WO2000072120A1 - Synthetiseur numerique a division coherente - Google Patents
Synthetiseur numerique a division coherente Download PDFInfo
- Publication number
- WO2000072120A1 WO2000072120A1 PCT/FR2000/001309 FR0001309W WO0072120A1 WO 2000072120 A1 WO2000072120 A1 WO 2000072120A1 FR 0001309 W FR0001309 W FR 0001309W WO 0072120 A1 WO0072120 A1 WO 0072120A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- divider
- modulo
- signal
- phase
- frequency
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/10—Frequency-modulated carrier systems, i.e. using frequency-shift keying
- H04L27/12—Modulator circuits; Transmitter circuits
- H04L27/122—Modulator circuits; Transmitter circuits using digital generation of carrier signals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/02—Digital function generators
- G06F1/03—Digital function generators working, at least partly, by table look-up
- G06F1/0321—Waveform generators, i.e. devices for generating periodical functions of time, e.g. direct digital synthesizers
- G06F1/0328—Waveform generators, i.e. devices for generating periodical functions of time, e.g. direct digital synthesizers in which the phase increment is adjustable, e.g. by using an adder-accumulator
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F7/00—Methods or arrangements for processing data by operating upon the order or content of the data handled
- G06F7/60—Methods or arrangements for performing computations using a digital non-denominational number representation, i.e. number representation without radix; Computing devices using combinations of denominational and non-denominational quantity representations, e.g. using difunction pulse trains, STEELE computers, phase computers
- G06F7/72—Methods or arrangements for performing computations using a digital non-denominational number representation, i.e. number representation without radix; Computing devices using combinations of denominational and non-denominational quantity representations, e.g. using difunction pulse trains, STEELE computers, phase computers using residue arithmetic
Definitions
- the present invention relates to a digital frequency synthesis device, and in particular to a device performing direct digital frequency synthesis.
- Digital synthesis is a frequency synthesis technique which consists in digitally calculating, at regular times, the value of the samples of the signal to be generated and in converting these samples using a digital analog converter to generate an analog signal.
- Analog digital converters are commonly designated by the abbreviation CNA or DAC according to English terminology.
- the frequency synthesizers obtained by this technique are very attractive as regards their volume, their weight and their energy consumption because they can benefit from a significant integration. Their other advantages include a very high resolution and very short switching times.
- phase law from frequency information.
- the phase law is developed by an accumulator.
- a table transforms the phase law into a digital sinusoidal signal.
- the digital sinusoidal signal is converted into a analog signal by a digital analog converter.
- the use of a digital to analog converter introduces faults which induce the creation of parasitic spectral components.
- the defects are linked on the one hand to the quantization of the signal and on the other hand to the non-linearities of the DAC.
- the DAC is limited to a number of bits NB which is generally lower than the number of bits N with which the table calculates the digital sinusoidal signal.
- the transition from N bits to NB bits generates a quantization error which results in the presence of parasitic lines on the signal at the output of the DAC,
- the transfer function of the DAC that is to say the output voltage as a function of the digital input words, is a step function.
- the differences in height between the steps of the staircase and the existence of irregular phenomena during the transition between steps induce non-linearities. These non-linearities are added to the above quantization error.
- a known method for improving the spectral purity of this type of synthesizer consists in dividing the signal coming from the digital to analog converter. The division is preceded by a filtering of the spectral components introduced by the sampling. This method has the major drawback of losing phase coherence due to the division operation.
- An object of the invention is to remedy this drawback. Thanks to the use of a coherent accumulator and a synchronizable divider, a device according to the invention improves the spectral purity, by dividing the output signal of the digital synthesizer, while preserving the phase coherence, by synchronizing the divider by a synchronization signal from the coherent accumulator. Thus, the device retains the phase coherence between bursts of the same frequency of a wave train, even if, between these bursts, the synthesizer emitted a burst whose frequency is different.
- a device for direct digital frequency synthesis comprises: a coherent accumulator, of modulo M, for generating a first phase law from a frequency control word, a table, addressed by a second phase law derived from the first phase law, for generating a digital sinusoidal signal, a digital analog converter for converting the digital sinusoidal signal into an analog sinusoidal signal, a filter for filtering the analog sinusoidal signal, and a divider, of a certain order less than M, for dividing the filtered signal, the divider having a synchronization input controlled by a synchronization pulse to resynchronize the signal after division, the synchronization pulse being produced from the phase law.
- the invention consists in using the most significant bit of the phase law from the coherent accumulator, having PI additional bits compared to the usual devices, to synchronize the divider.
