CA2165034A1 - Transmission process having spectrum spread phase differential modulation and demodulation using orthogonal pseudorandom sequences - Google Patents
Transmission process having spectrum spread phase differential modulation and demodulation using orthogonal pseudorandom sequencesInfo
- Publication number
- CA2165034A1 CA2165034A1 CA002165034A CA2165034A CA2165034A1 CA 2165034 A1 CA2165034 A1 CA 2165034A1 CA 002165034 A CA002165034 A CA 002165034A CA 2165034 A CA2165034 A CA 2165034A CA 2165034 A1 CA2165034 A1 CA 2165034A1
- Authority
- CA
- Canada
- Prior art keywords
- sequence
- signal
- sequences
- symbols
- pseudorandom
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04J—MULTIPLEX COMMUNICATION
- H04J13/00—Code division multiplex systems
- H04J13/16—Code allocation
- H04J13/18—Allocation of orthogonal codes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
- H04B1/69—Spread spectrum techniques
- H04B1/707—Spread spectrum techniques using direct sequence modulation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04J—MULTIPLEX COMMUNICATION
- H04J13/00—Code division multiplex systems
- H04J13/0007—Code type
- H04J13/0022—PN, e.g. Kronecker
Abstract
Transmission process having spectrum spread phase differential modulation and demodulation using orthogonal pseudorandom sequences.
Use is made of two orthogonal pseudorandom sequences (CE1, CE2) for spread-ing the spectrum of the symbols (dk) to be transmitted and detection takes place of the phase difference between the received signals filtered by filters matched to the two sequences, one being delayed.
Use is made of two orthogonal pseudorandom sequences (CE1, CE2) for spread-ing the spectrum of the symbols (dk) to be transmitted and detection takes place of the phase difference between the received signals filtered by filters matched to the two sequences, one being delayed.
Description
216~03~
-TRANSMISSION PROCESS HAVING SPECTRUM SPREAD PHASE DIFFERENTIAL MODULATION
AND DEMODULATION USING ORTHOGONAL PSEUDORANDOM SEQUENCES
DESCRIPTION
TECHNICAL FIELD
The present invention relates to a transmission process having spectrum spread phase differential modulation and demodulation using orthogonal pseudorandom sequences.
The direct sequence spectrum spread modulation method has already been used for many years, more particularly in radiocommunications with satellites and in the military field. It is also used in radiolocation, such as in the so-called global positioning system or GPS.
This method has numerous advantages:
a) Discretion: This discretion is linked with the spread of the informa-tion transmitted over a broad frequency band. This leads to a low spectral density of the transmitted power. This feature makes this method partic-ularly attractive for military applications, which require maximum dis-cretion transmission systems. It is also of interest for civil applications, because it makes it possible to allocate the same frequency band to services using conventional spectrum spread modulation methods.
b) The resistance to scramblers and interference: A spectrum spread receiver effects the correlation operation between the message received and the spread pseudorandom sequence before any demodulation operation. This correlation operation makes it possible to extract the message to be received and rejects any scrambler or interference. The longer the pseudo-random sequence used for spreading the spectrum the greater the resistance to scramblers. This feature is vital in military applications. It is also useful in civil applications, because it is necessary to avoid, apart from interference due to users operating in a narrow band, the broad band inter-ference produced by other spectrum spread transmissions.
c) The excellent behaviour in fading channels: Provided that use is made B 12073.3 RS
216503~
of an asynchronous receiver structure, this method makes it possible to transmit digital data in reliable manner between mobiles and in the presence of obstacles generating multiple propagation paths. This is the consequence of the diversity effect inherent in the system making it possible to demodulate independently of one another all the informations carried by all the propagation paths. This feature makes it possible to envisage the use of this method for all transmissions between vehicles in an urban environ-ment, or for radio connections within buildings, where fading effects can be produced by the presence of personnel constituting mobile obstacles for radio waves.
d) The possibility of using a multiple access protocol with distribution by codes: This method consists of allocating orthogonal spread pseudorandom sequences (i.e. having low intercorrelation coefficients) to the different users. However, it remains difficult to implement, because it imposes an effective control of the transmission power.
The direct sequence spectrum spread modulation method has been described to a significant extent in the specialized literature. Reference can be made to two works, namely "Spread spectrum communications" by Marvin K. Simon et al, published by Computer Science Press, USA and the second entitled "Spread spectrum systems" by ~obert C. Dixon and published by John Wiley and Sons, USA.
PRIOR ART
In a conventional data transmitter using a conventional modulation method, modulation takes place of a radio frequency carrier by phase, frequency or amplitude modulation, or by a mixed modulation. For simplification purposes reference will only be made to phase modulations, which are nowadays the most frequency used. The digital data to be transmitted are binary elements or bits, which have a period Tb, i.e. a new bit must be transmitted for every Tb. With these bits it is possible to constitute symbols having a period T . These are symbols which will modulate the radio frequency car-rier. The symbol flow rate is expressed in bauds (or symbols/second).
B 12073.3 RS
2165()~
This procedure can be illustrated by two examples:
a) Binary phase shift keying (BPSK), which consists of allocating a phasestate, e.g. O, to the O bits and another phase state, e.g. ~ to the 1 bits.
In this case, the symbol is the bit and T =Tb. The phase state of the radio frequency carrier is imposed for every bit.
b) Quaternary phase shift keying or QPSK consists of using symbols formed by two successive bits. Thus, these symbols can assume four states (00, 01, 10, 11). A phase state of the carrier is allocated to each of these states and in this case T =2Tb. Therefore the phase state of the radio frequency carrier is imposed every other bit.
It is possible to improve the transmitted radio si2nal spectrum and in particular reduce the power in parasitic side lobes by using more sophisti-cated phase modulations, for which the modulating signal is shaped (filtered) prior to modulation and reference is then made to MSK, GMSK, SRC4 modula-tions.
On the reception side a distinction can be made between two large demodula-tion families: coherent demodulation and non-coherent demodulation. The coherent demodulation methods consists of implementing in the receiver a subassembly, whose function is to estimate the average phase of the carrier, so as to reconstitute a phase reference, which is then mixed with the signal received in order to demodulate the data.
This method has the optimum performance characteristics in radio channels, where a Gaussian noise is added to the useful signal. This is e.g. the case with transmissions with satellites. However, in the presence of multiple paths, this method gives very poor results.
The non-coherent demodulation method is based on the observation according to which it is sufficient for the phase reference of the current symbol is the phase of the preceding symbol. In this case, the receiver does not estimate the phase of the symbols, but instead the phase difference between two successive symbols. Thus, there is differential phase shift keying B 12073.3 RS
21650~
-(DPSK) or differential quadrature phase shift keying (DQPSK).
Fig. 1 shows the block diagram of a DPSK transmitter, which has an input Ee, which receives the data bk to be transmitted, which have a period Tb and it comprises a differential coder 10 formed by a logic circuit 12 and a delay circuit 14, a local oscillator 16 supplying a carrier P and a modulator 18 connected to an output Se supplying the DPSK signal.
The logic circuit 12 receives the binary data bk and supplies other binary data called symbols and designed dk. The logic circuit 12 also receives the symbols delayed by a period Tb, i.e. dk 1- The logic operation per-formed in the circuit 12 is the exclusive-OR operation on the data bk and on the compliment of dk delayed by one serial position (i.e. dk 1)' which gives: dk=bk dk-1 The modulator 18 phase modulates the carrier P as a function of each symbol dk .
Fig. 2 shows the block diagram of a corresponding receiver of the differ-ential demodulator type. This receiver has an input Er and comprises a band pass filter 20, a delay circuit 22 for a duration Tb, a multiplier 24, an integrator 26 operating on a period Tb and a logic decision circuit 28.
The receiver has an output Sr, which restores the data bk.
The pass band of the input filter 20 is between the NYQUIST band equal to 1/Tb and the width of the main lobe of the DPSK signal, which is 2/Tb.
By designating x(t) the signal received applied to the input Er, the multi-plier 24 received a filtered signal xF(t) emanating directly from the filter 20 and a delayed filtered signal xFR emanating from the delay circuit 22. It supplies a product signal m(t), which is integrated on a period Tb in the integrator 26, which supplies a signal, whose polarity makes it possible to determine the value of the transmitted bit, which is decided by the logic 28.
The direct sequence spectrum spreading method consists of multiplying the B 12073.3 RS
-TRANSMISSION PROCESS HAVING SPECTRUM SPREAD PHASE DIFFERENTIAL MODULATION
AND DEMODULATION USING ORTHOGONAL PSEUDORANDOM SEQUENCES
DESCRIPTION
TECHNICAL FIELD
The present invention relates to a transmission process having spectrum spread phase differential modulation and demodulation using orthogonal pseudorandom sequences.
The direct sequence spectrum spread modulation method has already been used for many years, more particularly in radiocommunications with satellites and in the military field. It is also used in radiolocation, such as in the so-called global positioning system or GPS.
This method has numerous advantages:
a) Discretion: This discretion is linked with the spread of the informa-tion transmitted over a broad frequency band. This leads to a low spectral density of the transmitted power. This feature makes this method partic-ularly attractive for military applications, which require maximum dis-cretion transmission systems. It is also of interest for civil applications, because it makes it possible to allocate the same frequency band to services using conventional spectrum spread modulation methods.
b) The resistance to scramblers and interference: A spectrum spread receiver effects the correlation operation between the message received and the spread pseudorandom sequence before any demodulation operation. This correlation operation makes it possible to extract the message to be received and rejects any scrambler or interference. The longer the pseudo-random sequence used for spreading the spectrum the greater the resistance to scramblers. This feature is vital in military applications. It is also useful in civil applications, because it is necessary to avoid, apart from interference due to users operating in a narrow band, the broad band inter-ference produced by other spectrum spread transmissions.
c) The excellent behaviour in fading channels: Provided that use is made B 12073.3 RS
216503~
of an asynchronous receiver structure, this method makes it possible to transmit digital data in reliable manner between mobiles and in the presence of obstacles generating multiple propagation paths. This is the consequence of the diversity effect inherent in the system making it possible to demodulate independently of one another all the informations carried by all the propagation paths. This feature makes it possible to envisage the use of this method for all transmissions between vehicles in an urban environ-ment, or for radio connections within buildings, where fading effects can be produced by the presence of personnel constituting mobile obstacles for radio waves.
d) The possibility of using a multiple access protocol with distribution by codes: This method consists of allocating orthogonal spread pseudorandom sequences (i.e. having low intercorrelation coefficients) to the different users. However, it remains difficult to implement, because it imposes an effective control of the transmission power.
The direct sequence spectrum spread modulation method has been described to a significant extent in the specialized literature. Reference can be made to two works, namely "Spread spectrum communications" by Marvin K. Simon et al, published by Computer Science Press, USA and the second entitled "Spread spectrum systems" by ~obert C. Dixon and published by John Wiley and Sons, USA.
PRIOR ART
In a conventional data transmitter using a conventional modulation method, modulation takes place of a radio frequency carrier by phase, frequency or amplitude modulation, or by a mixed modulation. For simplification purposes reference will only be made to phase modulations, which are nowadays the most frequency used. The digital data to be transmitted are binary elements or bits, which have a period Tb, i.e. a new bit must be transmitted for every Tb. With these bits it is possible to constitute symbols having a period T . These are symbols which will modulate the radio frequency car-rier. The symbol flow rate is expressed in bauds (or symbols/second).
