Methods for transmitting and receiving spectrally spread wireless communication with high potential spectral efficiency and associated devices

Abstract

The invention relates to a method for transmitting wireless communication, comprising: obtaining symbols by channel coding and digital modulation; DSSS-type spreading of each symbol obtained by a TDSSS period-spreading code applied with a spreading factor, the SF chips spreading a symbol being specific to the instant of the symbol's transmission; and obtaining, from each symbol, a set of spread SF chips as a function of the symbol's value, these chips being separated by a period; implementing on the chips: - compression of the period between the chips by a compression ratio strictly less than 1, the period between the chips after said compression being equal to , - filtering by a shaping filter, in the form of a Nyquist root with period of orthogonality; or a useful bandwidth of finite length. Figure for the abstract: Fig. 1

Description

Title of the invention: Methods for transmitting and receiving spectrally spread wireless communication with high potential spectral efficiency and associated devices. Technical field

[0001] The invention relates to the field of spectrally spread wireless telecommunications as well as to that of high spectral efficiency systems. Prior art

[0002] DSSS (Direct Sequence Spread Spectrum) is a spectral spreading technique using "pseudo-noise" spreading codes over a frequency band larger than the minimum required (R. Pickholtz, D. Schilling, and L. Milstein, “Theory of spread-spectrum communications - a tutorial,” IEEE Transactions on Communications, vol. 30, no. 5, pp. 855-884, 1982.).

[0003] In the case of spectrally spread radio transmission signals, a distinction is generally made between short codes whose period is less than the symbol duration (or by extension less than a few symbol durations), and long codes whose period corresponds at least to a high multiple of symbol durations.

[0004] In the civilian sector, this technique allows multiple users to employ orthogonal or quasi-orthogonal codes to simultaneously share the same frequency band while ensuring low interference (Code Division Multiple Access or CDMA [Kwang Soon Kim, Lickho Song, Yun Hee Kim, Yong Up Lee, and Jooshik Lee, “Analysis of quasi-ML multiuser detection of DS / CDMA Systems in asynchronous channels,” IEEE Transactions on Communications, vol. 47, pp. 1875–1883, Dec. 1999]). In the military sector, its main advantage is its robustness to jammers (thanks to the spreading gain) and interception (thanks to the low power spectral density, which can be lowered to induce reception levels below the noise level).

[0005] A corollary to this application is that such so-called "discrete" communications may be authorized to use allocated bands as a secondary user (SU). Indeed, this new user generates little interference (thanks to its low DSP) while being robust against the primary user's signal.

[0006] Unfortunately, by spreading the signal spectrum, the spectral efficiency is significantly reduced. However, this is one of the important criteria to take into account when designing a waveform.

[0007] This problem has already been raised in the context of non-spread links. Among the techniques improving spectral efficiency, one can cite, for example:

[0008] - The choice of the formatting filter:

[0009] By using, for example, a raised cosine square root shaping filter, it is possible to reduce the bandwidth of the useful signal and thus increase spectral efficiency by adjusting a parameter called roll-off. Unfortunately, reducing the bandwidth also increases the filter's time support and therefore the complexity of reception.

[0010] - The choice of modulation:

[0011] By increasing the modulation order at constant band (and constant antenna output power), the number of bits transmitted per symbol is increased, and therefore the spectral efficiency, but the performance is degraded in terms of bit error rate (BER).

[0012] - The use of Faster-Than-Nyquist (FTN) links

[0013] The FTN consists of transmitting at a higher baud rate than the so-called Nyquist rates. However, this benefit comes at the cost of the appearance of intrinsic intersymbol interference (IESI) from the moment of transmission, and of more complex receivers to process it.

[0014] There is therefore a need to have a spread transmission chain that can ensure a certain energy discretion while maintaining reasonable spectral efficiency. Summary of the invention

[0015] To this end, according to a first aspect, the present invention describes a method for transmitting wireless communication, said method being implemented in a wireless telecommunication transmitting device receiving as input a signal to be transmitted, said method comprising the following steps implemented by the transmitting device:

[0016] application of a channel coding, including error-correcting coding and / or interleaving, on the signal to be transmitted, followed by digital modulation of said coded signal to obtain symbols;

[0017] said process being characterized in that it further comprises the following steps implemented by the emission device:

[0018] DSSS type spreading of each symbol obtained by a period spreading code TDSSs applied with a spreading factor SF, the SF chips spreading a symbol being specific to the instant of emission of the symbol; and obtaining, from each symbol, a set of spread SF chips as a function of the value of the symbol, these chips being separated by a period Tc;

[0019] implementation on chips obtained after spreading a treatment comprising:

[0020] - the compression of the period between the chips by a strictly less than 1, the period between chips after said compression being equal to ^Tc;

[0021] - filtering by a formatting filter vj^t):

[0022] - in Nyquist root of orthogonality period Tc; or

[0023] - of useful strip of finite length B = 1 / Tc-

[0024] The invention thus proposes a new configurable waveform that increases the spectral efficiency of conventional DSSS links and provides a degree of energy discretion to conventional FTN links. Various reception schemes associated with different complexities are also proposed.