- Which divider, of rank 2 P1 participates in the purification of the spectrum of the signal coming from the digital analog converter.
- the invention consists in separating the coherent accumulator into several coherent accumulators, of modulo Mi less than modulo M, so as to form a base of modulo in a numbering system with residue, the outputs of the accumulators representing the signal phase in the modulo base.
- a first division means operating according to this residue numbering system, makes it possible to adapt the phase law; the adaptation makes it possible to go from a first representation on the basis of the modulos of the accumulators to a second representation on a second modulo base compatible with the resolution of the table, in order to be able to address the table.
- a second division means operating according to this residue numbering system, makes it possible to adapt the phase law a second time; the adaptation makes it possible to pass from the second representation to a third representation on a third modulo basis.
- the third modulo base contains a single modulo equal to the division rank.
- a means extracts a most significant bit, from a coherent signal, to synchronize the divider.
- the most significant bit is extracted from the third representation of the phase law.
- the signal for synchronizing the divider is a pulse; this pulse is generated by the differentiation of the falling edge of the most significant bit.
- the most significant bit is commonly designated by the acronym MSB, abbreviation of the English terms Most Significant Bit.
- the pulse performs a periodic reset of the divider.
- the synchronization pulse ensures the correct placement of the phase cycle of the divider by imposing the instant of the phase zero of the divider.
- the phase cycle of the divider does not depend on the history of frequency changes as in the known devices; it depends on the phase of the most significant bit which is itself coherent, since it comes from the coherent accumulator.
- the cycle of the phase states of the divider is also established.
- the synchronization pulse occurs during the zero phase state of the divider; the synchronization pulse has no effect on the progress of the phase states of the divider. Consequently, the synchronization pulse can even exhibit a temporal variation without this having any effect on the coherence, or on the spectral purity, of the output signal of the divider; provided that this temporal variation remains less than the duration of the zero phase state of the divider.
- FIG. 2 a first embodiment of a device according to the invention
- FIG. 3 an embodiment of a coherent accumulator included in the device of FIG. 2,
- FIG. 4 a table of the states of certain signals of the device of FIG. 3 whose coherent accumulator has a given modulo M
- FIG. 6 a second embodiment of a device according to the invention.
- FIG. 7 an embodiment of a coherent accumulator included in the device of FIG. 6.
- Figure 1 shows a block diagram, a direct digital synthesizer according to the prior art.
- the synthesizer comprises an accumulator 1, a table 2, a digital analog converter 3, a filter 4 and a divider 5.
- the accumulator 1 draws up a phase law from frequency information.
- Table 2 transforms the phase law into a sinusoidal digital signal.
- Table 2 can be a memory area addressed as a function of the value of the phase present at the output of accumulator 1.
- the digital analog converter 3 converts the digital sinusoidal signal into an analog signal.
- the filter 4 filters the output signal from the digital analog converter 3 to filter, in particular, the spectral components introduced by the sampling.
- the divider 5 divides the filtered signal to improve its spectral purity.
- FIG. 2 is a block diagram of a first embodiment of a direct digital synthesizer according to the invention.
- the synthesizer comprises an accumulator 1, a table 2, a digital analog converter 3, a filter 4 and a divider 5.
- the accumulator 1 develops a first phase law from a frequency information; table 2 transforms a second phase law derived from the first phase law into a sinusoidal digital signal; the digital to analog converter 3 converts the sinusoidal digital signal into an analog signal; the filter 4 filters the output signal from the digital-analog converter 3 to filter, in particular, the spectral components introduced by the sampling; the divider 5 divides the filtered signal to improve its spectral purity.
- the synthesizer comprises truncation means 6, 7, 8 and means 9, 10 for generating a synchronization pulse and, the divider 5 is synchronizable.
- Accumulator 1 receives as input a word K which codes the desired frequency of the synthesizer output signal.
- the input bus of accumulator 1 is made up of P1 + P2 bits.
- the word K is coded on only P2 bits, the P2 least significant bits; the most significant Pi bits are set to zero. Consequently K verifies the relation: K ⁇ 2 P2 - 1.
- P2 is commonly chosen to be greater than PI.
- the output bus of the accumulator is made up of P1 + P2 bits.
- the P1 + P2 bits encode the phase of the signal to be generated.
- the phase law generated by the accumulator 1 corresponds to a signal whose frequency F is given by the expression:
- F 2 P1 X 2 P2 FH (1) in which F H is the frequency of a clock signal H.
- the accumulator 1 comprises a counter 11 and a multiplier 12.