B 12073.3 RS
2165()~
This procedure can be illustrated by two examples:
a) Binary phase shift keying (BPSK), which consists of allocating a phasestate, e.g. O, to the O bits and another phase state, e.g. ~ to the 1 bits.
In this case, the symbol is the bit and T =Tb. The phase state of the radio frequency carrier is imposed for every bit.
b) Quaternary phase shift keying or QPSK consists of using symbols formed by two successive bits. Thus, these symbols can assume four states (00, 01, 10, 11). A phase state of the carrier is allocated to each of these states and in this case T =2Tb. Therefore the phase state of the radio frequency carrier is imposed every other bit.
It is possible to improve the transmitted radio si2nal spectrum and in particular reduce the power in parasitic side lobes by using more sophisti-cated phase modulations, for which the modulating signal is shaped (filtered) prior to modulation and reference is then made to MSK, GMSK, SRC4 modula-tions.
On the reception side a distinction can be made between two large demodula-tion families: coherent demodulation and non-coherent demodulation. The coherent demodulation methods consists of implementing in the receiver a subassembly, whose function is to estimate the average phase of the carrier, so as to reconstitute a phase reference, which is then mixed with the signal received in order to demodulate the data.
This method has the optimum performance characteristics in radio channels, where a Gaussian noise is added to the useful signal. This is e.g. the case with transmissions with satellites. However, in the presence of multiple paths, this method gives very poor results.
The non-coherent demodulation method is based on the observation according to which it is sufficient for the phase reference of the current symbol is the phase of the preceding symbol. In this case, the receiver does not estimate the phase of the symbols, but instead the phase difference between two successive symbols. Thus, there is differential phase shift keying B 12073.3 RS
21650~
-(DPSK) or differential quadrature phase shift keying (DQPSK).
Fig. 1 shows the block diagram of a DPSK transmitter, which has an input Ee, which receives the data bk to be transmitted, which have a period Tb and it comprises a differential coder 10 formed by a logic circuit 12 and a delay circuit 14, a local oscillator 16 supplying a carrier P and a modulator 18 connected to an output Se supplying the DPSK signal.
The logic circuit 12 receives the binary data bk and supplies other binary data called symbols and designed dk. The logic circuit 12 also receives the symbols delayed by a period Tb, i.e. dk 1- The logic operation per-formed in the circuit 12 is the exclusive-OR operation on the data bk and on the compliment of dk delayed by one serial position (i.e. dk 1)' which gives: dk=bk dk-1 The modulator 18 phase modulates the carrier P as a function of each symbol dk .
Fig. 2 shows the block diagram of a corresponding receiver of the differ-ential demodulator type. This receiver has an input Er and comprises a band pass filter 20, a delay circuit 22 for a duration Tb, a multiplier 24, an integrator 26 operating on a period Tb and a logic decision circuit 28.
The receiver has an output Sr, which restores the data bk.
The pass band of the input filter 20 is between the NYQUIST band equal to 1/Tb and the width of the main lobe of the DPSK signal, which is 2/Tb.
By designating x(t) the signal received applied to the input Er, the multi-plier 24 received a filtered signal xF(t) emanating directly from the filter 20 and a delayed filtered signal xFR emanating from the delay circuit 22. It supplies a product signal m(t), which is integrated on a period Tb in the integrator 26, which supplies a signal, whose polarity makes it possible to determine the value of the transmitted bit, which is decided by the logic 28.
The direct sequence spectrum spreading method consists of multiplying the B 12073.3 RS
2 1 6 5 0 3 ~
-sequence of symbols dk obtained after the differential coding by a pseudo-random sequence having a much higher flow rate than that of the data to be transmitted. This pseudorandom sequence has a binary rate ~ times higher than that of the binary data to be transmitted. The duration Tc of one bit of this pseudorandom sequence, said bit also being called a chip, is there-fore equal to Tb/N. The chip rate of the pseudorandom sequence can be several megachips, or even several dozen megachips per second.
Fig. 3 shows the block diagram of a direct sequence spectrum spread trans-mitter. The elements already shown in fig. 1 carry the same references.
In addition to those of fig. 1, the emitter comprises a pseudorandom (C) sequence generator 30 and a multiplier 32. The signal supplied by the multiplier has a pseudorandom sequence-spread spectrum. The modulator 18 then no longer operates on the symbol of origin dk, but on the corresponding spread symbol Dk.
The corresponding receiver has the same structure as that of fig. 2, except that the filter 20 is now a matched filter, whose pulse response reduced to the baseband is the time-reversed combined complex of the pseudorandom sequence used in the transmitter.
The pseudorandom sequence used on transmission must consequently have an autocorrelation function with a marked peak of value N for a zero delay and the smallest possible side lobes. This can be obtained by using maximum length sequences also known as m-sequences, or so-called GOLD or KASAMI
sequences.
The sequences usable in a spectrum spread system have been the subject matter of intense studies, which are summarized in the article entitled "Cross-correlation properties of pseudorandom and related sequences" by Dilip V. SARWATE and Michael B. PURSLEY, published in "Proceedings of the IEEE", vol. 68, No. 5, May 1980, pp 593 to 619.
The input filter used in a spectrum spread receiver has an equivalent base-band pulse response H(t) and this response must be the time-reversed com-bined complex of the pseudorandom sequence used on transmission:
B 12073.3 RS
21~5034 H(t) = c*(T - t) where C(t) is the pseudorandom sequence used on transmission.
Therefore the signal supplied by such a filter is:
xF(t) = x(t) * H(t) where the symbol * designates the convolution operation, so as to give:
~t XF(t) = ~/ x(s).C* (Ts-t+s)ds.
t-T
Thus, the matched filter performs the correlation between the signal applied to its input and the pseudorandom spread sequence.
In a Gaussian additive noise channel, the signal xF(t) will therefore be in the form of a pulse signal, whose repetition frequency is equal to 1/Tb.
The envelope of this signal is the autocorrelation function of the signal C(t). The information is carried by the phase difference between two succ-essive correlation peaks. Thus, the multiplier output will be formed by a succession of positive or negative peaks, as a function of the transmitted bit value.
In the case of a radiotransmission in the presence of multiple paths, the output of the matched filter will be formed by a succession of correlation peaks, each peak corresponding to a propagation path.
The different signals of the reception chain are shown in fig. 4. Theline (a) represents the filtered signal xF(t), the line (b) the correlation signal xF(t)*xF(Ts-t) and the line (c) the signal at the integrator output.
This known procedure is described in detail in the article entitled "Direct-Sequence Spread Spectrum with DPSK Modulation and Diversity for Indoor Wireless Communications" published by Mohsen KAVEHRAD and Bhaskar RAMAMURTHI
in "IEEE Transactions on Communications", vol. COM 35, No. 2, February 1987.
Thus, a description has been given of DPSK spectrum spread modulation. It is clear that this procedure can be applied in the same way to all differ-B 12073.3 RS
216~03~
-ential modulations. For example, in the case of DQPSK modulation, each correlation peak at the output of the matched filter could assume four phase states, whereas there are only two with a DPSK modulation.
Despite their great interest these procedures suffer from disadvantages.
Thus, the successive bits constituting the message to be transmitted are all multiplied by the same pseudorandom spread sequence, which can give rise to a coherent intercorrelation noise able to deteriorate the quality of the signal m(t) at the demodulator output.
The applicant has profoundly studied this question and the research carried out can be summarized as follows. Consideration must again be given to the transmitter shown in fig. 3 and the receiver of fig. 2, knowing that the latter operates with a matched filter 20 corresponding to a given pseudo-random sequence. In general terms, the sequence used on reception isdesignated CR and that used on transmission CE. When the transmitter and receiver are matched to one another, one obviously obtains CE-CR.
In baseband, the signal at the input of the matched filter 20 can be written:
x(t) = dCE(t) _ representing the sequence of bits modulating the carrier, said bits being obtained from binary data to be transmitted b following passage in the differential coder 10.
The filtered signal found at the output of the matched filter can then be written: t *
xFt() = ~ x(s)CR (s - t + Tb)ds t-Tb CR representing the combined complex of CR, which can then be written:
O *
x (t) = d 1 ~ CE(s + TbCR~s - t + Tb)ds + do f CE(s)CR(s - t + Tb)ds t-Tb o or once again x(t) = d 1CC C (-t) + d0CC C (Tb ~ t) where the notation Cc C represents the aperiodic correlation function between CE and CR.
B 12073.3 RS
21650~
In the case where CE-CR the same formula is obtained by replacing the CC C by Cc ~ whlch is the aperiodic autocorrelation function of CR.
The delayed filtered signal found at the output of the delay circuit 22 is simply shifted by one symbol and can consequently be written:
xFR(t) = d 2CC C (-t) + d lCC C (Tb ~ t).
Finally, at the output of the demodulator 24, the signal can be written:
m(t)=xF(t).xFR(t) or d d c2 (-t) + l d 1 + d-2dO CCE,CR CE,CR b +d ldOCC , CR ( b As the symbols d are obtained from data b by the relation dk = bk63 dk 1 and as d or _ can only assume the two values +1 or -1, we necessarily obtain:
d-2d-1 = b_l~ d_1-do = bo and d_1 + d 2.do = 1 + b 1.bo Thus, one once again obtains a new expression of the demodulation sig20 m(t):
m(t) = b lCc C (-t) + ll + bob 1~Cc C (-t)Cc C (Tb ~ t) bO CE, CR( b (t) Three cases can then occur as a function of the respective values of boand b 1-Case No. 1: bo = b-1 We then obtain: 2 ~ CE~ R E" R b ] E, R
in which 4C C is called the even correlation function of CE and CR.
Case No. 2: b 1 = bo = -1 -2 ^2 3s m(t) lCCE,CR( t) CE'CR( b ~ CE.CR b B 12073.3 RS
216503~
, g where C C is called the odd autocorrelation function of CE and CR.
Case No. 3: b 1 = -bo m(t) = bo ICC C (Tb ~ t) - Cc C (-t)] = bo~c C (Tb ~ t) ~C C (Tb ~ t) With these considerations developed, the deficiency of the conventional direct sequence spectrum spread method becomes very clear. Thus, at least in the first two cases, the demodulation signal m(t) is either always positive, or always negative. However, it is this signal which will be inte-grated into the circuit 26 prior to decision taking. Apart from the corre-lation peak obtained for t=O, this signal should ideally have a zero mean value. This is manifestly not the case and the integrator output will consequently have at the time of decision making, a significant negative or positive shift value as a result of the integration of a signal, whose mean value is not zero. These shifts will deteriorate the quality of the deci-sion taking and will increase the bit error rate and this will become worse as the integration time increases.
The object of the invention is to obviate this disadvantage and aims atimproving the quality of the output signal of the demodulator (signal m(t)) and therefore the transmission quality.
