Classifications

• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
  • H04L27/00—Modulated-carrier systems
  • H04L27/18—Phase-modulated carrier systems, i.e. using phase-shift keying
  • H04L27/20—Modulator circuits; Transmitter circuits
  • H04L27/2003—Modulator circuits; Transmitter circuits for continuous phase modulation

Definitions

• the invention relates to a method for the modulation of digital information. It concerns, for example, continuous phase modulations
• CCM CCM
• GMSK Gaussian Minimum Shift Keying
• the expression "interference between symbols” is used with reference to the phenomenon known to those skilled in the art describing the fact that a signal sample received at a time t does not depend on a single symbol. issued but also other neighboring symbols.
• state of a symbol is used to designate the representation in the complex plane of said symbol.
• tellation of a complex sequence of symbols is used to designate the representation of said symbols in the complex plane.
• adapted filter known to those skilled in the art, is used in the rest of the text to denote the optimal linear filter implemented by a receiver to maximize the signal-to-noise ratio.
• the use of constant amplitude modulations for example phase or frequency modulations, is preferred since the latter make it possible to maximize the range of said systems, the transmission power being constant and maximum in this case.
• This is for example the case of the GSM cellular system (Global System for Mobile Communications).
• the current trend of communications systems is to increase the payload without increasing the bandwidth of the transmitted signal. This can be done by increasing the number of information bits conveyed by a transmitted symbol.
• a transmitted symbol carries only one bit of information
• a modulation with 4 or 8 states a transmitted symbol carries the information of 2 or 3 bits respectively.
• continuous phase modulation methods such as modulation. GMSK.
• the aforementioned methods although effective, have the disadvantage of introducing significant inter-symbol interference into the received signal.
• the symbol received at time t is disturbed at least by its neighbors emitted at times t-1 and t + 1, which makes it more difficult for a receiver to decide on the initial state of said symbol.
• a symbol can take 8 different states. The receiver will have to make a decision among these 8 possibilities to discriminate the corresponding bit. Interference between symbols is even more troublesome and the receiver will be more complex as the number of possible states for a symbol is high.
• the methods known to those skilled in the art for demodulating a signal transmitted according to a continuous phase modulation of the GMSK type have the disadvantage of being complex because they often implement an equalizer as well as an estimator of the maximum of likelihood known as the MLSE (Maximum Likelihood Sequence Estimator) implemented, for example, via a Viterbi algorithm.
• the subject of the invention is a method for constant amplitude modulation and continuous phase digital data, said data being complex symbols conveying at least one information bit and whose theoretical constellation comprises a number of states N to less than two, comprising at least the following steps:
• a coefficient combining step w (i) which provides at its output a transformed sequence b (n) obtained by combining said symbols delayed between them,> modulating the sequence of symbols b (n) by a continuous phase modulation method composed of an impulse response filter h (t) and a frequency modulator for obtaining a modulated symbol sequence, said method being characterized in that said coefficients w (i) are obtained by a calculation step whose function is the minimization of the signal-to-noise ratio between said modulated symbols and the symbols of said theoretical constellation.
• the combining step is performed by a finite impulse response (RIF) filter with coefficients w (i), with i varying from 0 to K.
• the combining step is implemented by a read-only memory in which the values of the transformed sequence b (n) are precalculated from the possible values of the sequence a (n).
• the impulse response h (t) of the filter used to carry out the continuous phase modulation is a Gaussian function of duration T and standard deviation ⁇ .
• the continuous phase modulation used is a Gaussian Minimum Shift Keying (GMSK) modulation.
• GMSK Gaussian Minimum Shift Keying
• the invention also relates to a transmitter used to transmit modulated data via a continuous phase modulation and constant amplitude modulation method, the data being in the form of a sequence of complex symbols conveying at least one information bit and whose theoretical constellation has a number of states N at least equal to two, said transmitter comprising at least the following elements:
• a delay line comprising K cells producing a delayed symbol sequence
• a coefficient combining system w (i) adapted to execute a combination of the delayed symbol sequence to produce a new symbol sequence b (n),
• a continuous phase modulation system comprising a filter and a frequency modulator and modulating the sequence of symbols b (n) in order to obtain a sequence of modulated symbols
• said transmitter being characterized in that said coefficients w (i) are obtained by a calculation step whose function is the minimization of the signal-to-noise ratio between said modulated symbols and the symbols constituting said theoretical constellation.
• FIG. 1 an example of a GMSK modulator
• FIGS. 2a, 2b and 2c an example of different characteristics of a GMSK modulation, the impulse response of a filter h (t), the integral function f (t) of h (t) and the variation of the phase the modulated signal over time
• FIG. 3 an example of a device for differential demodulation of a signal modulated by a GMSK modulation according to the prior art
• FIGS. 4a, 4b and 4c several examples of constellations obtained by differential demodulation of a signal modulated by a GMSK modulation
• FIG. 5 an example of a device for coherent demodulation of a signal modulated by a GMSK modulation according to the prior art
