Phase Synchronizing Circuit

The subject matter of this application is related to my copending application Ser. No. 779,853, filed Mar. 21, 1977.

The present invention relates to a phase synchronizing circuit for use in a multi-level, multi-phase, superposition-modulated signal transmission system in which the modulated carrier waves are mutually synchronized and, more particularly, to a phase synchronizing circuit of this type having a circuit for compensating the phase and amplitude variation of the phase error signal caused by the superposition-modulation.

Now that digital transmission has been in practical use, efforts have been made recently to enhance the transmission capacity of a carrier wave and also to enhance the efficiency with which the frequency band is used. One of the proposals made as a result of such efforts is the multi-level, multi-phase, superposition-modulated wave transmission system in which a first and a second modulated carrier waves are synthesized under synchronized state to constitute a composite carrier wave for transmission.

An ordinary multi-level, multi-phase, superposition-demodulator apparatus for such transmission system includes the so-called re-modulation type demodulator synchronized with the incoming transmitted carrier wave, and a vector subtractor circuit coupled to the demodulator.

The demodulator is adapted to demodulate at its first demodulator section the first modulated carrier wave contained in the modulated carrier wave. Also, the second modulated carrier wave is obtained by deriving the remodulated first modulated carrier wave from the first modulator section and subjecting it to vector subtraction. This second modulated carrier wave is demodulated at the second demodulator section similar to the aforementioned first demodulator circuit.

This demodulator circuit, that is, the so-called re-modulation type phase synchronizing circuit functions as a phase synchronizing circuit for the first modulated carrier wave component. However, since the second modulated wave component behaves as an interference wave, the reproduced carrier wave tends to be associated with jitters. This adversely affects the quality of the demodulation output.

It is therefore an object of the present invention to provide an excellent phase synchronizing circuit that eliminates the aforementioned disadventage.

According to one feature of the present invention, there is provided a phase synchronizing circuit for use in a multi-level, multi-phase, superposition-modulated signal transmission system in which the transmission carrier wave is formed by combining a first modulated carrier wave and a second modulated carrier wave which are synchronized with each other, comprising a voltage-controlled oscillator having frequencies varying in response to a phase error signal, phase-detector means for detecting said superposition-modulated signal with respect to the output of said oscillator, a logic circuit for processing the outputs of said phase-detector means to produce a signal representing the phase difference between said first and second modulated carrier waves, and means responsive to the output of said logic circuit for controlling at least one of the phase variation and the amplitude variation in said phase error signal.

Now the present invention will be described in more detail with reference to the accompanying drawings, in which:

FIG. 1 is a block diagram of one example of a known modulator for a 16-level superposition-modulation transmission system;

FIG. 2 is a vector diagram for the description of the modulator shown in FIG. 1;

FIG. 3 is a block diagram of one example of a known demodulator for a similar superposition-modulation transmission system;

FIG. 4 shows output beat waveforms for the description of a phase comparator circuit employed in the demodulator of FIG. 3 assuming that the remodulator therein is interrupted;

FIG. 5 shows output waveforms for the phase comparator circuit employed in the demodulator of FIG. 3,

FIGS. 6, 9 and 11 show preferred embodiments of the phase synchronizing circuit according to the present invention,

FIG. 7 is a more detailed circuit diagram of one example of an A-D converter and a logic circuit employed in the embodiment of FIG. 6,

FIG. 8 shows waveforms at various portions in the A-D converter in FIG. 7,

FIG. 10 shows a waveform at point H in FIG. 9, and

FIG. 12 is a vector diagram for explaining the operation of the circuit shown in FIG. 11.