DESCRIPTION OF THE INVENTION
The invention therefore proposes a novel transmission process by spectrum spread phase differential modulation and demodulation in which use is made of two pseudorandom sequences on transmission, which are orthogonal to one another, said two sequences being used for spreading the spectrum of the signals, which will then modulate a carrier. On reception, two filtering operations matched to these two orthogonal sequences will take place. In order to obtain the demodulation signal, multiplication will take place of the delayed filtered signal associated with one of the sequences by the filtered, but not delayed signal associated with the other sequence. As these two signals result from filtering operations by filters, whose pulse responses are orthogonal in the NYQUIST sense, the interference affecting these signals will be decorrelated, which will decrease their influence on B 12073.3 RS
the demodulation signal. In the prior art, the two signals, respectively delayed and filtered and filtered emanated from the same filter, so that the interference affecting them was correlated.
More specifically, the present invention relates to a data transmissionprocess by spectrum spread phase differential modulation and demodulation, in which the data are organized in at least one binary data sequence having a certain period, characterized in that:
A) on transmission:
- a differential coding of the binary signals to be transmitted takes place in order to obtain symbols, use being made of two orthogonal pseudorandom sequences for spreading the spectrum of each symbol and a carrier is modu-lated by the spread spectrum symbols, B) on reception:
- two parallel filtering operations take place matched to the two orthogonal sequences and there is an estimation of the phase difference between the two signals obtained after filtering, one of which has been delayed, which restores the data.
The invention has several embodiments according to which transmission takes place of one, two or more than two data flows.
BRIEF DESCRIPTION OF THE DRAWINGS
Fig, 1, already described, shows the block diagram of a prior art, differ-ential coding transmitter.
Fig. 2, already described, shows the block diagram of a corresponding receiver.
Fig. 3, already described, shows the block diagram of a prior art, direct sequence spectrum spread transmitter.
Fig. 4, already described, shows the configuration of the different signals appearing in a direct sequence spectrum spread receiver.
B 12073.3 RS
Fig. 5 is a block diagram of a transmitter according to the invention in afirst variant with a single flow of data to be transmitted.
Fig. 6 is a block diagram of a corresponding receiver.
Fig. 7 symbolically shows the transmission principle of a single data flow.
Fig. 8 symbolically shows the principle of simultaneously transmitting two data flows.
Fig. 9 shows the structure of a transmitter corresponding to a double data transmission.
Fig. 10 shows the structure of the corresponding receiver.
Fig. 11 symbolically illustrates a transmission of three data flows with three pairwise orthogonal pseudorandom sequences.
Fig. 12 illustrates a transmission variant of two symbols with the half flow rates.
Fig. 13 illustrates an embodiment of a receiver having a surface acoustic wave component.
Fig. 5 shows a diagram of an embodiment of a transmitter according to the invention. This transmitter comprises the means already shown in fig. 3 and which carry the same references, namely an input Ee receiving the binary data bk to be transmitted, said data having a period Tb, a differential coder 10 with its logic circuit 12 and its delay circuit 14. This coder supplies the symbols dk. The transmitter shown also comprises a first generator 301 supplying a first pseudorandom sequence CEl applied to a first multiplier 321 and a second generator 32 supplying a second pseudorandom sequence CE2 applied to a second multiplier 322. The multipliers 321 and 322 supply symbols (Dk)1 and (Dk)2, whose spectrum is spread by the B 12073.3 RS
216503~
corresponding sequences. The transmitter also comprises a first modulator 181 receiving a carrier P produced by a local oscillator 16 and the symbols (Dk)1 spread by the first pseudorandom sequence CE1 and a second modulator 182 receiving said same carrier P and the symbols (Dk)2 spread by the second pseudorandom sequence CE2. Finally, an adder 31 is connected to the two modulators 181 and 182 and supplies the output Se of the transmitter. This output transmits the carrier modulated by the doubly spread signal. The modulation can be of a random nature (phase, frequency, mixed, etc.).
Naturally, it would be possible to modify this circuit by firstly adding the two spread signals (Dk)1 and (Dk)2 and by using a single modulator. It would also be possible to add the two pseudorandom sequences CE1 and CE2 as soon as they are formed and to spread the signal with the aid of the pseudo-random sum obtained.
The receiver shown in fig. 6 comprises the means already shown in fig. 2, namely an input Er receiving a signal x(t), a delay circuit 22 with a dura-tion equal to the period Tb of the transmitted bits, a multiplier 24 supply-ing a demodulation signal m(t), an integrating circuit 26 acting on the signal m(t) supplied by the multiplier 24, the integration taking place between 0 and Tb, a logic decision circuit 28 and a general output Sr supply-ing the original digital data bk. This receiver differs from the prior art in that it comprises a first filter 201 matched to the first pseudorandom sequence CE1 used in the transmitter and a second filter 202 matched to the second pseudorandom sequence CE2 used in the transmitter. The term filter matched to a pseudorandom sequence is understood to mean a filter, whose pulse response brought down to the baseband is the time-reversed combined complex of the pseudorandom sequence in question.
The first filter 201 supplies a filtered signal xF1, which is then delayed by Tb in a circuit 22 in order to give a delayed, filtered signal xFR1.
The second filter 202 supplies a filtered signal xF2. This signal is not delayed and is used for producing the demodulation signal m(t) by multipli-cation with the filtered, delayed signal xFR1.
This transmitter-receiver assembly functions as follows. As in the case of B 12073.3 RS
` 216503~
conventional modulation with direct sequence spectrum spreading, the binary quantities used for modulating the carrier are the symbols dk obtained from the original data bk following passage in the differential coder 10.
In the conventional process, the symbols dk are multiplied by a single pseudorandom sequence CE. In the process according to the invention, the symbols dk are multiplied by the two pseudorandom sequences CE1 and CE2, which are orthogonal to one another, in order to form a signal which will modulate the carrier.
The sequences CE1 and CE2 can have the same general characteristics as a sequence used in the prior art. They are both constituted by N elements or chips having a period T = b and their autocorrelation functions have a single marked peak for a zero delay. They have the lowest possible value elsewhere. These sequences are also chosen so as to have a low inter-correlation function, no matter what the time.
In the case of the receiver, the input signal x(t) can be written:
x(t)=d(CE1(t)+CE2(t)) The filtered, delayed signal xFR1 at the output of the delay circuit 22 can be written:
XFRI = d-2 ~CcEl~cEl( t) CCE2,CEl( t)]
+d 1 [CcEl CE1(Tb ~ t) + CcE2 CE1(Tb ~ t~
In the same way, the filtered signal xF2 at the output of the filter 202 can be written:
xF2 - d_1 [CcEl~cE2( t) CCE2,CE2( )]
+d lC (T - t) + C (T - t)~
At the output of the multiplier 24, the demodulation signal is of form:
m(t)=xFR1 xF2 Therefore four cases can be distinguished:
Case No. 1: bo =b 1=1 B 12073.3 RS
216~034 We then have d 2=d 1=do. It is then possible to rewrite xFR1 and xF2 in the following form:
xFR1 d-1 ~ CE1(Tb t) CE2,CE1(Tb t) and xF2 d-l [ CEl,CE2( b t) CE2(Tb t)]
where, after simplification and replacing OCEi CEj(TB - t) with iJ
m(t) ~ 12 ~ 22 ~21 ~12 ~21 ~22 For t=Tb, we clearly obtained m(0)=C11(0).022(0)+ .. i.e. N , which corres-ponds to the autocorrelation peak.
A detailed analysis of the signal m(t) makes it possible to see that in the case where the two sequences used for modulating the data are orthogonal, the signal m(t) has a zero mean value outside the correlation peak. Thus, in this case, the output of the integrator 26 does not have a shift as was the case with the conventional structure.
Case No. 2: bo = b 1 = -1 In this case, we obtain do=d 1=-d 2. Thus, it is possible to rewrite x and xF2 in the following form, with the same simplifying notations as hereinbefore:
r~ A
~ l [ ^ ]
A ~, A ,~ ~ A A A
so that m(t) 11 12 11 22 21 12 21 22 All the remarks made in connection with case No. 1 still apply.
Case No. 3: bo=1 b 1=-l In this case, we obtain do=d l=-d 2 and therefore _ F2 -1 _ 12 22 B 12073.3 RS
216503~
-.~ ~ ,'~ A
and m(t) ~11 12 21 12 11 22 21 22 The same remarks still apply.
Case ~o. 4: bo=-l b 1=l In this last case, we have do=-d l=-d 2 and xFRl = d 1 ¦~11 + ~21 10F2 -1 [ 12 22~
so that m(t) 11~12 ~ 22 ~21~12 ~21~22 The same remarks still apply.
The two signals received by the demodulator 24 are different both as respect to their serial position and by the pseudorandom sequence involved. The signal xFRl has been delayed by a period Tb and therefore corresponds, on transmission, to a symbol of serial position k-l, whereas the signal xF2 corresponds to the following symbol of serial position k. In addition, the signal xFR1 comes from a symbol which has been spread and then despread by the first sequence CEl, whereas the signal xF2 has been processed by the second sequence CE2. As a result of the delay of one serial position between the symbols in question, it can be seen that the sequences CEl and CE2 have not been transmitted at the same time. Each sequence is period-ically transmitted with the same period Tb as the symbols to be processed, but the sequence CEl linked wit the signal xFR1 has been transmitted one serial position prior to the sequence CE2 linked with the signal xF2. The two sequences CEl and CE2, although globally produced simultaneously by the generators 301 and 32' intervene in a manner shifted by one serial position as a result of the processing performed on reception of the sequences.
Thus, use is then made of a sequence CEl and then a sequence CE2.
This process is diagrammatically illustrated in fig. 7. The first sequence CEl is repeated periodically for each symbol dk 1~ dk~ dk+l, etc., as is thesecond sequence CE2. These sequences are symbolized by the rectangles of B 12073.3 RS
216503~
-respectively the upper and lower bands. On reception, the demodulation signal m(t) is formed from a signal xFR1, whose origin is a symbol of serial position k-l processed by CE1 and a signal xFl originating from a symbol of serial position k processed by CE2. The downward oblique arrow from a sequence CE1 of serial position k-1 to a sequence CE2 of serial position k symbolizes the chronology of the sequences used. This chronology is repea-ted on each formation of a new demodulation signal. All the downward oblique arrows symbolize the transmission of all the symbols dk, when k increases unitwise.
In the embodiment illustrated in figs. 5 and 6, this distinction between the serial position of the sequences involved has no importance because a sequence is identically repeated (the sequence CE1 of serial position k-1 being identical to the sequence CE1 of serial position k). However, the distinctions described hereinbefore give a better understanding of the improvement provided by the invention and which is as follows.
It is possible to conceive a receiver using a first matched filter on CE1 and which would not be followed by a delay circuit, the latter being associated with the second filter matched to CE2. Demodulation would then take place of the phase jumps caused by CE2 followed by CEl.
In other words, it is possible to exploit the presence of two pseudorandom sequences on transmission by simultaneously transmitting two data flows and by detecting the phase jumps in both directions. This is what is symbolic-ally shown in fig. 8, where at the top there is a succession of pseudo-random sequences CE1 and at the bottom a successive of pseudorandom sequen-ces CE2. Data b1 impose phase jumps between a sequence CE1 and the follow-ing sequence CE2. Data b2 impose phase jumps between a sequence CE2 and the following sequence CE1. The rising or falling oblique arrows symbolize the two processes.