[0025] In embodiments, such a method will further comprise at least one of the following features:

[0026] The signal Se(t) at the end of the filtering is defined by the following equation:

[0027] 100281 s e (t) (nSF+q)ÇT c )

[0029] with: - Ç: the compression ratio, between 0 and 1; - j_1: the chips resulting from the spread symbols; lCiiSF+ <iJ • in the case of a single-code spreading: cnSF+q = àn^nSF+q • in the case of a two-code spread: CnSF+q — -^e( &n ) UlqiSF+q + j^m( ) ^2,nSF+q - Tc: the time separating the uncompressed chips; - SF; the spreading factor; - Ve(t): the impulse response of the shaping filter. - an: the symbol to be transmitted (output of the channel coding and modulation) numerical) with Re( an) the real part of the symbol and Im( 3.a) the imaginary part of the symbol; - In louJ„ letj„ 1: the monocode or dual code lunSF+qJ (“XnSF+qJ lu2jzSF+qJ spreading used.

[0030] According to another aspect, the invention describes a computer program intended to be stored in the memory of a wireless telecommunications transmitting device further comprising a microcomputer, said computer program comprising instructions which, when executed on the microcomputer, implement the steps of the transmitting process described above.

[0031] According to another aspect, the invention describes a method for receiving wireless communication implemented in a receiving device wireless telecommunication comprising at least one receiving antenna receiving an input transmitted signal, said method comprising the following steps implemented by the receiving device on a signal obtained via the receiving antenna(s):

[0032] application of digital demodulation and channel decoding on the signal;

[0033] said process being characterized in that it further comprises the following steps implemented by the receiving device:

[0034] implementation of a signal processing including filtering by a filter -t ) and to the channel ]j*[_ such that the filter y* ( -t ) is adapted to a shaping filter Ve(t) in Nyquist root of orthogonality period Tc or of useful bandwidth of finite length B= 1 / Tc;

[0035] sampling period c of the signal from the filter adapted to the global channel

[0036] despreading of the sampled signal as a function of the spreading code taken over a length equal to the spreading factor SF, the series of SF chips corresponding to the spreading code on a symbol being specific to the time of reception of said symbol, the SF chips being separated by a period ^Tc; and obtaining symbols of duration ÇT=SFÇTC.

[0037] In some embodiments, such a receiving process will further comprise one at least the following characteristics:

[0038] in a SIMO receiver according to which the channel h(t) and its matching filter are replaced } by vectors h(t) = 'h^t) , hj / t) with N being the number of antennas upon receipt;

[0039] the process includes, after the despreading, at least one of the following steps:

[0040] an equalization of the despreaded symbols implemented by an equalizer block (28) of the receiving device;

[0041] a selection of an element from a constellation of elements, according to the unspread symbols, by the implementation of a Viterbi algorithm or by a decision block of said receiving device.

[0042] According to another aspect, the invention describes a computer program intended to be stored in the memory of a wireless telecommunications receiving device further comprising a microcomputer, said computer program comprising instructions which, when executed on the microcomputer, implement the steps of a receiving process as described above.

[0043] According to another aspect, the invention describes a wireless communication transmission device adapted to receive as input a signal to be transmitted, the transmission device being adapted to apply channel coding, including error-correcting coding and / or interleaving, to the signal to be transmitted, and then digital modulation of said coded signal to obtain symbols;

[0044] said transmission device characterized in that it is adapted to perform DSSS type spreading of each symbol obtained by a period spreading code Tdsss applied with a spreading factor SF, the SF chips spreading a symbol being specific to the instant of emission of the symbol, and to obtain, from each symbol, a set of spread SF chips as a function of the value of the symbol, these chips being separated by a period Tc;

[0045] said emission device being adapted to implement on the chips obtained after spreading, a treatment comprising:

[0046] - the compression of the period between the chips by a strictly less than 1, the period between the chips after said compression being equal to Ç Tc;

[0047] - filtering by a formatting filter v^t):

[0048] - in Nyquist root of orthogonality period Tc; or

[0049] - of useful strip of finite length B= 1 / Tc-

[0050] In one embodiment, the signal obtained by said transmitting device after filtering is defined by the following equation: 100511 s e (t) = (nSF+q)ÇT c )

[0052] with:

[0053] - : the compression ratio, between 0 and 1;

[0054] - r 1: the chips resulting from the spread symbols; ^nSF+qj

[0055] in the case of a single-code spread: CnSF+q = ^n^nSF+q

[0056] in the case of a two-code spread: CnSF+q — Re( ) UljiSF+q + ) ^2jiSF+q

[0057] - Tc: the time separating the uncompressed chips;

[0058] - SF; the spreading factor;

[0059] - ve(t): the impulse response of the shaping filter.

[0060] - an: the symbol to be transmitted (output of the channel coding and modulation digital) with Re( aü) the real part of the symbol and Im( the imaginary part of the symbol;

[0061] - louf„ letj„ 1: the single or two-code spreading \UnSF+qJ ]Ul,nSF+qJ l^nSF+qJ used.

[0062] According to another aspect, the invention describes a wireless communication receiving device, adapted to receive as input via one or more receiving antennas a transmitted signal, said receiving device being adapted to apply to the signal obtained, via the receiving antenna(s), a digital demodulation and a channel decoding;

[0063] said device being characterized in that it is adapted to implement signal processing comprising filtering by a filter y* ( -t ) and to the channel such that the filter y*( - t) is adapted to a shaping filter y^f) in Nyquist root of orthogonality period Tc or of useful bandwidth of finite length B = 1 / T c;

[0064] said device is adapted to perform period sampling ÇTc of the signal from the filter adapted to the global channel g*( — v*( *h*(- ;

[0065] said device is adapted to perform a despreading of the sampled signal as a function of the spreading code taken over a length equal to the spreading factor SF, the series of SF chips corresponding to the spreading code on a symbol being specific to the time of reception of said symbol, the SF chips being separated by a period ^Tc; and to obtain symbols of duration ÇT = SFÇTC. Brief description of the drawings

[0066] The invention will be better understood and other features, details and advantages will become clearer from the following description, given by way of non-limiting reason, and from the accompanying figures, given by way of example.