Referring now to the drawings showing one example of the parallel-connection type modulator for the 16-level superposition-modulation system, reference numeral 1 designates a carrier wave oscillator; 2, a first 4-phase phase modulator; 3, a second 4-phase phase modulator; 4, a phase shifter; 5, an attenuator; and 6, an output terminal. In the phase modulator 2, a carrier wave is modulated by two series of first digital signals S.sub.11 and S.sub.12, the modulated signal is hereinafter called a first modulated signal or a first modulated carrier wave, and a demodulated signal for this modulated signal is called a first demodulated signal. Likewise, in the modulator 3, a carrier wave is modulated by two series of second digital signals S.sub.21 and S.sub.22, the modulated signal is hereinafter called a second modulated signal or a second modulated wave, and a demodulated signal for this modulated signal is called a second demodulated signal. The phase shifter 4 is provided for the purpose of adjusting the phases of the first and second modulated signals, and here it is assumed that they are adjusted to be in-phase. The attenuator 5 is provided for the purpose of adjusting the level difference between the first and second modulated signals, and here it is assumed that the first modulated signal level is equal to .alpha. and the second modulated signal level is equal to .beta.. FIG. 2 is a vector diagram of the modulated signals at the output terminal 6 (FIG. 1). Reference numeral 7 designates a first modulated signal, numeral 8 designates a second modulated signal, and a combined vector for these signals appears at the output terminal 6.

A conventional demodulator device for demodulating the thus superposition-modulated signal is shown in FIG. 3. In this figure, reference numeral 10 designates a 4-phase phase demodulator circuit employing a well-known remodulation-comparison type phase synchronizing circuit; 18, a vector subtractor circuit; and 19, a demodulator circuit similar to the demodulator circuit 10. The demodulator circuit, that is, phase synchronizing circuit 10 is composed of a 4-phase phase detector (P. DET) 11, discriminator-shaper circuits 12 and 13, a remodulator 14, a phase comparator circuit (P. COM) 15, a low-pass filter (LPF) 16 and a voltage-controlled oscillator (VCO) 17. Since the operation of this device is well-known, it will be explained only briefly. In the discriminator-shaper circuits 12 and 13, the first demodulated signal S.sub.11 and S.sub.12 is discriminated and reproduced. On the other hand, the second demodulated signal is obtained in the following manner. In the re-modulator 14, the reproduced carrier wave from the voltage-controlled oscillator 17 is modulated by the first demodulated signals S.sub.11 and S.sub.12 to provide the first modulated carrier wave. In the vector subtractor 18 (a directional coupler being available therefor), the first modulated carrier wave of the remodulator circuit 14 is subtracted from the input superposition-modulated carrier wave branched to provide only the second modulated carrier wave. The second demodulated signals S.sub.21 and S.sub.22 can be obtained by demodulating the second modulated carrier wave in the demodulator circuit 19.

However, in the case that a superposition-modulated carrier wave is applied to the input of this phase synchronizing circuit 10, this phase synchronizing circuit operates as a phase synchronizing circuit for the first modulated wave component, but the second modulated wave is deemed as an undesired wave causing an interference. This operation mode will be described with reference to FIGS. 4 and 5. FIG. 4 shows output beat waveforms of the phase comparator circuit 15 with the remodulator circuit 14 interrupted, in which the abscissa represents a phase difference between the input signal at the input of the phase comparator circuit 15 and the output of the VCO 17. In FIG. 4, a beat waveform A represents the case where the phase difference between the first and second modulated waves is equal to .pi./2, a waveform B represents the case of in-phase, a waveform C represents the case of opposite phase, and a waveform D represents the case of 3/2.pi.. In this way, the beat signal at the output of the phase comparator circuit 15 is subjected to phase variation of Q = tan.sup.-1 .beta./.alpha. and a level variation of 20 log (.alpha. + .beta./.alpha. - .beta.) (MAX) depending upon the phase relation between the first modulated wave and the second modulated wave. Consequently, starting from the state shown in FIG. 4, when the remodulator 14 is operated and a 4-phase phase error signal is obtained at the output of the phase comparator 15, the output waveforms take the forms shown in FIG. 5 which are directly subjected to the aforementioned phase variation and level variation. Here, point S represents a phase stable point.

As described above, in the phase synchronizing circuit 10 for the superposition-modulated wave, the phase error signal is directly affected by the phase and amplitude variations caused by the superposition-modulation system. Therefore, even if jitter components are suppressed by the low-pass filter 16, the phase error signal would contain a large amount of jitter components, making it impossible to obtain a reference carrier wave having an excellent carrier-to-noise ratio.

According to the present invention, there is provided an improved phase synchronizing circuit free of those disadvantages in which a phase variation component or an amplitude variation component contained in a phase error signal can be sufficiently compensated.

