frame generated by the base station which has a duration of 10 ms and is composed of sub-frames of duration 1 ms according to the LTE standard.

(4) FIG. 3 is a diagram of a sub-frame which comprises 14 DFT-s-OFDM symbols, each DFT-s-OFDM symbol being preceded by a cyclic prefix CP according to the LTE standard.

(5) FIG. 4 is a diagram illustrating the uplink synchronization mechanism of the LTE standard showing the return time A&R between the base station SB and the terminal UE.

(6) FIG. 5 is a diagram showing a time offset to between the reference frame (base station) for the synchronization and the received frame.

(7) FIG. 6 is a diagram showing the time offset to in the DFT-s-OFDM symbol, together with the positioning of the window for extraction of the CP and of the window for the FFT with respect to the DFT-s-OFDM symbol.

(8) FIG. 7 is a diagram identical to that in FIG. 6 with an offset to greater than the window for extraction of the CP.

(9) FIG. 8 is a diagram identical to that in FIG. 5 with a negative offset to.

(10) FIG. 9 is a diagram identical to that in FIG. 6 with a negative offset to.

(11) FIG. 10 is a diagram illustrating the synchronization mechanism between the base station and each terminal UE1, UE2, UE3 with the sending of a signal containing the time value ta1, ta2, ta3.

(12) FIG. 11 is a diagram illustrating a coarse synchronization mechanism for an IoT service.

(13) FIG. 12 is a diagram of the transmission technique according to the invention implemented by a terminal.

(14) FIG. 13 is a diagram illustrating the addition of a cyclic prefix and suffix onto a symbol at the IFFT output, together with the effect of the filtering according to one embodiment.

(15) FIG. 14 is a diagram illustrating a frame setup with time-domain overlap between successive symbols.

(16) FIG. 15 shows a sub-frame of 1 ms according to the LTE standard in which the preambles are multiplexed at the 4.sup.th and the 11.sup.th position.

(17) FIG. 16 illustrates the simultaneous time-domain and frequency-domain multiplexing of the preambles according to the invention.

(18) FIG. 17 is one example of a spread.

(19) FIG. 18 is a diagram of one example of implementation of a transmission method according to the invention.

(20) FIGS. 19 and 20 are diagrams respectively illustrating two embodiments of the reception according to the invention.

(21) FIG. 21 is a diagram of one embodiment of the folding involved in the embodiment in FIG. 19.

DESCRIPTION OF EMBODIMENTS OF THE INVENTION

(22) The transmission technique according to the invention is illustrated in FIG. 12. The transmission method 1 according to the invention implemented by a terminal UE generates a new multi-carrier modulation format configurable according to the service with the implementation of the same frequency-time transform of size M and with the same processing after IFFT irrespective of the service which allow the simultaneous processing at the data reception associated with various services. This scheme is called FC-OFDM.

(23) The new modulation format is obtained with the addition 3 (+FIXs) of a cyclic prefix and/or suffix to the orthogonal multi-carrier symbols generated 2 by the frequency-time transform IFFT combined with a filtering 4 by a filter WDG applied over M+L samples. L and M are non-zero natural numbers. Denoting s.sub.1 [ ] the sequence to be mapped at the input of IFFT, s.sub.2 [ ] the sequence prior to addition of a cyclic prefix and/or suffix and s.sub.3 [ ] the sequence after the filtering WDG of function , then the following may be written:
s.sub.2[m] for m[0,M1]
s.sub.3[n]=s.sub.2[mod(nU,M)][n] for n[0,M L1] and U=L.sub.1.

(24) According to this expression, the sequence obtained s.sub.3 [n] may either comprise a prefix of L.sub.1 samples, or a suffix of L.sub.2 samples, or a prefix and a suffix respectively of L.sub.1 and of L.sub.2 samples.