Figs. 9 and 10 illustrate how it is necessary to modify the diagrams of the transmitter and receiver compared with figs. 5 and 6 in order to effect this double transmission. The transmitter of fig. 9 has two inputs Ee1, Ee2 receiving two binary data flows respectively b1 k and b2 k. It comprises B 12073.3 RS
216~34 .
two logic circuits 121, 122 and two delay circuits 141, 142, whose input is connected to the output of one of the logic circuits and the output is connected to the input of the other logic circuit. A first multiplier 321 processes the symbols dl k~ present at the output of the logic circuit 122 and a second multiplier 322 the symbols d2 k present at the output of the logic circuit 121. There is a local oscillator 16 supplying the carrier, two modulators 181, 182 and an adder 31 supplying at its output Se the doubly modulated carrier.
On the receiver side and as is illustrated in fig. 10, there are two symmet-rical channels crossed at the outset, a first channel with a filter 201 matched to the sequence CE1 and a delay circuit 221 and a second channel with a second filter 202 matched to the sequence CE2 and a delay circuit 222.
Thus, the first channel comprises a multiplier 241, an integrator 261 and a decision logic 281. The second channel comprises a multiplier 242, an integrator 262 and a decision logic 282. The multiplier circuits 241, 242 receive in the first case the filtered, delayed signal xFR1 individual to the sequence CEl and the filtered signal xF2 individual to the sequence CE2 and for the second the filtered, delayed signal XFR2 individual to the sequence CE2 and the filtered signal xFl individual to the sequence CE1.
These circuits 241, 242 supply demodulation signals m1(t) and m2(t), which, as for the simple variant of figs. 5 and 6, are at a zero mean value outside t=O. The integrators 261, 262 integrate these signals, as for fig. 6, and two logic decision circuits 281, 282 restore on two outputs Sr1, Sr2, the two data flows b1 k and b2 k.
This transmitter-receiver assembly functions as follows. The symbols d1 k are the result of the logic combination by exclusive-OR of b2 k and the compliment of d2 k 1 In the same way, the symbols d2 k are the result of the logic combination of b1 k and the compliment of d1 k 1 and it is there-fore possible to write:
d1 k = b2,k ~ dZ,k-1 d2 k = b1,k ~ d1,k-1 The first pseudorandom sequence CE1 affects the symbols d1 k~ whereas the B 12073.3 RS
216503~
second pseudorandom sequence CE2 affects the symbols d2 k. These sequences spread the spectrum of the symbols and new spread spectrum symbols D1 k and D2 k are obtained.
In the receiver, the filter 201 of the upper channel, which is matched to CEl, supplies a filtered signal xF1, which will depend on the symbol d2 k affected by CE1, whereas the filter 202 of the lower channel, which is matched to CE2 supplies a signal xF2, which will depend on the symbol d2 k~ which has been affected by CE2. However, the delay circuit 222 delays the filtered signal xF2 in such a way that the filtered, delayed signal xFR2 is dependent on the preceding symbol, i.e. d2 k 1' which has been affected by CE2. The demodulation performed by the circuit 242 there-fore has an effect on a transition of CE2 to CE1, which corresponds to the binary data item b2 of the second data flow and in fig. 8 to one of the rising oblique arrows.
The same reasoning, but symmetrically, shows that the upper channel of the demodulator processes the transitions of CE1 to CE2 and therefore the data bl .
The preceding description relates to the transmission of two symbols and the use of two orthogonal pseudorandom sequences. It is perfectly possible to extend this principle to n symbols and n sequences, where n is an integer higher than 2. In this case, it is e.g. possible to use a transmission, whose modulation is based on the principle of fig. 11 (compare with fig. 8) for 3 symbols and 3 sequences.
In this case, the symbol b1 modulates the phase jumps between a sequence CEl and the following sequence CE2. The symbol b2 modulates the phase 30 j umps between a sequence CE2 and the following sequence CE3. The symbol b3 modulates the phase jumps between a sequence CE3 and the following sequence CEl.
All the advantages described hereinbefore are retained. It is merely neces-sary for this purpose that the n sequences used are pairwise orthogonal.
B 12073.3 RS
21650~4 Another embodiment of the principle is illustrated in fig. 12 and makes it possible to transit, as in the case of fig. 8, two symbols, but with twice lower rates and still using two orthogonal pseudorandom sequences. In this case, the two sequences are alternately transmitted and the symbol bl modu-lates the phase jumps between a sequence CE1 and the following sequence CE2,whereas the symbol b2 modulates the phase jumps between a sequence CE2 and the following sequence CE1. The advantage is that due to the fact that at a given instant a single sequence is transmitted instead of two, i.e. all the transmission power is used for this sequence, which leads to an improvement in the quality of the signal received. However, the disadvantage is the reduction of the symbol rate.
The construction of the corresponding transmitter is slightly simplified compared with that described relative to fig. 9, a single modulator being sufficient, there is no adder and the switching of the sequences can be very easily digitally performed. However, the receiver has a structure strictly identical to that described relative to fig. 10.
Such a process for the transmission of several bits with the aid of several pseudorandom sequences offers the following advantages compared with con-ventional processes:
- a better resistance to scramblers and interference in the wide and narrow bands, taking account of the passage in the two filters having orthogonal pulse responses prior to demodulation, - a reduction of the parasitic noise outside the correlation peaks (in mean, effective value).
All the preceding description has related to the transmission of binary symbols, i.e. a two-state phase modulation. However, the invention is not limited to this case and is applicable to phase modulations with a larger number of states. The most frequently encountered conventional construc-tions use DQPSK modulation with four phase states permitting the transmis-sion of two bit symbols. The present invention applies to said DQPSK modul-ation. As the phase jumps can assume four possible values they are jumps between two orthogonal sequences instead of being on the same sequence, as in the conventional case.
B 12073.3 RS
.
The pseudorandom sequences used for spreading the spectrum are not necess-arily binary sequences. The invention can be implemented with sequences having a larger number of states, e.g. quaternary sequences, or even more complex pseudorandom signals.
A particularly interesting case consists of using as a pair of orthogonal pseudorandom sequences, two reciprocal sequences (one being the time inverse of the other). Starting from a sequence in C(t), a reciprocal sequence is obtained by transforming t into Tb-t. The use of reciprocal sequences is e.g. described in WO 92/02997.
This use of two reciprocal sequences make it possible to construct a recei-ver using a surface acoustic wave component simultaneously performing the matched and delayed filtering operations. A receiver using such a component is described in fig. 13.
The component 40 has an input transducer Te and four output transducers T 1~ T'1, T 2' TS2, constituted by aluminium electrodes deposited on a piezoelectric, e.g. quartz substrate. These four transducers are connected to four outputs S1, Sr1, S2,Sr2.
The four output transducers have an identical pulse response of very short duration, equal to or below the duration of one element of the pseudorandom sequences. The global pulse response of the component is, under these conditions, imposed by the geometry of the input transducer. The latter comprises N elementary electrodes in split-finger transducer form, which are interconnected in such a way that the pulse response of the filter between the input E and an output S1 is equal to the time-reverse combined complex of the first sequence CE1. Obviously, if the two output transducers T 1 and TS2 are arranged symmetrically on either side of the input transducer Te, the pulse responses of the thus formed two filters will be reciprocal of one another.
In the same way, by placing two remote transducers T'1 and T'2 on the sub-strate, in such a way that the distance between T 1 and T 2 or betweenTs~ and T'2 is such that the time taken by the acoustic RAYLEIGH wave to B 12073.3 RS
~165034 .
cover the interval separating the transducers is equal to Tb, two delay lines of duration Tb will be obtained on the same component. Thus, this single component 40 can be used in a spectrum spread differential receiver using alternating reciprocal sequences.
s It is obviously possible to use two distinct components, each having an input transducer matched to one of the two orthogonal sequences and two output transducers making it possible to obtain filtered signals and filtered, delayed signals.
B 12073.3 RS
-sequence of symbols dk obtained after the differential coding by a pseudo-random sequence having a much higher flow rate than that of the data to be transmitted. This pseudorandom sequence has a binary rate ~ times higher than that of the binary data to be transmitted. The duration Tc of one bit of this pseudorandom sequence, said bit also being called a chip, is there-fore equal to Tb/N. The chip rate of the pseudorandom sequence can be several megachips, or even several dozen megachips per second.
Fig. 3 shows the block diagram of a direct sequence spectrum spread trans-mitter. The elements already shown in fig. 1 carry the same references.
In addition to those of fig. 1, the emitter comprises a pseudorandom (C) sequence generator 30 and a multiplier 32. The signal supplied by the multiplier has a pseudorandom sequence-spread spectrum. The modulator 18 then no longer operates on the symbol of origin dk, but on the corresponding spread symbol Dk.
The corresponding receiver has the same structure as that of fig. 2, except that the filter 20 is now a matched filter, whose pulse response reduced to the baseband is the time-reversed combined complex of the pseudorandom sequence used in the transmitter.
The pseudorandom sequence used on transmission must consequently have an autocorrelation function with a marked peak of value N for a zero delay and the smallest possible side lobes. This can be obtained by using maximum length sequences also known as m-sequences, or so-called GOLD or KASAMI
sequences.
The sequences usable in a spectrum spread system have been the subject matter of intense studies, which are summarized in the article entitled "Cross-correlation properties of pseudorandom and related sequences" by Dilip V. SARWATE and Michael B. PURSLEY, published in "Proceedings of the IEEE", vol. 68, No. 5, May 1980, pp 593 to 619.
The input filter used in a spectrum spread receiver has an equivalent base-band pulse response H(t) and this response must be the time-reversed com-bined complex of the pseudorandom sequence used on transmission:
B 12073.3 RS
21~5034 H(t) = c*(T - t) where C(t) is the pseudorandom sequence used on transmission.
Therefore the signal supplied by such a filter is:
xF(t) = x(t) * H(t) where the symbol * designates the convolution operation, so as to give:
~t XF(t) = ~/ x(s).C* (Ts-t+s)ds.
t-T
Thus, the matched filter performs the correlation between the signal applied to its input and the pseudorandom spread sequence.
In a Gaussian additive noise channel, the signal xF(t) will therefore be in the form of a pulse signal, whose repetition frequency is equal to 1/Tb.
The envelope of this signal is the autocorrelation function of the signal C(t). The information is carried by the phase difference between two succ-essive correlation peaks. Thus, the multiplier output will be formed by a succession of positive or negative peaks, as a function of the transmitted bit value.
In the case of a radiotransmission in the presence of multiple paths, the output of the matched filter will be formed by a succession of correlation peaks, each peak corresponding to a propagation path.
The different signals of the reception chain are shown in fig. 4. Theline (a) represents the filtered signal xF(t), the line (b) the correlation signal xF(t)*xF(Ts-t) and the line (c) the signal at the integrator output.
This known procedure is described in detail in the article entitled "Direct-Sequence Spread Spectrum with DPSK Modulation and Diversity for Indoor Wireless Communications" published by Mohsen KAVEHRAD and Bhaskar RAMAMURTHI
in "IEEE Transactions on Communications", vol. COM 35, No. 2, February 1987.