- the counter 11 counts from 0 to 2 P1 ⁇ 2 P2 - 1 in steps of one at the rate of a clock H.
- the output of the counter 11, coded on P1 + P2 bits, constitutes one of the inputs of the multiplier 12.
- the second input of the multiplier 12 is constituted by the frequency setpoint K, coded on P2 bits.
- the multiplier 12 multiplies its two inputs between them, modulo 2 P1 ⁇ 2 P2 .
- the output of the multiplier 12 is the result RE of the multiplication; it is coded on P1 + P2 bits.
- the operation performed is expressed by the following relationship:
- the truncation means 6, 7, 8 adapt the number of bits supplied by the accumulator 1 on the one hand, to the capacity of the table 2 and, on the other hand, to the means 9, 10 for generating a pulse synchronization.
- the truncation means 6, 7, 8 can be broken down into a first means 6, a second means 7 and a third means 8.
- the first means 6 takes into account the PI + P2 bits, coding the phase, supplied by the accumulator 1.
- the first means 6 calculates the rest of the value of the modulo 2 phase P2 . This calculation amounts to ignoring the most significant PI bits. This calculation is equivalent to a multiplication by 2 P1 of the phase law generated by the accumulator 1; which amounts to multiplying the frequency by 2 P1 .
- This signal is coded on P2 bits.
- the second means 7 truncates from R bits the P2 bits supplied by the first means 6. This operation consists in not taking into account the R least significant bits.
- This truncation operation is conventional in known digital synthesizers. Indeed, the tables 2 generally have a number of address bits less than the number of bits of the accumulator 1. Reference is made to the address bits, because generally the tables are located in memories; a memory box being accessed by the address bits.
- the truncation does not modify the frequency Fa of the input signal.
- the frequency Fa of the signal at the output of the second means 7 is given by expression (3).
- the output of the second means 7 addresses the table 2 with a number of bits equal to P2-R.
- the truncation operation generates parasitic signals. To get rid of it, there is a technique which consists in adding a random signal to the output signal of accumulator 1.
- the third means 8 extracts the most significant bit MSB, from the value of the phase at the output of the accumulator 1.
- the output signal of the third means 8 is obtained by a truncation of P1 + P2- 1 bits of the P1 + P2 output bits of accumulator 1.
- the MSB of the output bus of accumulator 1 has a frequency Fs identical to that of the output signal of the synthesizer. In in reality, the MSB is not "exactly" periodic as illustrated by the description opposite figure 4; the MSB contains non-harmonic frequencies.
- the MSB presents on certain fronts a variable delay, between 0 and a period of the clock H, with respect to a signal of frequency Fs. This variation is deterministic; it is a function of the relationship between the K value of the frequency setpoint and the value of the modulo 2 P1 x 2 P2 . It can therefore be compensated; compensation is the subject of a variant of the system.
- Table 2 transforms the phase law into a sinusoidal digital signal.
- Table 2 converts the phase samples to amplitude samples. Given the symmetries of the sine function, table 2 may contain only a quarter of a period of the sine function.
- the two most significant bits of the input bus, at P2-R bits, are used to reconstruct the entire period according to a process known to those skilled in the art.
- the digital analog converter 3 converts the digital sinusoidal signal, output from table 1, into an analog signal.
- the filter 4 filters the output signal from the digital analog converter 3 to filter, in particular, the spectral components due to the sampling.
- the sinusoidal analog signal has a frequency Fa given by equation (3).
- the filter 4 introduces into the output signal a phase variation as a function of the frequency. This phase variation can be broken down into:
- the adder 13 compensates for the offset. This arrangement is the subject of a variant of the invention.
- the adder 13 can also be incorporated into the table 2 in the form of taking the offset into account in the stored values.
- the means 9, 10 for generating a synchronization pulse generate a synchronization pulse, from the output signal of the third truncation means 8.
- the synchronization pulse synchronizes the synchronizable divider 5.
- the means 9, 10 can be broken down into a first means 9 and a second means 10 for generating an impulse.
- the first means 9 preferably performs a differentiation of the output signal from the third truncation means 8.
- the first means 9 can consist of a JK flip-flop.
- the differentiation makes it possible to obtain a pulse, at the output of the first means 9, which is triggered by a front; in the example used, this is the falling edge of the output signal from the third truncation means 8.
- the falling edge corresponds to the instant t, synchronous with the clock H, where the content of the accumulator becomes greater than or equal to the modulo of the accumulator. In the example, this modulo is chosen equal to 2 P1 x 2 P2 . That is to say that at the previous clock stroke, ie at t-t H , the content of the accumulator has a value A.