(25) FIG. 13 illustrates the addition, according to one embodiment, of a cyclic prefix and suffix to an orthogonal multi-carrier symbol, together with the filtering according to the invention. The method copies the last L.sub.1 samples of a multi-carrier symbol supplied by the IFFT of size M at the start of this symbol thus creating a cyclic prefix. The method furthermore copies the first L.sub.2 samples of the same multi-carrier symbol at the end of this symbol thus creating a cyclic suffix. The symbol then comprises M+L samples, with L=L.sub.1+L.sub.2. The filtering is applied over the M+L samples, which amounts to filtering the samples n of the sequence with a function (n):

(26) $f\left(n\right)=\{\begin{array}{cc}a\left(n\right)& n\left[0,L-1\right]\\ 1& n\left[L,M-1\right]\\ b\left(n\right)& n\left[M,M+L-1\right]\end{array}$

(27) with a.sup.2 (n)+b.sup.2 (n+M)=1 for n[0, L1] and with a(n) and b(n) monotonic functions. Hence, a(n)={square root over (1b.sup.2 (n+M))} for n[0, L1].

(28) According to one embodiment,

(29) $a\left(n\right)=\cos \left(\frac{}{2L}\left(L-1-n\right)\right)\mathrm{and}b\left(n\right)=\sin \left(\frac{}{2L}\left(M+L-1-n\right)\right).$

(30) According to one embodiment,

(31) $a\left(n\right)=\cos \left(\frac{}{2L^2}{\left(L-1-n\right)}^2\right)\mathrm{and}b\left(n\right)=\sin \left(\frac{}{2L^2}{\left(M+L-1-n\right)}^2\right).$

(32) According to one embodiment,

(33) $b\left(n\right)=e^{-\left({\left(\frac{n-M}{L}\right)}^{}\right)}$
and a(n)={square root over (1b.sup.2 (n+M))}, n[0, L1], and are parameters having a real value strictly greater than zero. Typically, their value is adjusted during simulations.

(34) FIG. 14 illustrates the frame setup for three successive symbols with a time-domain overlap D between two successive symbols. D is an integer greater than or equal to zero whose value is typically determined during simulations. D is a parameter which corresponds to a number of samples of a FC-OFDM symbol onto which other samples of the next FC-OFDM symbol are superposed.

(35) The transmission complies with a time-frequency frame with preambles multiplexed within the frame. According to one embodiment, the multiplexing is configurable, depends on the service and is under the constraint of a maximum number of preambles. Depending on the service, the configuration allows, with reference to FIG. 12, a time-domain multiplexing of the preambles or a time-domain and frequency-domain multiplexing of the preambles to be selected 5. In an uplink (UL) frame according to the LTE standard, the preambles are always transmitted at a fixed position with an interval of 0.5 ms. FIG. 15 shows a sub-frame of 1 ms complying with the LTE standard in which the preambles are multiplexed at the 4.sup.th and the 11.sup.th position. The preambles are therefore only multiplexed in time. According to the invention, the multiplexing may be carried out simultaneously in time and in frequency as illustrated in FIG. 16. According to the example illustrated, the time interval between the preambles is reduced which increases the robustness with respect to the Doppler effect and allows a time-domain estimation of the channel compatible for example with a V2X service. In contrast, the preambles are multiplexed in frequency which allows the same band consumed (overhead) by the preambles to be conserved.

(36) The method according to the invention is flexible with an adaptation of the data before frequency-time transformation according to a configuration depending on the service in order to generate data symbols.

(37) The configuration depending on the service allows, with reference to FIG. 12, no, one or several processing operations to be selected from amongst a padding 7, 0 PAD, a discrete Fourier transform 8, DFT, of size N, a spread 9, SPG.

(38) The padding consists in adding zeros (zero padding in English) in order to adjust the number of data at the input.

(39) The spread allows the diversity of the data to be increased. One example of a spread is illustrated by FIG. 17. According to this example, the data values at the input are taken in pairs, a and b, and each one is weighted with P weightings, w.sub.p, p=1, . . . , P, in order to generate two times P spread data: aw.sub.1, . . . , aw.sub.p, bw.sub.1, . . . , bw.sub.p. The spread factor P is strictly greater than one. The weightings may have a constant value: |w.sub.1|=|w.sub.2|= . . . =|w.sub.P| or a non-constant value and a constant sign or a non-constant sign.