• FIGS. 6a and 6b several examples of constellations obtained by coherent demodulation of a signal modulated by a modulation GMSK according to the prior art
• FIG. 7 an example of a device implementing a modulation method according to the invention
• FIG. 8 an exemplary embodiment of a method making it possible to determine the parameters of the modulation defined by the method according to the invention
• FIGS. 9a and 9b an example of constellations obtained by differential demodulation of a signal modulated via the method according to the invention
• FIG. 10 an example of constellations obtained by coherent demodulation of a signal modulated via the method according to the invention
• FIG. 11 an example of a frequency spectrum of the signal emitted via the method according to the invention.
• CPM modulations are a family of modulations commonly used to transmit digital data especially in the context of wireless communications. In contrast to other modulation methods for which the phase of the modulated signal is subjected to steep transitions, the CPM modulations are used to modulate the phase of the transmitted symbols continuously.
• Figure 1 illustrates a continuous phase modulation method.
• the binary data to be transmitted are first converted into symbols a (n) via a mapping operation, 1, or mapping in English. This step makes it possible to associate one or more bits with a symbol to be transmitted.
• the symbols obtained can be characterized by the representation in the complex plane of their constellation. If the symbol a (n) conveys an information bit, its constellation will comprise two states, if it conveys two bits of information, its constellation will comprise four states, and in general, if said vehicle symbol n bits of information its constellation will have 2 n states.
• the symbols a (n) are transmitted via a series of Dirac pulses spaced apart from each other by a duration equal to T, where T is the duration of a symbol.
• the Dirac pulse at time n is represented by the term a (n) ⁇ (t-nT).
• the sequence of Dirac pulses is then filtered by an impulse response filter defined by the function h (t), 2, and then transmitted to a frequency modulator 3.
• the pulse h (t) is defined over a duration LT, where L is generally an integer greater than or equal to 1.
• the values of this pulse are zero outside the interval [0, LT] and its integral between 0 and LT is 1.
• the frequency modulator 3 outputs a signal S (t) which can be represented by the following formula, where F 0 is the transmission frequency:
• the function f (t) represents the integral of h (t), equal to 0 for t negative and 1 if t is greater than LT.
• FIG. 2c represents the variation of the phase of the signal S (t) over time which is continuous, which contributes to concentrating the frequency spectrum of the signal emitted around the carrier.
• FIG. 3 schematizes a differential demodulation method known to those skilled in the art. Said method makes it possible to decide which are the values of the symbols a (n) which were most probably emitted during the reception of the signal S (t). Said received signal S (t), to which is added noise related to the transmission is filtered by a filter adapted to the signal 10. The role of this filter is to maximize the signal-to-noise ratio at the time of the decision by the receiver . The result of this operation X (t) is then transmitted to a module January 1 which applies a delay equal to the duration of the symbol T and the delayed signal is transmitted to a module 12 which calculates the conjugate of the complex signal and outputs the result X * (tT).
• a decision-making module 14 coupled to a synchronization module of the signal at the symbol rate 15 makes it possible to determine the value of the decided symbol D (nT) received at the instant nT after sampling of the signal.
• FIGS. 4a, 4b and 4c represent the constellations of the signals D (nT) obtained after demodulation of a modulated signal by a modulation GMSK via the method described in FIG. 3 for transmitted symbols a (n) comprising respectively 2, 4 and 8 states.
• the constellation of said symbols D (nT) should have only 2, 4 or 8 regularly distributed points on a circle centered at the origin. Except practically this is not the case, even in the simplest case of symbols a (n) having 2 states, the constellation obtained in reception has 6 distinct points (of which 2 are in fact double which brings the number of states to 8 in reality). For cases corresponding to 4 or 8 states, the decision to determine which symbol is issued is difficult to make.
• FIG. 5 schematizes a coherent demodulation method known to those skilled in the art. This method requires knowing the reference phase of the carrier, to ensure that a signal transmitted with a given phase is well seen by the receiver with the same phase. It therefore requires, in place of the delay line 11 of FIG. 3, the use of a phase estimation system 20 which uses, for example, previously known symbols inserted in the modulated signal. The rest of the diagram is similar to the case of the differential demodulation shown in FIG. 3. A demodulated signal D (nT) is obtained at the output.
• FIGS. 6a and 6b represent the constellations of the demodulated signal according to the method shown schematically in FIG. 5.
• FIG. 6a schematizes the case of transmitted symbols comprising 4 states
• FIG. 6b corresponds to the 8-state case.
• the constellations obtained comprise a large number of states much greater than 4 (respectively 8), the decision on the received symbols is complex to implement.
• FIG. 7 schematizes an exemplary embodiment of the method according to the invention in which the GMSK modulation method described in FIG. 1 is taken up and modified as follows:
• the symbols a (n) are stored in a delay line comprising K cells, K being a non-zero positive integer.
• K being a non-zero positive integer.
• the values available in the delay line are a (n-i), for i varying from 0 to K.