- the first column of the table contains the values taken by the output of the counter 11. The output changes between 0 and 31 at the rate of the clock H.
- the second column contains the values taken by the output of the multiplier 12 as well as the state of the MSB of this output, this for two frequencies different.
- the third column contains the values taken by the output of the differentiator 9, this for each of the preceding frequencies.
- the output bit MSB of the multiplier 12 is coherent since it is reset synchronously to the counter 11. Given the manner in which the output pulse of the differentiator 9 is generated, this pulse is also coherent, as the MSB bit, and it has the same frequency as the MSB bit.
- the second means 10 has the function of compensating for certain delays.
- the compensation includes an integer number of clock ticks to which is added a fractional part of a clock tick which can be performed in analog.
- the delays mainly come from: - the pipeline differential between the synchronization channel, which includes the means 8 and 9, and the signal synthesis channel, which includes the truncation means 6 and 7, the adder 13, the table 2 and the DAC 3,
- Compensation for the delay is controlled via a setpoint. This can be developed, for example, from the following information: - the content of the accumulator when the MSB bit goes to 1,
- the synchronizable divider 5 preferably divides by a power of two the output signal of the filter 4, which has the frequency Fa.
- the signal on output Q is a division by two of the signal which feeds the clock input of the rocker.
- FIG. 5a is a chronogram of the phase states of the non-coherent divider of FIG. 1.
- FIG. 5b is a chronogram of the phase states of the coherent divider of FIG. 4.
- the input signal of the divider is assumed to be produced from a coherent accumulator.
- the signal 20 represents the clock input of the divider 5 when the frequency setpoint at the input of the device is equal to F,; signal 20 has frequency Fa,.
- the signal 21 represents the phase states of the divider 5.
- the divider 5 in this example performs a division by four.
- the divider 5 thus comprises four phase states, noted from 0 to 3, which follow one another at the rate of its clock input.
- the signal 22 represents the clock input of the divider 5 when the frequency setpoint at the input of the device is equal to F 2 ; signal 22 has the frequency Fa 2 .
- the signal 23 represents the clock input of the divider 5 when the frequency setpoint at the input of the device is again equal to F,.
- the signal 24 represents the time scale.
- the output signal Before time t 0 , the output signal has a frequency F, the input signal of the divider 5 has a frequency Fa, and it is represented by the signal 20.
- the frequency setpoint K changes so that the output signal has a frequency F 2 .
- the clock input of the divider is represented by the signal 22.
- the signal 20 is continued in dotted lines beyond t 0 .
- the phase states of the divider evolve continuously, but at the rate of the frequency Fa 2 after the instant t 0 .
- the frequency setpoint K changes so that the output signal again has a frequency F,.
- the clock input of the divider is represented by the signal 23.
- the signal 22 is continued in dotted lines beyond t,.
- the divider input signal is assumed to be produced from a coherent accumulator.
- the signal 23 takes the form that the signal 20 would have had if no frequency change had taken place; a phase break generally exists at time t, between signals 22 and 23.
- the phase states of the divider evolve continuously, but at the rate of the frequency Fa, after time t,.
- the comparison of the phase states 21 of the divider, after time t illustrates the loss of coherence after a frequency change.
- the phase states of the divider may be different from the phase states, shown in dotted lines below signal 20, that the divider would have had if the frequency had not been changed.
- the phase states of the divider take place at a different speed; this destroys the coherence because when returning to the initial frequency no information allows the divider to resume its phase cycle as if there had been no change in frequency.
- the usual dividers do not, in general, provide the signals describing the phase states of the divider; in the case of a divider by 2 P1 , no signal describes the Pi phase states. On the other hand, the usual dividers generally have a reset input.
- the synchronization pulse generated by the means 9, 10 for generating a synchronization pulse, controls the reset input of the divider, to impose the instant of the zero state of the synchronizable divider 5.
- the reset pulse makes it possible to correctly place the phase cycle of the divider 5 by imposing the instant of the phase zero of the divider 5.
- FIG. 5b represents the phase states of a divider by four , of a device according to the invention, to which a frequency change is applied and then a return to the initial frequency F,.
- the synchronization pulse falls during the zero phase instant of the divider; this impulse has no effect.
- the frequency is modified and goes from F, to F 2 .
- Changing the frequency changes the duration of the phase states of the divider.
- the signal at the input of the divider, of initial frequency Fa is continued in dotted lines during the time when the frequency is Fa 2 .