[0067] [Fig.1] Fig.1 is an illustration of a transmitting device in an embodiment implemented by the invention;

[0068] [Fig.2] The [Fig.2] is a diagram representing the steps of the emission process according to an embodiment of the present invention;

[0069] [Fig.3] The [Fig.3] is a diagram of a receiving device according to an embodiment of the present invention;

[0070] [Fig.4] The [Fig.4] is a diagram representing the steps of the process according to another embodiment of the invention.

[0071] Identical references may be used in different figures when they refer to identical or comparable elements. Description of the implementation methods

[0072] A wireless telecommunication system implementing the invention, in one embodiment, comprises a wireless telecommunication transmitting device 10, hereinafter referred to as transmitter 10, and a wireless telecommunication receiving device 20, hereinafter referred to as receiver 20.

[0073] The wireless telecommunications transmitting device 10 according to the invention is adapted to, as a function of a source signal supplied to it as input, determine symbols modulated by a digital modulation in a known way, then to apply to said symbols a direct sequence spread spectrum, called DSSS (for Direct Sequence Spread Spectrum), and finally to apply a Faster-than-Nyquist type processing, called FTN, in transmission.

[0074] The wireless telecommunications receiver 20 is adapted to receive a communication signal transmitted by the transmitter 10, and then to apply to the received signal Faster-than-Nyquist (FTN) spread processing, followed by despadding processing. The characteristics of the FTN processing at the receiver correspond to the characteristics of the FTN processing performed in the transmitter 10. The characteristics used for the despadding performed by the receiver 20 are similar to the characteristics of the spreading performed by the transmitter 10.

[0075] In one embodiment of the invention, the transmitter 10 comprises, with reference to [Fig.1], a coding block 11, an interlacing block 12, a modulation block 13, a DSSS spreading block 14, a compression block 15, a filtering block 16 and a radio frequency (RF) transmitting antenna 17.

[0076] The function of the blocks is indicated below and is further completed later, with reference to [Fig.2].

[0077] The three blocks 11, 12, 13 thus perform channel coding and source coding, by methods known to those skilled in the art.

[0078] The coding block 11 is adapted to receive a source signal consisting of a series of bits and is adapted to perform channel coding.

[0079] The interleaving block 12 is adapted to receive the bits from the channel coding block 11, to interleave the received bits together in order to introduce temporal diversity, in order to protect the signal from the phenomenon known as fading.

[0080] The modulation block 13 is adapted to apply digital modulation to the bit sequences output from the interleaving block 12, i.e., to assign to each bit sequence a state (or symbol), corresponding to the bit sequence, from among the M possible states in the constellation associated with the chosen modulation. The digital modulation is, for example, of the ASK, mQAM, mPSK, mAPSK type.

[0081] The DSSS 14 spreading block is adapted to implement a direct sequence spreading (DSSS) technique, which is a well-known spreading technique to those skilled in the art: the input signal of the DSSS 14 block is combined with arbitrary modulation spreading codes, which may differ in the real and imaginary parts of the symbol and have a higher rate than the useful symbols, thus dividing the signal data according to a spreading ratio: cf. R. Pickholtz, D. Schilling and L. Milstein, "Theory of Spread- Spectrum Communications - A Tutorial," IEEE Transactions on Communications, vol. 30, no. 5, pp. 855-884, May 1982. Tc is the period separating the chips of the spreading code.

[0082] The compression block 15 and the filtering block 16 are adapted to implement the FTN technique in transmission. This FTN technique is described in detail, in particular, in J. Mazo, Faster-Than-Nyquist signaling, Bell System Tech Journal, vol. 54, no. 8, pp. 1451–1462, October 1975. According to one embodiment of the present invention, the Faster-Than-Nyquist (FTN) technique is implemented with a shaping filter, for example, of the root-raised cosine (RRC) type. In another embodiment of the invention, the use of the FTN technique is coupled with any other frequency-bounded (B-band) shaping filter.

[0083] The compression block 15 is adapted to modify the period separating the spread symbols according to a fixed compression ratio (0 < ^ < 1). At the output of the compression block 15, the spread symbols therefore each consist of SF pulses (chips) separated this time by a period ^Tc.

[0084] According to one embodiment of the present invention, the filter block 16 is considered as a Nyquist root filter with orthogonality time Tc: it applies, to the received input signal, a filter matched, yf), to the filter y^t) such that (f) = ye( Σ) *i.e., a Nyquist filter for the period Tc, that is, such that rVe(t) is zero every Σ = kTc kg Z* ■ 'c filter of setting in form is in RRC, ve(t) is associated with a roll-off parameter between 0 and 1, a compromise between spectral efficiency and the temporal support of the filter (the lower the roll-off, the smaller the band, the better the spectral efficiency but the more the temporal support of the filter increases, which complicates reception).