• the K + 1 available values are provided at the input of a combination system 31 which outputs samples b (n) obtained by a suitable combination of the values a (ni), i varying from 0 to K. samples b (n) are produced at the symbol rate, and replace the symbols a (n) in the original schema. They are then filtered by the filter h (t), 32, and modulated by the frequency modulator 33.
• the combination system 31 is a symmetrical finite impulse response filter (RIF).
• the value of b (n) is then obtained by linear combination of the values a (n) according to the following formula: where the values w (i) are the coefficients of said filter, p is an integer taking values between 0 and K / 2 and K is the value of the previously defined delay.
• the combination system 31 is a read-only memory containing a limited set of values b (n). At each sequence [a (n) ... a (nK)] is associated a pre-calculated value of b (n) stored in said ROM.
• This variant embodiment has the advantage of not performing the filtering calculation in real time.
• any other sequence of values can be used for the initialization of the coefficients of the filter 40.
• a randomly generated sequence of symbols D (n) is provided at the input of said filter 40. Said symbols are themselves created at from a binary sequence to which a correspondence rule has been applied as described in FIG. 1 in order to generate symbols conveying one, two, three or more generally N bits.
• a reference constellation 41 is associated with a number of states defined by the number of information bits conveyed per symbol. By way of example, for symbols carrying two information bits, the reference constellation will comprise four states for the symbols transmitted at the instant k and four other states out of phase by ⁇ / 4 for the symbols transmitted at the instant k. 1.
• the sequence of symbols D (n) is thus filtered via the filter 40 and then transmitted to a modulator 42 which may be, for example a GMSK modulator defined by its impulse response h (t) of length T and standard deviation ⁇ , such as previously described in Figure 2a.
• the generated signal S (t) is then supplied to a module 43 for calculating the frequency spectrum of said signal and then to a module 44 for calculating the matched filter which outputs the impulse response of said filter.
• This is then implemented by the device 45 which filters the signal S (t) to obtain a received demodulated sequence R.
• a step 46 then makes it possible to calculate the constellation of the received sequence R which will be compared with the reference constellation 41. This comparison is made via step 47 through, for example, a mean squared error calculation E between the symbols of constellation 41 and those of constellation 47:
• This operation is performed for each coefficient w (i) independently and is repeated until the value of E has reached a minimum value E min which indicates that the constellation 46 of the received signal is in accordance with the reference constellation 41 where, as far as possible, the nearest possible.
• the function of calculating the coefficients w (i) therefore has the object of minimizing the signal-to-noise ratio between the symbols generated by the method according to the invention and the symbols of the reference constellation 41.
• the modulation method according to the invention described in FIGS. 7 and 8 makes it possible to use receivers known in the prior art such as those described in FIGS. 3 and 5 for demodulating the received signal.
• FIGS. 9a and 9b show the final constellations obtained after differential demodulation of a modulated signal via the method according to the invention. Cases corresponding to modulations with 4 and 8 states respectively are considered. The decision step making it possible to retrieve the transmitted symbols is clearly facilitated since the number of states of the constellation is considerably reduced with respect to FIGS. 4b and 4c. Similarly, the results concerning a coherent demodulation of a modulated signal via the method according to the invention are shown in FIG. 10 for the case corresponding to 8 possible states for an initially transmitted symbol.
• the intrinsic signal-to-noise ratio of the constellation is of the order of 20 dB, which is all the more necessary as the point clouds are oriented in such a way that they minimize the error on the phase, which has a certain advantage in differential demodulation.
• the frequency spectrum of the emitted signal is also shown in FIG. 11, it has the advantage of not having net side lobes, which would be the case with a conventional continuous phase modulation method.
• the Gaussian frequency pulse used to realize the GMSK modulator has the advantage of having a frequency spectrum which decreases rapidly and continuously which minimizes the influence on the adjacent channels.
• the method and the system according to the invention have the advantage of virtually eliminating any interference between symbols in reception of the modulated signal.
• This aspect is important because it makes it possible to use simplified receivers compared to the systems used to demodulate a signal modulated by a continuous phase modulation of the GMSK type. Indeed, a technique known to those skilled in the art and usually used is to apply a maximum likelihood estimation algorithm to determine which symbol has been issued, this type of algorithm is complex to implement, the method according to the invention. the invention makes it possible to avoid it.

Landscapes

• Engineering & Computer Science (AREA)
• Computer Networks & Wireless Communication (AREA)
• Signal Processing (AREA)
• Digital Transmission Methods That Use Modulated Carrier Waves (AREA)

Priority Applications (2)

Applications Claiming Priority (2)

Publications (2)

Family

ID=41061256

Family Applications (1)

Country Status (4)

Cited By (1)

Families Citing this family (1)

Citations (3)

• 2008
  • 2008-11-21 FR FR0806554A patent/FR2938988B1/fr not_active Expired - Fee Related
• 2009
  • 2009-11-17 ES ES09752190T patent/ES2392608T3/es active Active
  • 2009-11-17 WO PCT/EP2009/065275 patent/WO2010057871A2/fr active Application Filing
  • 2009-11-17 EP EP09752190A patent/EP2351305B1/de active Active

Patent Citations (3)

Also Published As

Similar Documents

Legal Events