Thus, a description has been given of DPSK spectrum spread modulation. It is clear that this procedure can be applied in the same way to all differ-B 12073.3 RS
216~03~
-ential modulations. For example, in the case of DQPSK modulation, each correlation peak at the output of the matched filter could assume four phase states, whereas there are only two with a DPSK modulation.
Despite their great interest these procedures suffer from disadvantages.
Thus, the successive bits constituting the message to be transmitted are all multiplied by the same pseudorandom spread sequence, which can give rise to a coherent intercorrelation noise able to deteriorate the quality of the signal m(t) at the demodulator output.
The applicant has profoundly studied this question and the research carried out can be summarized as follows. Consideration must again be given to the transmitter shown in fig. 3 and the receiver of fig. 2, knowing that the latter operates with a matched filter 20 corresponding to a given pseudo-random sequence. In general terms, the sequence used on reception isdesignated CR and that used on transmission CE. When the transmitter and receiver are matched to one another, one obviously obtains CE-CR.
In baseband, the signal at the input of the matched filter 20 can be written:
x(t) = dCE(t) _ representing the sequence of bits modulating the carrier, said bits being obtained from binary data to be transmitted b following passage in the differential coder 10.
The filtered signal found at the output of the matched filter can then be written: t *
xFt() = ~ x(s)CR (s - t + Tb)ds t-Tb CR representing the combined complex of CR, which can then be written:
O *
x (t) = d 1 ~ CE(s + TbCR~s - t + Tb)ds + do f CE(s)CR(s - t + Tb)ds t-Tb o or once again x(t) = d 1CC C (-t) + d0CC C (Tb ~ t) where the notation Cc C represents the aperiodic correlation function between CE and CR.
B 12073.3 RS
21650~
In the case where CE-CR the same formula is obtained by replacing the CC C by Cc ~ whlch is the aperiodic autocorrelation function of CR.
The delayed filtered signal found at the output of the delay circuit 22 is simply shifted by one symbol and can consequently be written:
xFR(t) = d 2CC C (-t) + d lCC C (Tb ~ t).
Finally, at the output of the demodulator 24, the signal can be written:
m(t)=xF(t).xFR(t) or d d c2 (-t) + l d 1 + d-2dO CCE,CR CE,CR b +d ldOCC , CR ( b As the symbols d are obtained from data b by the relation dk = bk63 dk 1 and as d or _ can only assume the two values +1 or -1, we necessarily obtain:
d-2d-1 = b_l~ d_1-do = bo and d_1 + d 2.do = 1 + b 1.bo Thus, one once again obtains a new expression of the demodulation sig20 m(t):
m(t) = b lCc C (-t) + ll + bob 1~Cc C (-t)Cc C (Tb ~ t) bO CE, CR( b (t) Three cases can then occur as a function of the respective values of boand b 1-Case No. 1: bo = b-1 We then obtain: 2 ~ CE~ R E" R b ] E, R
in which 4C C is called the even correlation function of CE and CR.
Case No. 2: b 1 = bo = -1 -2 ^2 3s m(t) lCCE,CR( t) CE'CR( b ~ CE.CR b B 12073.3 RS
216503~
, g where C C is called the odd autocorrelation function of CE and CR.
Case No. 3: b 1 = -bo m(t) = bo ICC C (Tb ~ t) - Cc C (-t)] = bo~c C (Tb ~ t) ~C C (Tb ~ t) With these considerations developed, the deficiency of the conventional direct sequence spectrum spread method becomes very clear. Thus, at least in the first two cases, the demodulation signal m(t) is either always positive, or always negative. However, it is this signal which will be inte-grated into the circuit 26 prior to decision taking. Apart from the corre-lation peak obtained for t=O, this signal should ideally have a zero mean value. This is manifestly not the case and the integrator output will consequently have at the time of decision making, a significant negative or positive shift value as a result of the integration of a signal, whose mean value is not zero. These shifts will deteriorate the quality of the deci-sion taking and will increase the bit error rate and this will become worse as the integration time increases.
The object of the invention is to obviate this disadvantage and aims atimproving the quality of the output signal of the demodulator (signal m(t)) and therefore the transmission quality.
DESCRIPTION OF THE INVENTION
The invention therefore proposes a novel transmission process by spectrum spread phase differential modulation and demodulation in which use is made of two pseudorandom sequences on transmission, which are orthogonal to one another, said two sequences being used for spreading the spectrum of the signals, which will then modulate a carrier. On reception, two filtering operations matched to these two orthogonal sequences will take place. In order to obtain the demodulation signal, multiplication will take place of the delayed filtered signal associated with one of the sequences by the filtered, but not delayed signal associated with the other sequence. As these two signals result from filtering operations by filters, whose pulse responses are orthogonal in the NYQUIST sense, the interference affecting these signals will be decorrelated, which will decrease their influence on B 12073.3 RS
the demodulation signal. In the prior art, the two signals, respectively delayed and filtered and filtered emanated from the same filter, so that the interference affecting them was correlated.
More specifically, the present invention relates to a data transmissionprocess by spectrum spread phase differential modulation and demodulation, in which the data are organized in at least one binary data sequence having a certain period, characterized in that:
A) on transmission:
- a differential coding of the binary signals to be transmitted takes place in order to obtain symbols, use being made of two orthogonal pseudorandom sequences for spreading the spectrum of each symbol and a carrier is modu-lated by the spread spectrum symbols, B) on reception:
- two parallel filtering operations take place matched to the two orthogonal sequences and there is an estimation of the phase difference between the two signals obtained after filtering, one of which has been delayed, which restores the data.
The invention has several embodiments according to which transmission takes place of one, two or more than two data flows.
BRIEF DESCRIPTION OF THE DRAWINGS
Fig, 1, already described, shows the block diagram of a prior art, differ-ential coding transmitter.
Fig. 2, already described, shows the block diagram of a corresponding receiver.
Fig. 3, already described, shows the block diagram of a prior art, direct sequence spectrum spread transmitter.
Fig. 4, already described, shows the configuration of the different signals appearing in a direct sequence spectrum spread receiver.
B 12073.3 RS
Fig. 5 is a block diagram of a transmitter according to the invention in afirst variant with a single flow of data to be transmitted.
Fig. 6 is a block diagram of a corresponding receiver.
Fig. 7 symbolically shows the transmission principle of a single data flow.
Fig. 8 symbolically shows the principle of simultaneously transmitting two data flows.
Fig. 9 shows the structure of a transmitter corresponding to a double data transmission.
Fig. 10 shows the structure of the corresponding receiver.
Fig. 11 symbolically illustrates a transmission of three data flows with three pairwise orthogonal pseudorandom sequences.
Fig. 12 illustrates a transmission variant of two symbols with the half flow rates.
Fig. 13 illustrates an embodiment of a receiver having a surface acoustic wave component.
Fig. 5 shows a diagram of an embodiment of a transmitter according to the invention. This transmitter comprises the means already shown in fig. 3 and which carry the same references, namely an input Ee receiving the binary data bk to be transmitted, said data having a period Tb, a differential coder 10 with its logic circuit 12 and its delay circuit 14. This coder supplies the symbols dk. The transmitter shown also comprises a first generator 301 supplying a first pseudorandom sequence CEl applied to a first multiplier 321 and a second generator 32 supplying a second pseudorandom sequence CE2 applied to a second multiplier 322. The multipliers 321 and 322 supply symbols (Dk)1 and (Dk)2, whose spectrum is spread by the B 12073.3 RS
216503~
corresponding sequences. The transmitter also comprises a first modulator 181 receiving a carrier P produced by a local oscillator 16 and the symbols (Dk)1 spread by the first pseudorandom sequence CE1 and a second modulator 182 receiving said same carrier P and the symbols (Dk)2 spread by the second pseudorandom sequence CE2. Finally, an adder 31 is connected to the two modulators 181 and 182 and supplies the output Se of the transmitter. This output transmits the carrier modulated by the doubly spread signal. The modulation can be of a random nature (phase, frequency, mixed, etc.).
Naturally, it would be possible to modify this circuit by firstly adding the two spread signals (Dk)1 and (Dk)2 and by using a single modulator. It would also be possible to add the two pseudorandom sequences CE1 and CE2 as soon as they are formed and to spread the signal with the aid of the pseudo-random sum obtained.
The receiver shown in fig. 6 comprises the means already shown in fig. 2, namely an input Er receiving a signal x(t), a delay circuit 22 with a dura-tion equal to the period Tb of the transmitted bits, a multiplier 24 supply-ing a demodulation signal m(t), an integrating circuit 26 acting on the signal m(t) supplied by the multiplier 24, the integration taking place between 0 and Tb, a logic decision circuit 28 and a general output Sr supply-ing the original digital data bk. This receiver differs from the prior art in that it comprises a first filter 201 matched to the first pseudorandom sequence CE1 used in the transmitter and a second filter 202 matched to the second pseudorandom sequence CE2 used in the transmitter. The term filter matched to a pseudorandom sequence is understood to mean a filter, whose pulse response brought down to the baseband is the time-reversed combined complex of the pseudorandom sequence in question.
The first filter 201 supplies a filtered signal xF1, which is then delayed by Tb in a circuit 22 in order to give a delayed, filtered signal xFR1.
The second filter 202 supplies a filtered signal xF2. This signal is not delayed and is used for producing the demodulation signal m(t) by multipli-cation with the filtered, delayed signal xFR1.
This transmitter-receiver assembly functions as follows. As in the case of B 12073.3 RS
` 216503~
conventional modulation with direct sequence spectrum spreading, the binary quantities used for modulating the carrier are the symbols dk obtained from the original data bk following passage in the differential coder 10.
In the conventional process, the symbols dk are multiplied by a single pseudorandom sequence CE. In the process according to the invention, the symbols dk are multiplied by the two pseudorandom sequences CE1 and CE2, which are orthogonal to one another, in order to form a signal which will modulate the carrier.
The sequences CE1 and CE2 can have the same general characteristics as a sequence used in the prior art. They are both constituted by N elements or chips having a period T = b and their autocorrelation functions have a single marked peak for a zero delay. They have the lowest possible value elsewhere. These sequences are also chosen so as to have a low inter-correlation function, no matter what the time.
In the case of the receiver, the input signal x(t) can be written:
x(t)=d(CE1(t)+CE2(t)) The filtered, delayed signal xFR1 at the output of the delay circuit 22 can be written:
XFRI = d-2 ~CcEl~cEl( t) CCE2,CEl( t)]
+d 1 [CcEl CE1(Tb ~ t) + CcE2 CE1(Tb ~ t~
In the same way, the filtered signal xF2 at the output of the filter 202 can be written:
xF2 - d_1 [CcEl~cE2( t) CCE2,CE2( )]
+d lC (T - t) + C (T - t)~
At the output of the multiplier 24, the demodulation signal is of form:
m(t)=xFR1 xF2 Therefore four cases can be distinguished:
Case No. 1: bo =b 1=1 B 12073.3 RS
216~034 We then have d 2=d 1=do. It is then possible to rewrite xFR1 and xF2 in the following form:
xFR1 d-1 ~ CE1(Tb t) CE2,CE1(Tb t) and xF2 d-l [ CEl,CE2( b t) CE2(Tb t)]
where, after simplification and replacing OCEi CEj(TB - t) with iJ
m(t) ~ 12 ~ 22 ~21 ~12 ~21 ~22 For t=Tb, we clearly obtained m(0)=C11(0).022(0)+ .. i.e. N , which corres-ponds to the autocorrelation peak.