- the phase cycle of the divider 5 no longer depends on the history of frequency changes but on the phase of the MSB signal which is itself coherent; the MSB signal for generating the synchronization pulse.
- Figure 5b illustrates how the synthesizer output signal is kept consistent despite a frequency change.
- Signal 20 represents the clock input of divider 5 when the frequency setpoint at the input of the device is equal to F,; signal 20 has frequency
- the signal 21 represents the phase states of the divider 5.
- the divider 5 in this example performs a division by four.
- the divider 5 thus comprises four phase states, noted from 0 to 3, which follow one another at the rate of its clock input.
- the signal 22 represents the clock input of the divider 5 when the frequency setpoint at the input of the device is equal to F 2 ; signal 22 has the frequency Fa 2 given by equation (3).
- the signal 23 represents the clock input of the divider 5 when the frequency setpoint at the input of the device is again equal to F,.
- the signal 24 represents the time scale.
- the signal 25 is an illustration of the synchronization signal of the divider 5. In steady state, the synchronization signal 25 has the same frequency as the output signal of the synthesizer.
- the frequency of the synchronizing signal is Fa, / 2 2 to t 0, then F 2/2 2, between t 0 and t, and again Fa, / 2 2 beyond t.
- the synchronization pulse drops at the same time as the zero state of the divider 5; for example at times t 2 , t 3 , t 4 , t 6 , t 7 and t 8 .
- the output signal Before time t 0, the output signal has a frequency F i, the input signal from the divider 5 has a frequency Fa, and it is represented by the signal 20.
- the frequency setpoint K changes so that the output signal has a frequency F 2 .
- the clock input of the divider is represented by the signal 22.
- the signal 20 is continued in dotted lines beyond t 0 .
- the phase states 21 of the divider and the synchronization signal 25 are continued in dotted lines beyond t 0 as if the setpoint K had not been modified.
- the output signal of filter 4 changes frequency, the frequency becomes equal to Fa 2 .
- the phase states of the divider evolve continuously, but at the rate of the frequency Fa 2 after the instant t 0 .
- the frequency setpoint K changes so that the output signal again has a frequency F,.
- the clock input of the divider is represented by the signal 23.
- the signal 22 is continued in dotted lines beyond t,.
- the phase states 21 of the divider and the synchronization signal 25 are continued in dotted lines beyond t, as if the setpoint K had not been modified
- the output signal of the filter 4 changes frequency, the frequency again becomes equal to Fa,.
- the signal 23 being produced from the coherent accumulator 1, the signal 23 is coherent.
- there is generally a phase break in the output signal of the filter 4 so that the signal 23 is identical to the signal 20 followed by dotted lines.
- the phase states of the divider evolve continuously, but at the rate of the frequency Fa, after the instant t,.
- time t 5 there is a break in the phase states 21.
- phase states 21 evolve continuously regardless of the frequency changes.
- the phase states become again those which would have continued after t 0 if there had been no change in frequency at times t 0 and t,.
- the time between instants t, and t 5 corresponds to a transition period. The duration of this period depends on: - the frequency F,, - the instant t, where the frequency change occurs.
- This duration is less than a period of the frequency F,.
- this transitional period also exists during the first frequency change, after time t 0 . But in Figure 5b it does not appear. Beyond the instant t 5 the output signal of the divider is consistent with the output signal of the divider present before the instant t 0 and this despite the frequency change between the instants t 0 and t,. The frequency regime of the synthesizer output signal is established; the synchronization pulses arriving after the instant t 5 have no effect, they fall during the zero state of the divider 5.
- FIG. 6 gives a representation of such an alternative in a device according to the invention.
- the device comprises elements identical to the device described with reference to FIG. 2. These elements have the same reference number; they are not re-described.
- the device implements the residue algebra, this algebra is better known by the name RNS abbreviation of the Anglo-Saxon terms Residues Number System.
- RNS abbreviation of the Anglo-Saxon terms Residues Number System.
- the article by WA Chren “One-Hot Residue Coding for Low Delay-Power Product CMOS Design” with reference IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS-II: ANALOG AND DIGITAL SIGNAL PROCESSING, Vol.45, NO.3.MARCH 1998 gives a description of this system.
- the battery 1 of FIG. 2 is replaced, in this embodiment, by several small batteries 30 s .
- Each accumulator 30j has for modulo Mi; the accumulator 30 0 preferably having for modulo 4.2 P1 , in the case where the synchronizable divider 5 would have a rank of division, or order, equal to 2 P1 .