FIG. 6 shows one preferred embodiment of the remodulation type phase synchronizing circuit according to the present invention. This phase synchronizing circuit can be deemed also as a demodulator device for the superposition modulated carrier wave. In this figure, reference numeral 21 designates an A-D converter circuit; 22, a logic circuit; 23, a first phase modulator; 24, a second phase modulator; 25, a first amplitude modulator; and 26, a second amplitude modulator. Component elements designated by reference numerals 11 and 14 to 17, respectively have the same functions as those represented by like reference numerals in FIG. 3. The first phase modulator 23 is for giving a phase shift of +tan.sup.-1 .beta./.alpha., which is actuated by a drive signal G1 issued from the logic circuit 22 when the phase difference between the first modulated wave and the second modulated wave is equal to 3/2.pi.. By this operation, the phase error signal D' in FIG. 5 is brought to phase coincidence with the phase error signal C'. The second phase modulator 24 is for giving a phase shift of -tan.sup.-1 .beta./.alpha., which is actuated by a drive signal G2 when the phase difference between the first modulated wave and the second modulated wave is equal to .pi./2. By this operation, the phase error signal A' in FIG. 5 is brought to phase coincidence with the phase error signal C'. The first amplitude modulator 25 is for giving an attenuation of 20 log ##EQU1## which is actuated by a drive signal G3 when the phase difference between the first modulated wave and the second modulated wave is equal to .pi./2 or 3/2.pi.. By this operation, the amplitude values of the phase error signals A' and D' in FIG. 5 are brought to coincidence with that of the phase error signal C'. In addition, the second amplitude modulator 26 is for giving an attenuation of 20 log (.alpha. + .beta./.alpha. - .beta.) dB, which is actuated by a drive signal G4 when the phase difference between the first modulated wave and the second modulated wave is equal to zero. By this operation, the amplitude value of the phase error signal B' in FIG. 5 is made to coincide with that of the phase error signal C'. Owing to the above-described operations, all the phase error signals A' to D' in FIG. 5 coincide in phase values and amplitude values with each other, so that the amplitude variation as well as the phase variation caused by the superposition-modulation can be compensated. The drive signals G1 to G4 are generated by the A-D converter 21 and the logic circuit 22.

FIG. 7 shows a more detailed circuit arrangement of the A-D converter circuit 21 and the logic circuit 22. The A-D converter circuit 21 is for reproducing a first demodulated signal and a second demodulated signal from the superposition-demodulated signal, and is composed of squaring circuits constituted by full-wave rectifier circuits 30 and 31, discriminator-shaper circuits 32 to 35, and Exclusive-OR circuits 36 and 37. The operation of this A-D converter circuit 21 will be explained with reference to FIG. 8. The output signal from the 4-phase detector circuit 11 in FIG. 6 is a signal demodulated in the form such that the second modulated signal is superposed on the first modulated signal, and so, it is a 4-level demodulated signal. Assuming that the demodulated signal at the channel CH1 has a pattern series as shown by waveform 1 in FIG. 8, then the signal formed by superposing the second demodulated signal as shown by waveform 3 on the first demodulated signal as shown by waveform 2. Accordingly, it is only necessary to separate these two signals by any appropriate means. At first, if the waveform 1 is discriminated with respect to a voltage level of 0 in the discriminator-shaper circuit 32, then the waveform 4 can be obtained, which coincides with the waveform 2 and provides a first demodulated signal S.sub.11. On the other hand, if the waveform 1 is subjected to full-wave rectification in the full-wave rectifier circuit 30, then a waveform 5 can be obtained. Further, if the waveform 5 is discriminated with respect to a voltage level of L in the discriminator-shaper circuit 33, then a waveform 6 can be obtained. Referring to the waveform 6, although this waveform contains a second demodulated signal component, its polarity is reversed when the first demodulated signal is at a negative level (assumed as 0 level). Therefore if an Exclusive-OR function of the waveforms 6 and 4 is obtained in the Exclusive-OR circuit 36, then a waveform 7 is obtained as its complementary output. This waveform 7 coincides with the waveform 3, and it can serve as a second demodulator signal S.sub.21. By the above-mentioned operations, a first demodulated signal and a second demodulated signal can be obtained. With respect to the channel CH2, the operations are the same as those described above. Through the above-mentioned processes, first demodulated signal S.sub.11 and S.sub.12 and second demodulated signal S.sub.21 and S.sub.22 can be reproduced.