A detailed analysis of the signal m(t) makes it possible to see that in the case where the two sequences used for modulating the data are orthogonal, the signal m(t) has a zero mean value outside the correlation peak. Thus, in this case, the output of the integrator 26 does not have a shift as was the case with the conventional structure.
Case No. 2: bo = b 1 = -1 In this case, we obtain do=d 1=-d 2. Thus, it is possible to rewrite x and xF2 in the following form, with the same simplifying notations as hereinbefore:
r~ A
~ l [ ^ ]
A ~, A ,~ ~ A A A
so that m(t) 11 12 11 22 21 12 21 22 All the remarks made in connection with case No. 1 still apply.
Case No. 3: bo=1 b 1=-l In this case, we obtain do=d l=-d 2 and therefore _ F2 -1 _ 12 22 B 12073.3 RS
216503~
-.~ ~ ,'~ A
and m(t) ~11 12 21 12 11 22 21 22 The same remarks still apply.
Case ~o. 4: bo=-l b 1=l In this last case, we have do=-d l=-d 2 and xFRl = d 1 ¦~11 + ~21 10F2 -1 [ 12 22~
so that m(t) 11~12 ~ 22 ~21~12 ~21~22 The same remarks still apply.
The two signals received by the demodulator 24 are different both as respect to their serial position and by the pseudorandom sequence involved. The signal xFRl has been delayed by a period Tb and therefore corresponds, on transmission, to a symbol of serial position k-l, whereas the signal xF2 corresponds to the following symbol of serial position k. In addition, the signal xFR1 comes from a symbol which has been spread and then despread by the first sequence CEl, whereas the signal xF2 has been processed by the second sequence CE2. As a result of the delay of one serial position between the symbols in question, it can be seen that the sequences CEl and CE2 have not been transmitted at the same time. Each sequence is period-ically transmitted with the same period Tb as the symbols to be processed, but the sequence CEl linked wit the signal xFR1 has been transmitted one serial position prior to the sequence CE2 linked with the signal xF2. The two sequences CEl and CE2, although globally produced simultaneously by the generators 301 and 32' intervene in a manner shifted by one serial position as a result of the processing performed on reception of the sequences.
Thus, use is then made of a sequence CEl and then a sequence CE2.
This process is diagrammatically illustrated in fig. 7. The first sequence CEl is repeated periodically for each symbol dk 1~ dk~ dk+l, etc., as is thesecond sequence CE2. These sequences are symbolized by the rectangles of B 12073.3 RS
216503~
-respectively the upper and lower bands. On reception, the demodulation signal m(t) is formed from a signal xFR1, whose origin is a symbol of serial position k-l processed by CE1 and a signal xFl originating from a symbol of serial position k processed by CE2. The downward oblique arrow from a sequence CE1 of serial position k-1 to a sequence CE2 of serial position k symbolizes the chronology of the sequences used. This chronology is repea-ted on each formation of a new demodulation signal. All the downward oblique arrows symbolize the transmission of all the symbols dk, when k increases unitwise.
In the embodiment illustrated in figs. 5 and 6, this distinction between the serial position of the sequences involved has no importance because a sequence is identically repeated (the sequence CE1 of serial position k-1 being identical to the sequence CE1 of serial position k). However, the distinctions described hereinbefore give a better understanding of the improvement provided by the invention and which is as follows.
It is possible to conceive a receiver using a first matched filter on CE1 and which would not be followed by a delay circuit, the latter being associated with the second filter matched to CE2. Demodulation would then take place of the phase jumps caused by CE2 followed by CEl.
In other words, it is possible to exploit the presence of two pseudorandom sequences on transmission by simultaneously transmitting two data flows and by detecting the phase jumps in both directions. This is what is symbolic-ally shown in fig. 8, where at the top there is a succession of pseudo-random sequences CE1 and at the bottom a successive of pseudorandom sequen-ces CE2. Data b1 impose phase jumps between a sequence CE1 and the follow-ing sequence CE2. Data b2 impose phase jumps between a sequence CE2 and the following sequence CE1. The rising or falling oblique arrows symbolize the two processes.
Figs. 9 and 10 illustrate how it is necessary to modify the diagrams of the transmitter and receiver compared with figs. 5 and 6 in order to effect this double transmission. The transmitter of fig. 9 has two inputs Ee1, Ee2 receiving two binary data flows respectively b1 k and b2 k. It comprises B 12073.3 RS
216~34 .
two logic circuits 121, 122 and two delay circuits 141, 142, whose input is connected to the output of one of the logic circuits and the output is connected to the input of the other logic circuit. A first multiplier 321 processes the symbols dl k~ present at the output of the logic circuit 122 and a second multiplier 322 the symbols d2 k present at the output of the logic circuit 121. There is a local oscillator 16 supplying the carrier, two modulators 181, 182 and an adder 31 supplying at its output Se the doubly modulated carrier.
On the receiver side and as is illustrated in fig. 10, there are two symmet-rical channels crossed at the outset, a first channel with a filter 201 matched to the sequence CE1 and a delay circuit 221 and a second channel with a second filter 202 matched to the sequence CE2 and a delay circuit 222.
Thus, the first channel comprises a multiplier 241, an integrator 261 and a decision logic 281. The second channel comprises a multiplier 242, an integrator 262 and a decision logic 282. The multiplier circuits 241, 242 receive in the first case the filtered, delayed signal xFR1 individual to the sequence CEl and the filtered signal xF2 individual to the sequence CE2 and for the second the filtered, delayed signal XFR2 individual to the sequence CE2 and the filtered signal xFl individual to the sequence CE1.
These circuits 241, 242 supply demodulation signals m1(t) and m2(t), which, as for the simple variant of figs. 5 and 6, are at a zero mean value outside t=O. The integrators 261, 262 integrate these signals, as for fig. 6, and two logic decision circuits 281, 282 restore on two outputs Sr1, Sr2, the two data flows b1 k and b2 k.
This transmitter-receiver assembly functions as follows. The symbols d1 k are the result of the logic combination by exclusive-OR of b2 k and the compliment of d2 k 1 In the same way, the symbols d2 k are the result of the logic combination of b1 k and the compliment of d1 k 1 and it is there-fore possible to write:
d1 k = b2,k ~ dZ,k-1 d2 k = b1,k ~ d1,k-1 The first pseudorandom sequence CE1 affects the symbols d1 k~ whereas the B 12073.3 RS
216503~
second pseudorandom sequence CE2 affects the symbols d2 k. These sequences spread the spectrum of the symbols and new spread spectrum symbols D1 k and D2 k are obtained.
In the receiver, the filter 201 of the upper channel, which is matched to CEl, supplies a filtered signal xF1, which will depend on the symbol d2 k affected by CE1, whereas the filter 202 of the lower channel, which is matched to CE2 supplies a signal xF2, which will depend on the symbol d2 k~ which has been affected by CE2. However, the delay circuit 222 delays the filtered signal xF2 in such a way that the filtered, delayed signal xFR2 is dependent on the preceding symbol, i.e. d2 k 1' which has been affected by CE2. The demodulation performed by the circuit 242 there-fore has an effect on a transition of CE2 to CE1, which corresponds to the binary data item b2 of the second data flow and in fig. 8 to one of the rising oblique arrows.
The same reasoning, but symmetrically, shows that the upper channel of the demodulator processes the transitions of CE1 to CE2 and therefore the data bl .
The preceding description relates to the transmission of two symbols and the use of two orthogonal pseudorandom sequences. It is perfectly possible to extend this principle to n symbols and n sequences, where n is an integer higher than 2. In this case, it is e.g. possible to use a transmission, whose modulation is based on the principle of fig. 11 (compare with fig. 8) for 3 symbols and 3 sequences.
In this case, the symbol b1 modulates the phase jumps between a sequence CEl and the following sequence CE2. The symbol b2 modulates the phase 30 j umps between a sequence CE2 and the following sequence CE3. The symbol b3 modulates the phase jumps between a sequence CE3 and the following sequence CEl.
All the advantages described hereinbefore are retained. It is merely neces-sary for this purpose that the n sequences used are pairwise orthogonal.
B 12073.3 RS
21650~4 Another embodiment of the principle is illustrated in fig. 12 and makes it possible to transit, as in the case of fig. 8, two symbols, but with twice lower rates and still using two orthogonal pseudorandom sequences. In this case, the two sequences are alternately transmitted and the symbol bl modu-lates the phase jumps between a sequence CE1 and the following sequence CE2,whereas the symbol b2 modulates the phase jumps between a sequence CE2 and the following sequence CE1. The advantage is that due to the fact that at a given instant a single sequence is transmitted instead of two, i.e. all the transmission power is used for this sequence, which leads to an improvement in the quality of the signal received. However, the disadvantage is the reduction of the symbol rate.
The construction of the corresponding transmitter is slightly simplified compared with that described relative to fig. 9, a single modulator being sufficient, there is no adder and the switching of the sequences can be very easily digitally performed. However, the receiver has a structure strictly identical to that described relative to fig. 10.
Such a process for the transmission of several bits with the aid of several pseudorandom sequences offers the following advantages compared with con-ventional processes:
- a better resistance to scramblers and interference in the wide and narrow bands, taking account of the passage in the two filters having orthogonal pulse responses prior to demodulation, - a reduction of the parasitic noise outside the correlation peaks (in mean, effective value).
All the preceding description has related to the transmission of binary symbols, i.e. a two-state phase modulation. However, the invention is not limited to this case and is applicable to phase modulations with a larger number of states. The most frequently encountered conventional construc-tions use DQPSK modulation with four phase states permitting the transmis-sion of two bit symbols. The present invention applies to said DQPSK modul-ation. As the phase jumps can assume four possible values they are jumps between two orthogonal sequences instead of being on the same sequence, as in the conventional case.
B 12073.3 RS
.
The pseudorandom sequences used for spreading the spectrum are not necess-arily binary sequences. The invention can be implemented with sequences having a larger number of states, e.g. quaternary sequences, or even more complex pseudorandom signals.
A particularly interesting case consists of using as a pair of orthogonal pseudorandom sequences, two reciprocal sequences (one being the time inverse of the other). Starting from a sequence in C(t), a reciprocal sequence is obtained by transforming t into Tb-t. The use of reciprocal sequences is e.g. described in WO 92/02997.
This use of two reciprocal sequences make it possible to construct a recei-ver using a surface acoustic wave component simultaneously performing the matched and delayed filtering operations. A receiver using such a component is described in fig. 13.
The component 40 has an input transducer Te and four output transducers T 1~ T'1, T 2' TS2, constituted by aluminium electrodes deposited on a piezoelectric, e.g. quartz substrate. These four transducers are connected to four outputs S1, Sr1, S2,Sr2.