[0085] For further details on the FTN link, reference may be made to the publication J. Fan, S. Guo, X. Zhou, Y. Ren, GY Li and X. Chen, "Faster-Than-Nyquist Signaling: An OverView," in IEEE Access, vol. 5, pp. 1925-1940, 2017, doi: 10.1109 / ACCESS.2017.2657599. This filter limits the signal band to B (where B is the band of the shaping filter) and is defined so that, when = 1, transmitter 10 corresponds to a conventional DSSS transmitter.

[0086] Faster-Than-Nyquist (FTN) transmission consists of transmitting information faster than the limit imposed by the orthogonality time (Tc) of the shaping filter 16. Exceeding this limit has the advantage of increasing the data rate while maintaining the same bandwidth B (there is an increase in spectral efficiency), but has the disadvantage of introducing intrinsic symbol interference (IESI). In the specific case of an FTN-DSSS link according to the invention, the transmission time of the chips is compressed. This means that we will introduce

[0087]

[0088] intrinsic inter-chip interference associated with both the current symbol, but also potentially with the preceding and following symbols. Even if the shaping filter does not correspond to a Nyquist root filter (the autocorrelation of the Nyquist root shaping filter pv(t) = Ve(t) * Ve(-t) must vanish every kTc, kZ, (k^0) to satisfy the Nyquist criterion), it is possible to define λ = 1 / B from the finite useful bandwidth B of the filter (for example, with an upper limit at the -3dB point, or the cutoff frequency...). By setting y = , the results of the invention as described below can be extended to any shaping filter with a useful bandwidth B; cf. L. Mounsif and D. Roque, "Optimal Pilot Sequences for Timing Estimation in Faster-Than-Nyquist Systems," in IEEE Communications Letters, vol. 25, no. 4, pp. 1236-1240, April 2021, doi: 10.1109 / LCOMM.2020.304551. Figure 2 illustrates the steps of a method for emitting a wireless communication signal in one embodiment of the invention.

[0089] Blocks 11 to 16 of the transmitter 10 are adapted to implement those steps, described below with reference to [Fig. 2], which are their responsibility. In one embodiment, these blocks (or at least one of them) comprise a microprocessor and a memory storing software instructions, which, when executed on the block's microprocessor, implement the steps of the process 100 that are the responsibility of said block.

[0090] In a step 101, the coding block 11 receives the digital signal sn to be transmitted and first implements on this signal sn an error detection or error correction coding of any efficiency R (turbocodes, convolutional code ...), in order to protect the signal from noise during its transmission.

[0091] In a step 102, the interleaver 12 II interleaves the received bits and provides resulting identically distributed useful bits.

[0092] In a step 103, the modulation block 13 delivers symbols an according to the constellation states corresponding to bit sequences supplied by the interleaver 12. The symbols an delivered by the modulation block 13 each have the same duration equal to T.

[0093] In a subsequent step 104, each symbol an is spread by the spreading block 14: the spreading block 14 modulates each of the symbols supplied to it as input by a set of SF chips called the spreading code. This spreading code can be identical for each symbol (short codes) or different from one symbol to another (long codes). The length of a short code is therefore the symbol duration, while that of a long code can be several centuries. Spreading consists of multiplying the symbols (separated by a period T) that one wishes to spread by Spreading codes (pseudo-random sequences) consisting of SF chips separated by a smaller period 71 = -L (where SF is the spreading factor). The length The LDSss and TDSSs (LDSSs / Tc) period of the code vary depending on the application: the value of TDSss can range from the duration of a single symbol (TDSss = 1 / 2: short spreading codes) to several centuries (long codes). The larger the TDSSs and the more arbitrary the DSSS code, the more difficult it is to recover the spreading code and despread the signal, and the more discrete the transmission. The spreading code can be real or complex (of any modulation) and can be identical or different for the real and imaginary parts of the symbols.

[0094] The spreading factor SF (for Spreading Factor) is the ratio between the transmission frequency of the chips (1 / Tc) and of the symbols (1 / T).

[0095] The spreading code L, 1 associated with the symbol an is a long code: 1 nSF+qjg=0 £ SF4 The spreading code changes every symbol, i.e. it is different for each symbol an and is therefore also a function of n. These SF chips spreading a symbol are specific to the constitution of said spreading code, its initial synchronization and the instant of emission of the symbol.

[0096] For example SF = 3 and the same spreading code (long: different all times symbol) is used on the real and imaginary part of the symbol.

[0097] With 6 being the Dirac delta function, the DSSS symbol that we wish to transmit (at time nT = nSFTc) is:

[0098] an(t) = anô(t-nT)

[0099] The associated spreading code is:

[0100] Uu(t) = UhSp+q6(t) + WbSF+1Ô(t-Tc) + UnSp+2Ô(t-2Tc)

[0101] The spread symbol is therefore:

[0102]

[0103] cD(t) = an(t^un(t) = aniinSf+QÔ( t-nT) + anunSp+]6[ t-Tc-nT) + anunSp+2Ô( t- "2Tc- nT) = anunSF+oô( t-(nSF + 0)Tc) + anunSF+1ô( t-(.nSF + l')Tc') + anunSF+2ô(t-(nSF+2)Tc)

[0104] Depending on the spreading code used, different needs can be met:

[0105] If a link with high energy discretion is desired, it is necessary to: - Use an independent and identically distributed (iid) long code on each symbol with a large spreading factor (SF); or - Use a short code with the addition of a scrambler for each symbol (transforms the short code into a long code) with a large SF spreading factor.