In the logic circuit 22 shown in FIG. 7, the phase relationship is determined between the first modulated signal and the second modulated signal, on the basis of the demodulated signals S.sub.11, S.sub.12, S.sub.21 and S.sub.22. This logic circuit 22 is composed of Exclusive OR circuits 40 to 44 and AND circuits 45 to 48. It is to be noted here that between the phase of the first modulated wave and the signal S.sub.11, S.sub.12 and between the phase of the second modulated wave and the signal S.sub.21, S.sub.22, respectively, there exist the relations represented by Table Ia. Accordingly, if the relations represented by Table Ia are used, the phase relation between the first modulated wave and the second modulated wave can be determined. Table Ib shows the relations between the signals S.sub.11, S.sub.12 and S.sub.21, S.sub.22 in the case where the phase difference between the first modulated wave and the second modulated wave is zero, Table Ic shows the same in the case where the phase difference is .pi., Table Id shows the same in the case where the difference is .pi./2, and Table Ie shows the same in the case where the difference is 3/2.pi.. In accordance with the conditions given by Tables Ib - Ie, the decision circuit 22 can be formed in the following manner. The Exclusive-OR function of the signals S.sub.11 and S.sub.21 is obtained in an Exclusive-OR circuit 40. Also the Exclusive-OR function of the signals S.sub.12 and S.sub.22 is obtained in an Exclusive-OR circuit 41.

Further, if the Exclusive-OR function of the complementary outputs of the circuit 40 and the circuit 41 is obtained in the Exclusive-OR circuit 44, then a signal G3 representing a phase difference of .pi./2 or 3/2.pi. can be obtained at the output of the circuit 44, and if a logic product of the complementary outputs of the circuits 40 and 41 is taken by the AND circuit 46, then a signal G4 representing an in-phase relationship is obtained at the output of the circuit 46. In addition, the Exclusive-OR function of the signals S.sub.12 and S.sub.21 is obtained in the Exclusive-OR circuit 42, the Exclusive-OR function of the signals S.sub.11 and S.sub.22 is obtained in the Exclusive-OR circuit 43, and a logic product of an output of the circuit 42 and a complementary output of the circuit 43 is taken by the AND circuit 47. Then at the output of this circuit is obtained a signal G1 representing a phase difference of 3/2.pi.. Furthermore, if a logic product of a complementary output of the circuit 42 and an output of the circuit 43 is taken by the AND circuit 48, then a signal G2 representing a phase difference of .pi./2 can be obtained at the output of the circuit 48. Still further, if the logic product of the outputs of the circuits 40 and 41 is taken by the AND circuit 45, then a signal G5 representing a phase difference of .pi. can be obtained at the output of the circuit 45. Through the above-mentioned operations, signals G1 to G5 can be derived from the logic circuit 22.

When the phase modulators 23 and 24 and the amplitude modulators 25 and 26 are driven by the thus obtained signals G1 to G4, respectively, the phase variation as well as the amplitude variation which are generated in the phase error signal due to superposition modulation as described above, can be compensated. With regard to the signal G5, description will be given later.