- the accumulator 30 0 is separated into two accumulators: a first modulo accumulator 4 and a second modulo accumulator equal to the division rank. If the modulo of this second accumulator is not prime with the other Mi modulos, it must be broken down according to its prime factors. And these different factors must be distributed, by combining them with the different Mi modulos, so as to respect the condition of obtaining primary accumulators between them.
- the division rank of the synchronizable divider is equal to 15.
- the different modulo 4.2 P1 , M1 Mm are all prime among themselves; that is, whatever two modulo Mj and Mk, from the list of modulo 4.2 P1 , M1, ..., Mm, the only common divisor for Mj and Mk is 1.
- the word control command Ki at the input of each accumulator 30j is equal to the remainder of the division of the frequency word K by the modulo of the corresponding accumulator. All the outputs of the 3O s accumulators represent the phase of the signal in the modulo base (4.2 P , M1, ..., Mm).
- phase states equal to the product of the modulo, that is to say (4 2 PI ) ⁇ (Ml) ⁇ ... (Mm).
- the number of phase states is equal to 2 P1 x 2 P2 .
- Table 2 generally has a phase resolution lower than that defined by all of the accumulators. In this case, it is necessary to carry out a truncation operation which consists in passing from a representation of the phase on the basis of the modulos (4.2 P1 , M1, ..., Mm) to a representation on the basis of the modulos ( 4.2 P1 , M1, ..., Ms) with s ⁇ m.
- the device comprises a first dividing block 31 for performing this truncation.
- the output signal of the first division block 31 is the integer part of the quotient represented on the remaining modulos. This imposes a new restriction on the choice of modulos: each modulo from M (s + 1) to Mm must present an inverse for each of the remaining modulos from 4.2 P1 to Ms, to make the division operation possible.
- the output of the first division block 31 is represented on the basis of the modulos (4.2 P1 , M1, ..., Ms).
- the device includes a second division block 32.
- This second block 32 divides by the product 4 ⁇ Ml ⁇ M2 ⁇ ... ⁇ Ms in order to be able to represent the signal only on modulo 2 P1 .
- the output signal of this division is therefore a signal on P1 bits.
- a third truncation means 8 extracts the MSB bit from the output of the second block 32. This extraction has been described with reference to FIG. 2.
- a first means 9 for generating a synchronization pulse preferably performs: a differentiation of the output signal from the third truncation means 8 as in the first embodiment of the invention.
- the second means 10 for generating a synchronization pulse has the function of compensating for certain delays as in the first embodiment of the invention.
- the device includes a multiplier operator 33.
- This operator 33 multiplies the signal frequency by 2 P1 .
- the multiplication relates only to the first modulo (4 ⁇ 2 pl ).
- the output signal of the operator 33 is the remainder modulo 4 of the input signal of the operator 33.
- the signal, at the input of the table 2 is represented on the basis of the modulos (4, M1,. .., Ms).
- the modulo 4 on this base saves material for coding the sinus. Indeed, it suffices to code a quarter of a sinus period and to use the symmetries to reconstruct the whole of the sinusoid. By choosing to use the modulo 4 for this purpose and by choosing to divide by a power of two at the output of the DAC, it becomes necessary to combine these two factors, 4 and 2 P1 , in the same accumulator.
- modulo 4 is not prime with a modulo 2 P1 .
- the grouping of modulo 4 and 2 P in the same accumulator leads to the embodiment described above.
- Other embodiments are possible; for example modes which do not use the symmetries of the sine, or for example modes in which the order of the divisor is an odd number.
- FIG. 7 illustrates an embodiment of a coherent accumulator 30 s of modulo Mi, of the device of FIG. 6.
- the accumulator 30j comprises a counter 33 s and a multiplier 34j.
- the counter 33j counts from 0 to Mi-1 in steps of one at the rate of a clock H.
- the output Ci of the counter 33j constitutes one of the inputs of the multiplier 34j.
- the second input of the multiplier 34j is constituted by the frequency setpoint Ki.
- the multiplier 34j multiplies its two inputs between them, modulo Mi.
- the output of the multiplier 34j is the result Ri of the multiplication.