[0106] If a link with improved spectral efficiency and performance is desired, it is necessary to: - choose a spreading code or subset of codes that minimizes interference introduced by the transmitting FTN; and / or - Choose the constellation used by the spreading code to limit intrinsic reception errors (pathological cases) and those introduced by the FTN

[0107] The use of a short code, phase modulation, or a sufficiently large spreading factor makes it possible to reduce the complexity of reception.

[0108] In another embodiment, the long spreading code is obtained by scrambling a short code, thus making it specific to the symbol to which it will be multiplied.

[0109] In a step 105, the compression block 15 then compresses, according to the compression ratio, the period Tc separating the chips of the spread symbols (0 < < 1). At the output of the compression block 15, the spread symbols thus consist of SF pulses linked to the spreading code separated by a period ^Tc. In one embodiment, is chosen to be strictly less than 1.

[0110] Thus, using the example above, the FTN-DSSS signal at the output of compression block 15 is: [YES]

[0112] cn(t) =^anunSF+o6(t4nSF + O)^J^ (nSF+l)^c)+^anu^F+26(t-(nSF + 2)Çrc)

[0113] Note: the factor allows for obtaining an average emission power constant regardless of the value of Ç.

[0114] In a step 106, these spread symbols consisting of SF pulses separated by a period ^Tc are then filtered by the shaping filter, which can be a Nyquist root filter with orthogonal time Tc or any filter with a finite useful bandwidth B=UTC

[0115] In a step 107, the resulting constant power FTN-DSSS transmission signal Se(t) is then emitted via the RF antenna 17.

[0116] In general:

[0117] 101181 Se(t) = t- (nSF+q)ÇT c )

[0119] with: - Ç: the compression ratio, between 0 and 1; - r 1: the chips from the spread symbols; '■SiSF+gj - Tc: the time separating the chips;

[0120]

[0121]

[0122]

[0123]

[0124]

[0125]

[0126]

[0127]

[0128]

[0129]

[0130]

[0131]

[0132]

[0133]

[0134]

[0135]

[0136]

[0137]

[0138]

[0139] - SF; the spreading factor; - Ve(t): the impulse response of the shaping filter. We can consider a spreading code (i.e., a bi-code) with arbitrary and different modulation on the real and imaginary parts of the symbol. In this case, the expression for cnSF+q is given by: ^nSF+q — Re( an ) ^IqiSF+q + j ( an ) a2qiSF+q with : to: the transmitted symbol ai,nSF+q ; the spreading code over the real part, Re( of the symbol U2,nSF+q ; the spreading code over the imaginary part, Im(an) ,of the symbol In such a case, of course, there are symbols with a non-zero real part and symbols with a non-zero imaginary part. Equivalently, the above equation can be rewritten as: cnSF+q=an vUiSF+q+an v2jiSF+q with : T7 _ H,, । „ 1 : the spreading code associated with the symbol vIjiSF+q ~ 2\Ul,nSF+q + ^qiSF+q) J year T7 _ JT V the spreading code associated with the symbol v2,nSF+q ~ 2\UUiSF+q' U^nSF+q) J conjugated a*r If we consider the special case where the spreading code on the real part is identical to the code on the imaginary part, then we have CnSF+q = an^nSF+q and the equation giving Se( t ) becomes: se( t) = unSF+q ve( t- (nSF + q)ÇTc) with : - : the compression ratio - an: the symbols obtained after modulation - sF+q} ' 'cc° of spreading associated with the symbol a) - Tc: the uncompressed chip period - SF; the spreading factor - Ve(t): The formatting filter. Specifically, continuing with the example outlined above:

[0140] se(t) = cn(t)*ve(t) = auuI1SFWve(t- (nSF+0)^Tc) +^anunSF+1ve(t- (nSF+l)^Tc)+^ anunSF+2ve(t- (nSF + 2)^Tc)

[0141] From the equation giving Se(t), we note that the proposed FTN-DSSS signal does indeed depend on eight main parameters in transmission: - : the compression ratio - SF; the spreading factor - B; the occupied band; the shaping filter

[0142] the constellation symbol - the error-correcting code used - the interlacer used - {unsF+q}: The 2 spreading codes

[0143] In one embodiment of the invention, with reference to [Fig.3], the wireless telecommunications receiver 20 comprises an RF receiving antenna 21, a filtering block 24 for receiving shaping, a channel synchronization and estimation module 50 and a demodulation and decoding module 51.

[0144] The channel synchronization and estimation module 50 includes a frame synchronization block 40, a fine synchronization block 41, a channel estimation block 42. The demodulation and decoding module 51 includes a filtering block adapted to the estimated channel 25, a sampler block 26, a despreading block 27, an equalization block 2 (optional), a decision block 29, a demodulation block 30, a deinterlacing block 31, a decoding block 32.

[0145] These receiver blocks 20 are adapted to implement the steps, described below with reference to [Fig.4], which are their responsibility.

[0146] In one embodiment, the receiver blocks 20 (or at least one of them) include a microprocessor and a memory storing software instructions, which, when executed on the microprocessor of the block, implement the steps of the process 100 which are the responsibility of said block.

[0147] With reference to [Fig.4], the steps of a method for receiving a wireless communication signal implemented in an embodiment using receiver 20 are described.