(40) FIG. 18 illustrates one example of implementation of a method of transmission according to the invention. Four terminals UE transmit data associated with different services. The base station BS simultaneously receives (FDMA access) the data coming from the four terminals. The 1.sup.st terminal transmits data associated with an MBB service. The 2.sup.nd terminal transmits data associated with a V2X service. The 3.sup.rd terminal transmits data associated with a MCC service. The 1.sup.st, 2.sup.nd and 3.sup.rd terminals benefit from a synchronization mechanism (LTE SYNC) with respect to the base station BS. The 4.sup.th terminal transmits data associated with an IoT service and does not benefit from a synchronization mechanism or benefits from a coarse synchronization mechanism (coarse SYNC) with respect to the base station BS. Each terminal has its own configuration that it transmits to the base station. The 1.sup.st terminal is configured with activation of the DFT (p2, FIG. 12) and of the time-domain multiplexing of the preambles (PREAM LTE). The 2.sup.nd terminal is configured with activation of the DFT (p2, FIG. 12) and of the time-domain and frequency-domain multiplexing of the preambles (PREAM Interl). The 3.sup.rd terminal is configured with activation of the DFT (p2, FIG. 12), of the spread (p3, FIG. 12) and of the time-domain multiplexing of the preambles (PREAM LTE). The 4.sup.th terminal is configured with activation of the padding, of the DFT (p2, FIG. 12) and of the time-domain multiplexing of the preambles (PREAM LTE).

(41) The reception at the base station is illustrated in FIGS. 19 and 20 which respectively correspond to two embodiments. Irrespective of the service, the reception implements the same time-frequency transform FFT.

(42) According to the 1.sup.st embodiment, FIG. 19, the received FC-OFDM symbols of length M+L samples are filtered by the filter WDG, then undergo a folding implemented by a computer FOLD+ prior to the FFT of size M. The filter WDG is the same as that used in transmission, and implements the function (n). The folding according to one embodiment is illustrated in FIG. 21. The first L.sub.1 samples of the filtered FC-OFDM symbol are added to the samples that precede the last L.sub.2 samples of this symbol and the first L.sub.1 samples of the filtered FC-OFDM symbol are eliminated. The last L.sub.2 samples of the filtered FC-OFDM symbol are added to the samples that follow the first L.sub.1 samples of the filtered FC-OFDM symbol and the last L.sub.2 samples of the filtered FC-OFDM symbol are eliminated. This embodiment is adapted to the transmission of a FC-OFDM symbol with a cyclic prefix and suffix of L.sub.1 and L.sub.2 samples, respectively.

(43) The FFT is applied over the M samples obtained after additions which, according to the embodiment illustrated in FIG. 21, correspond to the M central samples, i.e. according to the example L.sub.1=L.sub.2. The samples at the FFT output are de-mapped by a de-mapper MAP then the processing that follows depends on the configuration used in the transmission by the terminal.

(44) According to the 2.sup.nd embodiment, FIG. 20, the received symbols FC-OFDM of length M+L are padded with zeros in order to obtain a sequence of length 2M. This sequence is transformed in the frequency domain with a FFT of size 2M. The frequency-domain samples are filtered with a filter FLG of function F(m) such that:

(45) 0$F\left(m\right)=\underset{n=0}{\overset{2M-1}{.Math.}}f\left(n\right)e^{-\frac{j2\mathrm{mn}}{2M}}.$

(46) After filtering, only the samples of even index are conserved at the output of the selector EEI. The samples at the output of the selector are de-mapped by a de-mapper MAP then the processing that follows depends on the configuration used in the transmission by the terminal.

(47) [IEEE Access] Hao LIN, Flexible Configured OFDM for 5G Air Interface, IEEE ACCESS, 1 Jan. 2015, XP055276445

(48) [3GPP] Motivation for new WI on Low Complexity and Enhanced Coverage LTE UE for MTC, 3GPP TSG RAN Meeting #64 RP-140845, Sophia Antipolis, France, 10-13 Jun. 2014

(49) Although the present disclosure has been described with reference to one or more examples, workers skilled in the art will recognize that changes may be made in form and detail without departing from the scope of the disclosure and/or the appended claims.