The four output transducers have an identical pulse response of very short duration, equal to or below the duration of one element of the pseudorandom sequences. The global pulse response of the component is, under these conditions, imposed by the geometry of the input transducer. The latter comprises N elementary electrodes in split-finger transducer form, which are interconnected in such a way that the pulse response of the filter between the input E and an output S1 is equal to the time-reverse combined complex of the first sequence CE1. Obviously, if the two output transducers T 1 and TS2 are arranged symmetrically on either side of the input transducer Te, the pulse responses of the thus formed two filters will be reciprocal of one another.
In the same way, by placing two remote transducers T'1 and T'2 on the sub-strate, in such a way that the distance between T 1 and T 2 or betweenTs~ and T'2 is such that the time taken by the acoustic RAYLEIGH wave to B 12073.3 RS
~165034 .
cover the interval separating the transducers is equal to Tb, two delay lines of duration Tb will be obtained on the same component. Thus, this single component 40 can be used in a spectrum spread differential receiver using alternating reciprocal sequences.
s It is obviously possible to use two distinct components, each having an input transducer matched to one of the two orthogonal sequences and two output transducers making it possible to obtain filtered signals and filtered, delayed signals.
B 12073.3 RS
Claims (10)
1. Process for data transmission by spectrum spread phase differential modulation and demodulation, the data being organized in at least one binary data sequence (bk having a certain period (Tb), characterized in that:
A) on transmission:
- there is a differential coding of the binary data to be transmitted (bk) to obtain symbols (dk), use being made of two orthogonal pseudorandom sequences (CE1, CE2) in order to spread the spectrum of each symbol and a carrier (P) is modulated by the spread spectrum symbols (Dk), B) on reception:
- two parallel filtering operations matched to the two orthogonal sequences (CE1, CE2) are performed and the phase difference is estimated between the signals (xF, xFR) obtained after filtering, one (xF) having been delayed, which restores the data (bk).
A) on transmission:
- there is a differential coding of the binary data to be transmitted (bk) to obtain symbols (dk), use being made of two orthogonal pseudorandom sequences (CE1, CE2) in order to spread the spectrum of each symbol and a carrier (P) is modulated by the spread spectrum symbols (Dk), B) on reception:
- two parallel filtering operations matched to the two orthogonal sequences (CE1, CE2) are performed and the phase difference is estimated between the signals (xF, xFR) obtained after filtering, one (xF) having been delayed, which restores the data (bk).
2. Process according to claim 1, characterized in that with the data to be transmitted organized in a single binary data sequence (bk):
A) on transmission:
- a first pseudorandom sequence (CE1) is produced, - the symbols (dk) are multiplied by said first pseudorandom sequence (CE1) in order to obtain first spread spectrum symbols (Dk)1, - a second pseudorandom sequence (CE2) orthogonal to the first (CE1) is produced, - the symbols (dk) are also multiplied by said second pseudorandom sequence (CE2) in order to obtain second spread spectrum symbols (Dk)2, - the carrier (P) is modulated by the first spread spectrum symbols (Dk)1 and the second spread spectrum symbols (Dk)2, B) on reception:
- the modulated carrier (x(t)) is received, - a first filtering takes place of the received symbol (x(t)) by a first matched filter (201) having a pulse response, which is the time inverse of the first pseudorandom sequence used on transmission (CE1) and a first filtered signal (xF1) is obtained, - a second filtering of the signal received (x(t)) is performed by a second matched filter (202) having a pulse response, which is the time inverse of the second pseudorandom sequence used on transmission (CE2) and a second filtered signal (xF2) is obtained, - one of the filtered signals is delayed by one period (Tb) to obtain a filtered, delayed signal (xFR), - the filtered, delayed signal (xFR) is multiplied by the filtered signal (xF1) to obtain a demodulation signal (m(t)).
A) on transmission:
- a first pseudorandom sequence (CE1) is produced, - the symbols (dk) are multiplied by said first pseudorandom sequence (CE1) in order to obtain first spread spectrum symbols (Dk)1, - a second pseudorandom sequence (CE2) orthogonal to the first (CE1) is produced, - the symbols (dk) are also multiplied by said second pseudorandom sequence (CE2) in order to obtain second spread spectrum symbols (Dk)2, - the carrier (P) is modulated by the first spread spectrum symbols (Dk)1 and the second spread spectrum symbols (Dk)2, B) on reception:
- the modulated carrier (x(t)) is received, - a first filtering takes place of the received symbol (x(t)) by a first matched filter (201) having a pulse response, which is the time inverse of the first pseudorandom sequence used on transmission (CE1) and a first filtered signal (xF1) is obtained, - a second filtering of the signal received (x(t)) is performed by a second matched filter (202) having a pulse response, which is the time inverse of the second pseudorandom sequence used on transmission (CE2) and a second filtered signal (xF2) is obtained, - one of the filtered signals is delayed by one period (Tb) to obtain a filtered, delayed signal (xFR), - the filtered, delayed signal (xFR) is multiplied by the filtered signal (xF1) to obtain a demodulation signal (m(t)).
3. Process according to claim 1, characterized in that, the data being organized in at least two separate sequences of binary data (b1,k)(b2,k) of the same period (Tb), it comprises the following operations:
A) on transmission:
- a first and a second sequences of symbols (d1,k) (d2,k) is formed from two binary data sequences (b1,k)(b2,k), one symbol of a sequence (d2,k)(d1,k) being obtained by a logic combination between a data item (b1,k), (b2,k) and a symbol of the other sequence delayed by a period equal to the period (Tb) (d1,k-1,d2,k-1), - a first pseudorandom sequence (CE1) is produced and the symbols of the first sequence (d1,k) is multiplied by said first pseudorandom sequence (CE1) in order to obtain a first sequence of spread spectrum symbols (D1,k), - a second pseudorandom sequence (CE2), orthogonal to the first sequence (CE1) is produced and the symbols of the second seequence (d2,k) are multi-plied by said second pseudorandom sequence (CE2) to obtain a second sequence of spread spectrum symbols (D2,k), - the carrier (P) is modulated on the one hand by the first sequence of spread spectrum symbols (D1,k) and on the other by the second sequence of spread spectrum symbols (D2,k), - the two modulated parts of the carrier are summated and the resulting carrier is transmitted, B ) on reception:
- the modulated carrier (x(t)) is received, - there is a first filtering of the signal received (x(t)) by a first matched filter (201) having a pulse response, which is the time inverse of the first pseudorandom sequence used on transmission (CE1) and a first filtered signal (xF1) is obtained, - there is a second filtering of the signal received (x(t)) by a second matched filter (202) having a pulse response, which is the time inverse of the second pseudorandom sequence used on transmission (CE2) and a second filtered signal (xF2) is obtained, - the first filtered signal (xF1) is delayed by a period equal to the period (Tb) of the binary data (b1,k) (b2,k) and a first filtered, delayed signal (xFR1) is obtained, - the second filtered signal (xF2) is delayed by a period equal to the period (Tb) of the binary data (b1,k)(b2,k) and a second filtered, delayed signal (xFR2) is obtained, - the first filtered, delayed signal (xFR1) is multiplied by the second filtered signal (xF2) to obtain a first demodulation signal (m1(t)), - the second filtered, delayed signal (xFR2) is multiplied by the first filtered signal (xF1) to obtain a second demodulation signal (m2(t)), - the first demodulation signal (m1(t)) is integrated for a duration equal to the period (Tb) of the binary data (b1,k)(b2,k), - the second demodulation signal (m2(t)) is integrated on a duration equal to the period (Tb) of the binary data (b1,k)(b2,k), - an evaluation takes place of the sign of the signal resulting from the integration of the first signal and as a consequence restoration takes place of the binary data of the first sequence (b1,k), - evaluation takes place of the sign of the signal resulting from the integration of the second signal and as a consequence the binary data of the second sequence (b2,k) is restored.
A) on transmission:
- a first and a second sequences of symbols (d1,k) (d2,k) is formed from two binary data sequences (b1,k)(b2,k), one symbol of a sequence (d2,k)(d1,k) being obtained by a logic combination between a data item (b1,k), (b2,k) and a symbol of the other sequence delayed by a period equal to the period (Tb) (d1,k-1,d2,k-1), - a first pseudorandom sequence (CE1) is produced and the symbols of the first sequence (d1,k) is multiplied by said first pseudorandom sequence (CE1) in order to obtain a first sequence of spread spectrum symbols (D1,k), - a second pseudorandom sequence (CE2), orthogonal to the first sequence (CE1) is produced and the symbols of the second seequence (d2,k) are multi-plied by said second pseudorandom sequence (CE2) to obtain a second sequence of spread spectrum symbols (D2,k), - the carrier (P) is modulated on the one hand by the first sequence of spread spectrum symbols (D1,k) and on the other by the second sequence of spread spectrum symbols (D2,k), - the two modulated parts of the carrier are summated and the resulting carrier is transmitted, B ) on reception:
- the modulated carrier (x(t)) is received, - there is a first filtering of the signal received (x(t)) by a first matched filter (201) having a pulse response, which is the time inverse of the first pseudorandom sequence used on transmission (CE1) and a first filtered signal (xF1) is obtained, - there is a second filtering of the signal received (x(t)) by a second matched filter (202) having a pulse response, which is the time inverse of the second pseudorandom sequence used on transmission (CE2) and a second filtered signal (xF2) is obtained, - the first filtered signal (xF1) is delayed by a period equal to the period (Tb) of the binary data (b1,k) (b2,k) and a first filtered, delayed signal (xFR1) is obtained, - the second filtered signal (xF2) is delayed by a period equal to the period (Tb) of the binary data (b1,k)(b2,k) and a second filtered, delayed signal (xFR2) is obtained, - the first filtered, delayed signal (xFR1) is multiplied by the second filtered signal (xF2) to obtain a first demodulation signal (m1(t)), - the second filtered, delayed signal (xFR2) is multiplied by the first filtered signal (xF1) to obtain a second demodulation signal (m2(t)), - the first demodulation signal (m1(t)) is integrated for a duration equal to the period (Tb) of the binary data (b1,k)(b2,k), - the second demodulation signal (m2(t)) is integrated on a duration equal to the period (Tb) of the binary data (b1,k)(b2,k), - an evaluation takes place of the sign of the signal resulting from the integration of the first signal and as a consequence restoration takes place of the binary data of the first sequence (b1,k), - evaluation takes place of the sign of the signal resulting from the integration of the second signal and as a consequence the binary data of the second sequence (b2,k) is restored.
4. Process according to claim 3, characterized in that alternate trans-mission takes place of the two orthogonal sequences, a first symbol (b1) modulating the phase jumps between a first sequence (CE1) and a second, following sequence (CE2), whereas a second symbol (b2) modulates the phase jumps between the second sequence (CE2) and the first sequence (CE1).
5. Process according to claim 3, characterized in that with the binary data organized in a number n of sequences, n being an integer higher than 2, n pairwise orthogonal pseudorandom sequences are produced and the data of the two sequences are transmitted with the aid of two of the orthogonal pseudorandom sequences.
6. Process according to any one of the preceding claims, characterized in that the symbols (dk) are one bit.
7. Process according to any one of the claims 1 to 5, characterized in that the symbols (dk) are two bits.
8. Process according to any one of the preceding claims, characterized in that the first and second pseudorandom sequences (CE1, CE2) are binary sequences.