FIG. 9 shows another preferred embodiment of the present invention as applied to a phase synchronizing circuit employing a base-band processing. In this figure, reference numerals 50 to 55 squaring circuits constituted by full-wave rectifier circuits; 56 to 59, phase shifters; 60 to 62, subtractors; 63 to 65 and 68, analog switches (SW); 66, an adder; 67, an Exclusive-OR circuit; 69, an amplitude modulator; and 70, the entire circuit illustrated in FIG. 7. In addition, numerals 11, 16 and 17 designate the circuits represented by like numerals in FIG. 6. The heretofore known phase synchronizing circuit is composed of the circuits 11, 16, 17, 50, 51, 60, 67 and 68. In connection to this known apparatus, reference should be made to U.S. Pat. No. 3,983,499 issued Sept. 28, 1976. In this heretofore known synchronizing circuit, the phase error signal for the superposition-modulated wave is as shown in FIG. 10, in which it is subjected to a phase variation as well as an amplitude variation similarly to the case of FIG. 5. In order to compensate for these variations, the circuit shown in FIG. 9 can be employed. In FIG. 9, the circuit arrangement is constructed in such manner that a signal appearing at point K may be phase-shifted by +tan.sup.-1 .beta./.alpha. with respect to a signal appearing at point J, while a signal appearing at point M may be phase-shifted by -tan.sup.-1 .beta./.alpha. with respect to the signal at point J. Therefore, in the case where the phase difference between the first modulated wave and the second modulated wave is equal to .pi./2, an output at the point J is switched off by the analog switch 63 in response to an inverted signal G4 for the output signal G4, while an output signal at the point M is switched on by the analog switch 65 in response to the signal G2. Through these operations, phase error signal A" in FIG. 10 is brought to phase coincidence with a phase error signal C". On the other hand, in the case where the phase difference between the first modulated wave and the second modulated wave is equal to 3/2.pi., the output at the point J is switched off by the analog switch 63 in response to the inverted signal G4 for the signal G4, while the output signal at the point K is switched on by the analog switch 64 in response to the signal G1. Through these operations, a phase error signal D" in FIG. 10 is brought to phase coincidence with the phase error signal C". In other words, if the output at the point J is selected when the phase difference between the first modulated wave and the second modulated wave is equal to 0 or the output at the point M is selected when the phase difference is equal to .pi./2, and the output at the point K is selected when the phase difference is equal to 3/2.pi., and if these signals are added together by the adder 66 and the sum signal is used as a phase error signal (a double frequency-multiplied signal), then all the phase error signals (4-fold frequency-multiplied signals) at the point H would coincide in phase with the phase error signal C" in FIG. 10. While the phase variation is compensated in the above-described manner, the amplitude variation can be compensated in the following way. In the case where the phase difference between the first modulated wave and the second modulated wave is equal to zero, the amplitude modulator 69 is actuated by the drive signal G3. The circuit 69 gives an attenuation of 20 log .alpha. + .beta./.alpha. - .beta., and through this operation the amplitude value of the phase error signal B" in FIG. 10 is brought to amplitude coincidence with the error signal C". In addition, if the outputs at the points K and M are adjusted so that the amplitude value of the phase error signal at the point H caused by the outputs at the points K and M may coincide with the amplitude value of the phase error signal C" in FIG. 10, then the amplitude values of the phase error signals A" and D" in FIG. 10 will coincide with the amplitude value of the phase error signal C". Through the above-mentioned operations, the amplitude variation can be compensated.

Nextly, a circuit for phase-shifting the output at the point K by +tan.sup.-1 .beta./.alpha. with respect to the output at the point J will be explained. The phase shift of +tan.sup.-1 .beta./.alpha. can be achieved by the phase shifters 56 and 57. The outputs CH1 and CH2 of the 4-phase phase detector 11 are orthogonal to each other, so that by varying the amplitude values of the respective signals and adding or subtracting them to or from each other, any arbitrary phase condition can be attained. For instance, in the case where a phase difference of +tan.sup.-1 .beta./.alpha. is desired, representing CH1 by cos .theta. and CH2 by sin .theta., then since cos ##EQU2## is equal to ##EQU3## it can be achieved by subtracting CH2 from CH1 at a ratio of .beta.:.alpha., respectively. Or else, since sin (.theta. + tan.sup.-1 .beta./.alpha.) is equal to ##EQU4## it can be achieved by adding CH1 and CH2 at a ratio of .beta.:.alpha., respectively. With respect to the phase shifters 58 and 59 also, similar operations are performed.

As described above, the phase variation and amplitude variation caused by the superposition-modulation can be compensated even in the embodiment of the present invention illustrated in FIG. 9 as applied to a phase synchronizing circuit employing the signal processing in the base-band frequency region. With regard to the method for compensating the phase variation, in a phase synchronizing circuit employing a base-band processing it is achieved by means of arithmetic circuits and analog switches, while in a phase synchronizing circuit employing a carrier band processing it is achieved by means of phase modulators.

The preferred embodiments illustrated in FIGS. 6 and 9 are circuits for completely compensating the phase and amplitude variation components contained in the phase error signal. However, since the jitter of a phase error signal is more largely affected by the phase variation component than by the amplitude variation component, the jitter component of the phase error signal can be reduced to a sufficiently small value in comparison to the heretofore known circuit, even if the amplitude modulators 25, 26 and 69 should be removed. Therefore, the amplitude modulator is not an essentially necessary element for practicing the present invention.