- the operation performed is expressed by the following relationship:
- the output of the multiplier 34 ( corresponds to the output of the accumulator 30 *.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Stabilization Of Oscillater, Synchronisation, Frequency Synthesizers (AREA)
Abstract
Description
Claims
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2000620447A JP2003500958A (ja) | 1999-05-21 | 2000-05-16 | コヒーレント分割によるデジタルシンセサイザ |
CA002374707A CA2374707A1 (fr) | 1999-05-21 | 2000-05-16 | Synthetiseur numerique a division coherente |
US09/926,587 US6597208B1 (en) | 1999-05-21 | 2000-05-16 | Digital synthesizer with coherent division |
EP00927344A EP1190295A1 (fr) | 1999-05-21 | 2000-05-16 | Synthetiseur numerique a division coherente |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
FR9906514A FR2793972B1 (fr) | 1999-05-21 | 1999-05-21 | Synthetiseur numerique a division coherente |
FR99/06514 | 1999-05-21 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2000072120A1 true WO2000072120A1 (fr) | 2000-11-30 |
Family
ID=9545898
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/FR2000/001309 WO2000072120A1 (fr) | 1999-05-21 | 2000-05-16 | Synthetiseur numerique a division coherente |
Country Status (6)
Country | Link |
---|---|
US (1) | US6597208B1 (fr) |
EP (1) | EP1190295A1 (fr) |
JP (1) | JP2003500958A (fr) |
CA (1) | CA2374707A1 (fr) |
FR (1) | FR2793972B1 (fr) |
WO (1) | WO2000072120A1 (fr) |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7343387B2 (en) * | 2002-02-26 | 2008-03-11 | Teradyne, Inc. | Algorithm for configuring clocking system |
US7302237B2 (en) * | 2002-07-23 | 2007-11-27 | Mercury Computer Systems, Inc. | Wideband signal generators, measurement devices, methods of signal generation, and methods of signal analysis |
FR2880219B1 (fr) * | 2004-12-23 | 2007-02-23 | Thales Sa | Procede et systeme de radiocommunication numerique, notamment pour les stations sol mobiles |
US7616141B2 (en) * | 2004-12-23 | 2009-11-10 | Jianzhong Chen | Digital-to-analog converter |
WO2012093386A2 (fr) | 2011-01-03 | 2012-07-12 | Savant Technologies Ltd | Procédé et système de synthèse de signaux |
CN113985761B (zh) * | 2021-09-30 | 2024-02-09 | 歌尔股份有限公司 | 基于fpga的脉冲产生控制方法、终端设备及可读存储介质 |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0469233A2 (fr) * | 1990-08-03 | 1992-02-05 | Hewlett-Packard Company | Synthétiseur numérique avec mémoire de phase |
US5757239A (en) * | 1995-03-16 | 1998-05-26 | Qualcomm Incorporated | Direct digital synthesizer driven phase lock loop frequency synthesizer with clean up phase lock loop |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
FR2748872B1 (fr) | 1990-08-21 | 1998-11-27 | Thomson Trt Defense | Synthetiseur de frequence a boucle a verrouillage de phase a division fractionnaire multiple |
FR2716312B1 (fr) | 1994-02-11 | 1996-03-22 | Thomson Csf | Dispositif de modulation de phase continue par synthétiseur de fréquences à boucle à verrouillage de phase. |
GB9616537D0 (en) * | 1996-08-06 | 1996-09-25 | Digi Media Vision Ltd | Digital synthesiser |
FR2763196B1 (fr) | 1997-05-07 | 1999-07-30 | Thomson Csf | Synthetiseur de frequence coherent a boucle de phase et pas fractionnaires |
FR2765419B1 (fr) | 1997-06-27 | 1999-09-17 | Thomson Csf | Dispositif de generation de signaux analogiques a partir de convertisseurs analogique-numerique, notamment pour la synthese numerique directe |
US6307441B1 (en) * | 1997-12-31 | 2001-10-23 | Texas Instruments Incorporated | Shape modulation transmit loop with digital frequency control and method for same |
FR2780831B1 (fr) | 1998-07-03 | 2000-09-29 | Thomson Csf | Synthetiseur numerique de signaux |
FR2794309B1 (fr) | 1999-05-28 | 2001-08-31 | Thomson Csf | Dispositif compensateur de la non-linearite d'un convertisseur analogique-numerique |
-
1999
- 1999-05-21 FR FR9906514A patent/FR2793972B1/fr not_active Expired - Fee Related
-
2000
- 2000-05-16 WO PCT/FR2000/001309 patent/WO2000072120A1/fr active Application Filing
- 2000-05-16 CA CA002374707A patent/CA2374707A1/fr not_active Abandoned
- 2000-05-16 JP JP2000620447A patent/JP2003500958A/ja not_active Withdrawn
- 2000-05-16 EP EP00927344A patent/EP1190295A1/fr not_active Withdrawn
- 2000-05-16 US US09/926,587 patent/US6597208B1/en not_active Expired - Fee Related