[0148] A difference between an FTN-DSSS reception according to the invention and a (Nyquist-)DSSS reception comes from the sampling which takes place every TCTC, both in the synchronization and channel estimation part and in the demodulation and decoding part, as described below.

[0149] In a step 201, the filter y^- adapted to the shaping filter is applied by the filtering block 24 to the signal supplied via the antenna 21. More precisely, y*( -t) corresponds to the conjugate of time-reversed.

[0150] In a step 202, the frame synchronization block 40 performs a frame synchronization of the signal delivered by the shaping filter 24 (the synchronization can, for example, be done via a dedicated channel or by using a preamble before data transmission or directly on a specific pilot spreading code): the frequency offset and the delays are thus estimated by the frame synchronization block 40 and delivered by the frame synchronization block 40 to the fine synchronization block 41 and the channel estimation block 42. The signal filtered with the filter adapted to the shaping filter is, for example, frequency-shifted in parallel for different frequency offsets. Subsequently, the received signal is sampled on each parallel branch at times (with a the oversampling factor).Thus, on each branch, we obtain a succession of samples which we then correlate with the preamble (known spread symbols present in a frame). Next, we select the signal with the largest envelope before comparing the correlation of the sectioned signal to a threshold. If the latter exceeds the set threshold, a path is detected. We then retrieve the delay and the estimated frequency shift that caused the threshold to be exceeded.

[0151] In a step 203, the channel estimation block 42 estimates the transmission channel parameters representing the artifacts brought by the channel (for example from pilot sequences appearing in the signal frame).

[0152] In a step 204, the fine synchronization block 41, based on the frequency shifts / Çf and delays received, performs the fine synchronization allowing to improve the estimation of the delays J.

[0153] These two steps 203 and 204 are for example carried out simultaneously as is classically the case in DSSS receiving processes.

[0154] Then in a step 205, the filtering block adapted to the estimated channel 25 applies to the signal delivered by the filtering block 24 the filter adapted to the propagation channel determined, where applicable, based on the estimates of the transmission channel parameters made in step 203.

[0155] In a subsequent step 206, the sampler block 26 applies a sampling period c on the coefficients delivered by the filtering block adapted to the estimated channel 25. Indeed, since at transmission Tc was compressed by a factor to recover what was transmitted at reception, it is necessary to sample all the c's (the compression rate £ at transmission and reception being identical).

[0156]

[0157]

[0158]

[0159]

[0160]

[0161]

[0162]

[0163]

[0164]

[0165]

[0166]

[0167]

[0168]

[0169]

[0170]

[0171]

[0172]

[0173]

[0174]

[0175] Next, in step 207, the signal samples as delivered by the sampler block 26 are despreaded by the despreading block 27, which thus determines symbol sequences of duration T = SFTC by cross-correlation of the received signal with the respective sequences of SF chips separated by a period TCc, a function of the spreading code and the symbol. In an example where SF = 3, the output of step 205 is the signal z(t), and then every c is sampled by 26. The resulting sampled signal is: zn ( t ) = znSP^ô( t- ( nSF + 0 ) F, Tc ) + znSP+1B( t - ( nSF +1 ) ^Tc ) + zbSP+26( t - ( nSF + 2 ) ÇTC ) The despreading step from the samples ■[ Zn§p+g, Zngp+p Zngp+2} is done by multiplying the latter by the conjugate spreading codes (used in emission and associated with the symbol n) and performing their summation (for a single code): yn — znSF+0unSF+0' znSF+lunSF+l + znSF+2unSF+2 Unlike a (Nyquist-)DSSS bond, the znSF+q sample no longer depends solely on the current chip (anunSF+q) but also on the preceding and following chips and symbols: In the case of a classic single-code DSSS link, we would have in the "perfect" case (perfect channel h(t) = 0(t) + No noise) __1 ZnSF+q ~ SF ^n^nSF+q Whereas in the FTN-DSSS case we have: u znSF+q — "sF ^nSF+q + InSF+q + ^nSF+q the interference of the spreading code associated with the current symbol and ij1Sp+q the interference of the spreading codes associated with the following and preceding symbols. In the case of a SISO link with a flat channel (h(t) = ) and assuming perfect synchronization and the same code on the real and imaginary parts, we obtain the following signal after despreading: Yn — znSF+q^nSF+q — I -N \ / ^(¾¾ + în + bD with : : the contribution of the spreading code associated with the current symbol

[0176] in: the interference associated with the preceding and following symbols (if the SF is greater than or equal to the inverse of the compression ratio - 1).

[0177] bQ: the received noise.

[0178] : the channel amplitude

[0179] We therefore obtain symbols with more or less interference. The optimal reception in the sense of the MLSE (Maximum Likelihood Sequence Estimation) criterion is given by using on the received symbols, a decision box implementing the Viterbi algorithm with a non-classical branch metric.

[0180] From the optimal scheme, suboptimal receiver configurations can be deduced by optionally adding an equalizer (Minimum Mean Square Error (MMSE), Expected Propagation (EP) ...), performed by the equalization block 28.

[0181] Then, in a step 208, the decision block 29 decides, based on the symbol sequences from step 207, which symbols of the constellation have been received (Viterbi or symbol-by-symbol decision).