The phase variation contained in the phase error signal arises when the phase difference between the first and the second modulated waves is equal to .pi./2 or 3/2.pi., and in the embodiments illustrated in FIGS. 6 and 9 phase compensation at that time is achieved. However, there is another approach, in which the voltage-controlled oscillator 17 is controlled by the phase error signal representing that the phase difference between the first and the second modulated waves is equal 0 or .pi.. However, it is not responsive to the phase error signal representing the phase difference of .pi./2 or 3/2.pi.. This approach applied to the embodiment of FIG. 3 is that a switch is inserted between the phase comparator 15 and the low-pass filter 16 and is turned off by the signal G3 shown in FIG. 7.

According to this approach, too, the jitter component of the phase error signal can be reduced in comparison to the heretofore known circuit.

FIG. 11 shows still another embodiment of the present invention as applied to a reverse modulation system, in which reference numeral 80 designates a 4-phase re-modulator, numeral 81 designates a vector subtractor, numeral 82 designates a 4-phase modulator, and numeral 83 designates a logic circuit. The operation of this circuit arrangement will be described hereunder.

In the re-modulator 80 (which differs from the re-modulator 14 shown in FIG. 3 in function), the input superposition-modulated carrier wave is reversely modulated by the first demodulated signals S.sub.11 and S.sub.12 to obtain the second modulated carrier wave.

The carrier wave derived by the re-modulator 80 is represented as shown in FIG. 12. A vector PQ represents the case where the phase difference between the first modulated wave and the second modulated wave is equal to .pi./2, and an angle .angle.QPR is represented by tan.sup.-1 .beta./.alpha. as explained above. Likewise, in case that the phase difference is equal to 3/2.pi., it is represented by a vector PS and an angle .angle.RPS is represented by tan.sup.-1 .beta./.alpha.. In addition, a vector PW has a phase difference of .pi., while a vector PV has a phase difference of O. To compensate for the vector variation, the 4-phase modulator 82 and the vector subtractor circuit 81 are employed. In the modulator 82, the carrier wave supplied from the oscillator 17 is modulated by the control signals X and Y to provide the signal having the vectors PQ, RV, RS and RW shown in FIG. 12. Accordingly, if the output vector of the 4-phase modulator 82 is controlled in accordance with the output vector of the re-modulator 80 as shown in Table IIa, and if they are subjected to subtraction in the vector subtractor 81, then at the output of the subtractor 81 there is obtained a constant output represented by a vector PR. It is to be noted that the control signals X and Y for the 4-phase phase modulator 82 are only required to satisfy the condition as shown in Table IIb. Here it is to be noted that the circuit 82 is assumed to be of Gray type (or parallel type). The Gray type modulator is disclosed in FIG. 2.4 of an article published by S. Yokoyama et al., in NEC Research and Development, No. 36, Jan. 1975, pp. 21 to 31, entitled "The Performance and the Features of the 4-Level PSK MODEM for an Ultralarge TDMA System." Accordingly, the control signals X and Y can be obtained at the output of the circuit 83 which is constructed of OR circuits 84 and 85 by making use of the output signals G1, G2 and G5 from the logic circuit 22. This circuit 83 could be considered to be included in the logic circuit 22.

As will be apparent from the above description, it is necessary to know the phase difference between the first modulated wave and the second modulated wave by whatever means. That is, an essentially necessary element for practicing the present invention is a logic circuit for determining the phase difference between the first modulated wave and the second modulated wave.

In the phase synchronizing circuits of the present invention, the 16-level superposition modulated wave is employed as an input wave. The phase sychronizing circuit capable of receiving IN(>16)-level superposition modulated wave may be arranged in the same manner described above.

The preferred embodiments illustrated in FIGS. 6, 9 and 11 also provide means for demodulating the second modulated signal. Therefore, if this demodulating means is employed, the circuit for demodulating the second modulated wave as used in the prior art apparatus such as, for example, the demodulator circuit 19 in FIG. 3, becomes unnecessary, and so, there is an advantage that the number of the component units can be reduced.