9. Process according to any one of the claims 1 to 7, characterized in that the first and second pseudorandom sequences (CE1, CE2) are quaternary sequences.
10. Process according to any one of the preceding claims, characterized in that the two orthogonal pseudorandom sequences (CE1, CE2) are reciprocal sequences.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| FR9415241A FR2728415B1 (en) | 1994-12-19 | 1994-12-19 | TRANSMISSION METHOD WITH DIFFERENTIAL SPREAD SPECTRUM PHASE MODULATION AND DEMODULATION USING ORTHOGONAL PSEUDORANDOM SEQUENCES |
| FR9415241 | 1994-12-19 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA2165034A1 true CA2165034A1 (en) | 1996-06-20 |
Family
ID=9469920
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA002165034A Abandoned CA2165034A1 (en) | 1994-12-19 | 1995-12-12 | Transmission process having spectrum spread phase differential modulation and demodulation using orthogonal pseudorandom sequences |
Country Status (5)
| Country | Link |
|---|---|
| US (1) | US5822363A (en) |
| EP (1) | EP0718983B1 (en) |
| CA (1) | CA2165034A1 (en) |
| DE (1) | DE69521671T2 (en) |
| FR (1) | FR2728415B1 (en) |
Families Citing this family (32)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH1022874A (en) | 1996-07-09 | 1998-01-23 | Hitachi Ltd | Cdma communication system and method therefor |
| US6289041B1 (en) | 1997-02-11 | 2001-09-11 | Snaptrack, Inc. | Fast Acquisition, high sensitivity GPS receiver |
| US6396880B1 (en) * | 1998-04-17 | 2002-05-28 | Analog Devices Inc | π/4 DQPSK encoder and modulator |
| US6208291B1 (en) * | 1998-05-29 | 2001-03-27 | Snaptrack, Inc. | Highly parallel GPS correlator system and method |
| US6567474B1 (en) * | 1999-03-02 | 2003-05-20 | Phonex Corporation | Digital wireless phone/modem jack capable of communications over the power lines using differential binary phase shift keying (DBPSK) |
| GB9922239D0 (en) * | 1999-09-20 | 1999-11-17 | Nokia Mobile Phones Ltd | A multiple access technique for mobile radio systems |
| FR2813474B1 (en) * | 2000-08-28 | 2002-12-13 | Commissariat Energie Atomique | NON-CONSISTENT DP-MOK RECEPTION METHOD WITH MULTIPLE PATH COMBINATION AND CORRESPONDING RECEIVER |
| JP4278332B2 (en) * | 2001-06-29 | 2009-06-10 | 日本電信電話株式会社 | Optical transmitter and optical transmission system |
| US7596190B2 (en) * | 2002-04-01 | 2009-09-29 | Qualcomm Incorporated | System, method, and apparatus for correction of code doppler shift |
| US7301993B2 (en) * | 2002-09-13 | 2007-11-27 | Broadcom Corporation | Channel estimation in a spread spectrum receiver |
| US7532682B1 (en) * | 2002-11-27 | 2009-05-12 | Schell Stephan V | Quadrature modulation without carrier |
| AU2003903271A0 (en) * | 2003-06-26 | 2003-07-10 | Commonwealth Scientific And Industrial Research Organisation | A device and method for encoding or decoding a datum |
| CN1940589B (en) * | 2005-09-30 | 2010-04-21 | 凌阳科技股份有限公司 | Method and system for synchronizing data bit of global satellite positioning system |
| US20110188389A1 (en) | 2008-07-04 | 2011-08-04 | Commonwealth Scientific And Industrial Research Organisation | Wireless Localisation System |
| US9057606B2 (en) | 2009-09-10 | 2015-06-16 | Nextnav, Llc | Wide area positioning system |
| US9035829B2 (en) | 2008-09-10 | 2015-05-19 | Nextnav, Llc | Wide area positioning systems and methods |
| US8130141B2 (en) | 2008-09-10 | 2012-03-06 | Commlabs, Inc. | Wide area positioning system |
| US8917209B2 (en) | 2009-09-10 | 2014-12-23 | Nextnav, Llc | Coding in a wide area positioning system (WAPS) |
| US8229041B2 (en) * | 2009-05-26 | 2012-07-24 | Broadcom Corporation | Direct detection of wireless interferers in a communication device for multiple modulation types |
| US8743848B2 (en) * | 2009-05-26 | 2014-06-03 | Broadcom Corporation | Hybrid location determination for wireless communication device |
| US9372266B2 (en) | 2009-09-10 | 2016-06-21 | Nextnav, Llc | Cell organization and transmission schemes in a wide area positioning system (WAPS) |
| US9291712B2 (en) | 2009-09-10 | 2016-03-22 | Nextnav, Llc | Cell organization and transmission schemes in a wide area positioning system (WAPS) |
| EP2732310B1 (en) * | 2011-06-28 | 2023-06-07 | Nextnav, LLC | Coding in a wide area positioning system (waps) |
| US9645249B2 (en) | 2011-06-28 | 2017-05-09 | Nextnav, Llc | Systems and methods for pseudo-random coding |
| JP2014529058A (en) * | 2011-06-28 | 2014-10-30 | ネクストナヴ，エルエルシー | Coding in Wide Area Positioning System (WAPS) |
| US9176217B2 (en) | 2011-08-02 | 2015-11-03 | Nextnav, Llc | Cell organization and transmission schemes in a wide area positioning system (WAPS) |
| AU2013271771B2 (en) | 2012-06-05 | 2016-07-28 | Nextnav, Llc | Systems and methods for location positioning of user device |
| US9286490B2 (en) | 2013-09-10 | 2016-03-15 | Nextnav, Llc | Systems and methods for providing conditional access to transmitted information |
| US9390279B2 (en) | 2012-09-11 | 2016-07-12 | Nextnav, Llc | Systems and methods for providing conditional access to transmitted information |
| US11026608B2 (en) * | 2017-01-09 | 2021-06-08 | Vox Biomedical Llc | Cerebral oximetry using time-gated direct sequence spread spectrum |
| CN112737619B (en) * | 2020-12-24 | 2022-05-13 | 重庆航天火箭电子技术有限公司 | Design method of full open loop PCM-DPSK-FM safety control receiver |
| CN113141195B (en) * | 2021-04-06 | 2022-08-23 | 重庆邮电大学 | Demodulation method for dispreading in direct sequence spread spectrum system and storage medium |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4253067A (en) * | 1978-12-11 | 1981-02-24 | Rockwell International Corporation | Baseband differentially phase encoded radio signal detector |
| US4730340A (en) * | 1980-10-31 | 1988-03-08 | Harris Corp. | Programmable time invariant coherent spread symbol correlator |
| US4443799A (en) * | 1981-07-02 | 1984-04-17 | Sperry Corporation | Spread spectrum radar |
| GB2233860B (en) * | 1989-07-13 | 1993-10-27 | Stc Plc | Communications systems |
| US5157686A (en) * | 1990-05-24 | 1992-10-20 | Cylink Corporation | Method and apparatus for the modulation of spread spectrum radio signals |
| US5103459B1 (en) * | 1990-06-25 | 1999-07-06 | Qualcomm Inc | System and method for generating signal waveforms in a cdma cellular telephone system |
| US5237587A (en) * | 1992-11-20 | 1993-08-17 | Magnavox Electronic Systems Company | Pseudo-noise modem and related digital correlation method |
| US5353301A (en) * | 1993-09-17 | 1994-10-04 | Motorola, Inc. | Method and apparatus for combining multipath spread-spectrum signals |
| JPH07170210A (en) * | 1993-12-16 | 1995-07-04 | Nec Corp | Spread spectrum modulating/demodulating method and modem using the same |
| US5592506A (en) * | 1994-10-17 | 1997-01-07 | Cylink Corporation | MSK spread-spectrum receiver which allows CDMA operations |
-
1994
- 1994-12-19 FR FR9415241A patent/FR2728415B1/en not_active Expired - Fee Related
-
1995
- 1995-11-21 US US08/560,211 patent/US5822363A/en not_active Expired - Lifetime
- 1995-12-12 CA CA002165034A patent/CA2165034A1/en not_active Abandoned
- 1995-12-14 DE DE69521671T patent/DE69521671T2/en not_active Expired - Lifetime
- 1995-12-14 EP EP95402818A patent/EP0718983B1/en not_active Expired - Lifetime
Also Published As
| Publication number | Publication date |
|---|---|
| FR2728415A1 (en) | 1996-06-21 |
| EP0718983A1 (en) | 1996-06-26 |
| US5822363A (en) | 1998-10-13 |
| DE69521671D1 (en) | 2001-08-16 |
| FR2728415B1 (en) | 1997-01-24 |
| EP0718983B1 (en) | 2001-07-11 |
| DE69521671T2 (en) | 2002-04-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US5822363A (en) | Transmission process having spectrum spread phase differential modulation adn demodulation using orthogonal pseudorandom sequences | |
| KR100782947B1 (en) | Pilot symbol assisted modulation and demodulation in wireless communication systems | |
| US7583582B2 (en) | M-ary orthogonal keying system | |
| US5276704A (en) | SAWC phase detection method and apparatus | |
| US5712869A (en) | Data transmitter and receiver of a spread spectrum communication system using a pilot channel | |
| US6404732B1 (en) | Digital modulation system using modified orthogonal codes to reduce autocorrelation | |
| CN105530214B (en) | Generate or receive the method and signal generator of spread-spectrum GMSK signal | |
| US10797920B1 (en) | High-entropy continuous phase modulation data transmitter | |
| EP2804327B1 (en) | Signal representing data, method and device for generating such signal and method and device for determining the represented data from such signal | |
| US7545845B2 (en) | Wireless communication system, wireless transmitter, wireless receiver, wireless communication method, wireless transmission method and wireless reception method | |
| USRE41931E1 (en) | Receiver module and receiver formed from several cascaded modules | |
| US5799035A (en) | Digital circuit for differential receiver for direct sequence spread spectrum signals | |
| CA2335020C (en) | One-bit correlator rake receiver | |
| US6198764B1 (en) | Method for the transfer of a digital data signal from a transmitter to a receiver | |
| KR20010022743A (en) | A Scheme for Spread Spectrum Multiple Access Coding | |
| JP3301724B2 (en) | Toothed Spectrum Communication System with Complementary Sequence Repetitive Modulation Comb | |
| AU712151B2 (en) | A method for direct sequence spreading of a data sequence (DSSS) | |
| JPH066324A (en) | Spread spectrum communication method | |
| Ozluturk et al. | Performances of DS/SSMA communications with MPSK signaling and complex signature sequences | |
| Elhakeem et al. | Performance of hybrid chirp/DS signals under Doppler and pulsed jamming | |
| JP3165233B2 (en) | Demodulator for spread spectrum communication | |
| JPH07221670A (en) | Saw spread spectrum demodulator | |
| JPH07321701A (en) | Spread spectrum communication equipment | |
| Cowl | Performance of code division multiple access on multipath channels-an exact analysis | |
| Frerking | Data Transmission |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| FZDE | Discontinued | ||
| FZDE | Discontinued |
Effective date: 20011212 |