[0182] Then, during a step 209, the reverse operations of those at transmission are carried out, including: symbol demodulation, deinterlacing, and finally channel decoding:

[0183] - thus, the demodulation of the symbol is carried out by the demodulation block 30, which thus provides for each symbol the bits corresponding to the symbol according to the type of modulation chosen;

[0184] - the deinterlacing block 31 performs the inverse interlacing processing carried out in step 102.

[0185] - and finally, channel decoding is performed by decoding block 32, to identify potential errors, even correct them and finally deliver an estimate of the transmitted bits.

[0186] The receiver 20 described above with reference to [Fig.2] implements switching demodulation, but of course the application of the invention is not limited to this type of receiver.

[0187] The invention thus proposes the combined use of DSSS and FTN techniques (with RRC or other shaping filters) to ensure a certain degree of energy discretion and robustness to interference for sufficient SF, while limiting the loss in spectral efficiency compared to a conventional DSSS link, and while seeking to achieve good reception performance at reasonable complexity. The combination of the two DSSS and FTN techniques makes it possible to take advantage of their complementary benefits and to reduce their drawbacks.

[0188] Indeed, in terms of spectral efficiency:

[0189] - the FTN increases spectral efficiency (especially as the compression ratio is little),

[0190] - while the DSSS reduces it (all the more so as the spreading factor is large);

[0191] and at the level of inter-symbol interference:

[0192] - the FTN adds intrinsic IES (especially as the compression rate is little),

[0193] - while the DSSS reduces the IES (all the more so as the spreading factor SF is big).

[0194] Thus, with this combination of the spreading factor and the compression ratio, compression limits the spectral efficiency loss induced by spreading and the reduced intrinsic 1TES spread generated by compression, potentially allowing for very good performance from receivers with reduced complexity. The behavior and performance of the FTN-DSSS coupling depend on the combination of values ​​assigned to eight parameters, defining at least partially the configuration of the communication system:

[0195] - the spreading factor;

[0196] - the compression ratio;

[0197] - gang occupation;

[0198] - the link balance in reception.

[0199] - the formatting filter

[0200] - the symbolic constellation

[0201] - the error-correcting code used

[0202] - the interlacer used.

[0203] This combination makes it possible to generate a configurable FTN-DSSS waveform, which one, depending on the configurations: - greatly improves spectral efficiency; - induces energy discretion - limits the complexity of the receptor; - or even in some cases, improves performance (at constant spectral efficiency).

[0204] Depending on the eight parameters defined above, it is possible, for example, to: - implement a CDMA network implementing the invention, which results in increased system spectral efficiency compared to a conventional CDMA network; or - to design an energy-efficient discrete IoT system

[0205] The optimal combinations of values ​​for these parameters are determined, for example, by system simulation; they will depend on the application in which the invention is implemented and the signal band. Generally, the form The waveform according to the invention allows for an increase in the spectral efficiency of the transmission. The choice of values ​​for the various parameters indicated and the spreading code used allows either for improved performance at constant spectral efficiency or for improved energy discretion.

[0206] Wireless telecommunications according to the invention can be implemented more generally in all applications, civil or military, where the notion of energy discretion is desired, for example civil applications relating to loT or 5G-6G networks, but also applications relating to CDMA networks with increased spectral efficiency.

[0207] The invention can be implemented in the context of:

[0208] simplex links (in one direction only) with the FOP (Parametric Waveform) associated with specific parameters and different synchronization / delay estimation processes (MMSE, etc.), particularly in radionavigation applications;

[0209] simplex links with the FOP associated with specific settings and different reception, demodulation decoding processing (RAKE, EP, MLSE, etc.), particularly in broadcasting applications;

[0210] duplex links (transmits and receives) with the FOP associated with specific parameters and different demodulation decoding processes (RAKE, EP, MLSE, etc.) particularly in single-user radio communication link applications;

[0211] multiple duplex links (several users transmit and receive) with multiple FOPs associated with specific settings and demodulation decoding processing (RAKE, EP, MLSE, Join Detection, etc.) particularly in multi-user Radiocommunication Network applications - CDMA networks.

[0212] The invention can be implemented in a SIMO receiver according to which one replaces the channel h(f) and its matched filter with the vectors h(t) = and h* ( -t ) with N the number of receiving antennas.

Claims

Demands

1. A method for transmitting wireless communication, said method being implemented in a wireless telecommunication transmitting device (10) receiving as input a signal to be transmitted, said method comprising the following steps implemented by the transmitting device: application of a channel coding, including error-correcting coding and / or interleaving, to the signal to be transmitted, followed by digital modulation of said coded signal to obtain symbols; said method being characterized in that it further comprises the following steps implemented by the transmitting device: DSSS-type spreading of each symbol obtained by a TDSSs period-spreading code applied with a spreading factor SF, the SF chips spreading a symbol being specific to the instant of the symbol's transmission; and obtaining, from each symbol, a set of spread SF chips as a function of the symbol's value, these chips being separated by a period Tc;implementation on the chips obtained after spreading a treatment comprising: compression of the period between the chips by a compression ratio Ç strictly less than 1, the period between the chips after said compression being equal to ^Tc; filtering by a shaping filter v^t): in Nyquist root of period of orthogonality Tc; or of useful band of finite length B= 1 / T c;

2. A method for transmitting wireless communication, according to claim 1, wherein the signal Se(t) after filtering is defined by the following equation: se(t) (nSF+q)ÇTc) with : - £: the compression ratio, between 0 and 1; - r 1: the chips from the spread symbols; tCnSF+çJ • in the case of a single-code spread: CnSF+q — ân^nSF+q • in the case of a two-code spread: CnSF+q= -^e( ^l^SF+q + 3n) U2,nSF+q - Tc: the time separating the uncompressed chips; - SF: the spreading factor; - Ve(t): the impulse response of the shaping filter. - a« : the symbol to be transmitted at the output of the channel coding and digital modulation, with Re( an) the real part of the symbol and Im( aü) the imaginary part of the symbol; - louj„ letj„ 1: the monocode l^nSF+qJ t^hnSF+qJ l^iSF+qJ or bicode of spreading used.