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0469233A2 (fr) * | 1990-08-03 | 1992-02-05 | Hewlett-Packard Company | Synthétiseur numérique avec mémoire de phase |
US5757239A (en) * | 1995-03-16 | 1998-05-26 | Qualcomm Incorporated | Direct digital synthesizer driven phase lock loop frequency synthesizer with clean up phase lock loop |
Also Published As
Publication number | Publication date |
---|---|
FR2793972A1 (fr) | 2000-11-24 |
JP2003500958A (ja) | 2003-01-07 |
FR2793972B1 (fr) | 2001-08-10 |
CA2374707A1 (fr) | 2000-11-30 |
EP1190295A1 (fr) | 2002-03-27 |
US6597208B1 (en) | 2003-07-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP1956714A1 (fr) | Procédé d'ajout d'un bruit aléatoire dans un circuit convertisseur temps-numérique et circuits pour mettre en oeuvre le procédé | |
FR2489554A1 (fr) | Circuit de traitement numerique en arithmetique distribuee utilisant des multiplexeurs a l'entree d'une memoire | |
FR2666463A1 (fr) | Procede et appareil de synthese a n fractionnaire avec reduction de l'erreur residuelle et application a un radioemetteur. | |
FR2486334A1 (fr) | Procede pour numeriser un signal video discret dans le temps avec utilisation d'une transformation d'image | |
KR20070025937A (ko) | 샘플링 레이트 변환 방법 및 그 회로 | |
FR2471094A1 (fr) | Circuit interpolateur augmentant le debit de mots d'un signal numerique du type utilise dans les systemes telephoniques a commutation numerique et joncteur de ligne equipe d'un tel circuit | |
FR2587819A1 (fr) | Dispositif de calcul d'une transformee de fourier discrete, glissante et non recursive, et son application a un systeme radar | |
FR2545673A1 (fr) | Systeme de stockage d'information respectivement de transmission pour television en couleurs a codage en multiplex a division dans le temps et generateur et recepteur d'information | |
EP0147307B1 (fr) | Synthétiseur de fréquences à division fractionnaire, à faible gigue de phase et utilisation de ce synthétiseur | |
FR2588680A1 (fr) | Dispositif de calcul d'une transformee de fourier discrete, et son application a la compression d'impulsion dans un systeme radar | |
WO2000072120A1 (fr) | Synthetiseur numerique a division coherente | |
EP0753941B1 (fr) | Synthétiseur de fréquences | |
CA2276850C (fr) | Synthetiseur numerique de signaux | |
FR2666184A1 (fr) | Horloge a division de frequence fractionnaire et asservissement de cette horloge. | |
CA1308449C (fr) | Dispositif de retard d'au moins un train de donnees binaires a haut debit | |
EP0298804B1 (fr) | Procédé et dispositif de synthèse numérique d'un signal d'horloge | |
FR2613559A1 (fr) | Appareil de modulation par impulsions codees differentielle et son procede de fonctionnement | |
EP0476592A2 (fr) | Générateur d'adresses pour la mémoire de données d'un processeur | |
EP0053064A1 (fr) | Système de transmission numérique à codage adaptatif d'informations analogiques échantillonnées et transformées par transformation orthogonale | |
EP3048730A1 (fr) | Dispositif de synthèse de fréquence à boucle de rétroaction | |
CA2033156C (fr) | Dispositif d'insertion d'elements binaires d'information dans une structure de trame determinee | |
JPH1079650A (ja) | クロック発生回路 | |
EP0175623A1 (fr) | Dispositif de traitement en temps réel de signal numérique par convolution | |
EP0643502B1 (fr) | Dispositif de filtrage de gigue de déjustification positive d'un train numérique et application au filtrage de gigue de déjustification positive et positive-négative d'un train numérique | |
FR2538647A1 (fr) | Circuit de demultiplexage d'un signal numerique en trames et a haut debit |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): CA JP US |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
DFPE | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101) | ||
ENP | Entry into the national phase |
Ref document number: 2374707 Country of ref document: CA Ref country code: CA Ref document number: 2374707 Kind code of ref document: A Format of ref document f/p: F |
|
ENP | Entry into the national phase |
Ref country code: JP Ref document number: 2000 620447 Kind code of ref document: A Format of ref document f/p: F |
|
WWE | Wipo information: entry into national phase |
Ref document number: 09926587 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2000927344 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 2000927344 Country of ref document: EP |