3. A method for receiving wireless communication implemented in a wireless telecommunications receiving device (20) comprising at least one receiving antenna receiving a transmitted signal as input, said method comprising the following steps implemented by the receiving device on a signal obtained via the receiving antenna: - application of digital demodulation and channel decoding on the signal; said process being characterized in that it further comprises the following steps implemented by the receiving device: implementation of a signal processing comprising filtering by a filter y* ( -t ) and at the channel - fj, such that the filter y* ( -f ) is adapted to a shaping filter v^t) in Nyquist root of orthogonality period Tc or of useful bandwidth of finite length B — 1 / TC; period sampling ^Tc of the signal from the filter adapted to the global channel g*( ) = y*( -t) t) ' °where is strictly less than 1; despatting of the sampled signal as a function of the spreading code taken over a length equal to the spreading factor SF, the series of SF chips corresponding to the spreading code on a symbol being specific to the instant of reception of said symbol, the SF chips being separated by a period Te; and obtaining symbols of duration ÇT=SFÇTC.

4. Method of receiving a wireless communication, according to the preceding claim, in a SIMO receiver wherein the channel and its matched filter are replaced by the vectors 'h^t)' and jj* ( ) with N the number of antennas in h(t) = reception.

5. Method for receiving a wireless communication according to claim 3 or 4, comprising, after despreading, at least one of the following steps: - an equalization of the despreaded symbols implemented by an equalizer block (28) of the receiving device; - a selection of an element from a constellation of elements, according to the despreaded symbols, by the implementation of a Viterbi algorithm by a decision block of said receiving device.

6. Computer program, intended to be stored in the memory of a wireless telecommunications transmitting device (10) further comprising a microcomputer, said computer program comprising instructions which, when executed on the microcomputer, implement the steps of a method according to any one of claims 1-2.

7. Computer program, intended to be stored in the memory of a wireless telecommunications receiving device (20) further comprising a microcomputer, said computer program comprising instructions which, when executed on the microcomputer, implement the steps of a method according to any one of claims 3-5.

8. A wireless communication transmitting device (10), adapted to receive as input a signal to be transmitted, the transmitting device being adapted to apply channel coding, comprising a error-correcting coding and / or interleaving, on the signal to be transmitted, then digital modulation of said coded signal to obtain symbols; said transmission device characterized in that it is adapted to perform DSSS-type spreading of each symbol obtained by a TDSSs period spreading code applied with a spreading factor SF, the SF chips spreading a symbol being specific to the instant of emission of the symbol, and to obtain, from each symbol, a set of spread SF chips as a function of the value of the symbol, these chips being separated by a period TC; said transmission device being adapted to implement on the chips obtained after spreading, a treatment comprising: - compression of the period between the chips by a compression ratio strictly less than 1, the period between the chips after said compression being equal to ^Tc;- filtering by a shaping filter v^t): - in Nyquist root with period of orthogonality Tc; or - of useful band of finite length B= 1 / Tc-;

9. Wireless communication transmission device (10) according to claim 8, wherein the signal Se(t) obtained after filtering is defined by the following equation: Se(t) = fe^'oCnSF+qVef t- (nSF + q)ÇTC) with : - ? : the compression ratio, between 0 and 1; - r 1: the chips from the spread symbols; '-SjSF+çJ • in the case of a single-code spread: CnSF+q — ^n^nSF+q • in the case of a two-code spread: CnSF+q= -^e( ^l^SF+q + 3n) U2,nSF+q - Tc: the time separating the uncompressed chips; - SF; the spreading factor; - Ve(t): the impulse response of the shaping filter. - a« : the symbol to be transmitted at the output of the channel coding and digital modulation, with Re( an) the real part of the symbol and Im( aü) the imaginary part of the symbol; - louj„ letj„ 1: the monocode l^nSF+qJ t^hnSF+qJ l^iSF+qJ or bicode of spreading used.

10. A wireless communication receiving device (20) adapted to receive a transmitted signal via a receiving antenna, said receiving device being adapted to apply digital demodulation and channel decoding to the signal obtained via the receiving antenna; said device being characterized in that it is adapted to implement signal processing comprising filtering by a filter y*(-t) and to the channel such that the filter v*(-t) is adapted to a Nyquist root shaping filter v^t) with orthogonality period Tc or a finite-length useful bandwidth B = 1 / Tc; said device is adapted to perform sampling with period ^Tc of the signal from the filter adapted to the overall channel — v*( -t) *h\-1)' °ù ? is strictly less than 1 ; said device is adapted to perform a despreading of the sampled signal as a function of the spreading code taken over a length equal to the spreading factor SF, the series of SF chips corresponding to the spreading code on a symbol being specific to the instant of reception of said symbol, the SF chips being separated by a period ^Tc ; and to obtain symbols of duration ÇT=SFÇTC.

Images