JPH0541717A - Demodulator for digitally modulated waves - Google Patents

Abstract

(57) [Abstract] [Purpose] To prevent the deterioration of receiver performance even if the input frequency is detuned. [Structure] A QPSK modulated wave is quadrature detected by detectors 2 and 3, digitized by A / D converters 6 and 7, and frequency converted by a complex multiplier 33 for AFC. Next is a digital LP
The spectrum is shaped in F8 and F9, and is multiplied by the reproduced carrier in the complex multiplier 11. The output of the multiplication is the phase detector 1
It is detected in 2 and phase information is obtained from this. The phase information is used for data demodulation in the data determination circuit 13.
The loop filter 14, the numerically controlled oscillator 15, the sine, and the cosine data converters 16 and 17 are carrier reproducing PLLs.
To form the frequency error detection circuit 19 and the loop filter 2
0, the latch circuit 21, the numerically controlled oscillator 22, and the data converters 31 and 32 form an AFC loop. When the device starts up, the AFC loop operates, and when the frequency error becomes sufficiently small, switching to the PLL operation is performed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a demodulation device for digitally modulated waves used in satellite communication, satellite broadcasting and the like.

[0002]

2. Description of the Related Art In a system for transmitting a video signal, an audio signal, etc., there is a digital modulation technique which is useful for high quality transmission and improvement of frequency utilization efficiency. Has been. As a digital modulation system, 16QPAM (16-value quadrature amplitude modulation), 64QPAM (64-value quadrature amplitude modulation), etc., which have good frequency utilization efficiency in terrestrial microwave circuits, etc. are used, and BPS with a small transmission code error rate in satellite circuits
The K (two-phase phase modulation) and QPSK (four-phase phase modulation) systems have been generally used.

By the way, in recent years, such digital transmission technology has come to be used also for consumer use, and it is more and more popular because it is superior in terms of high-quality signal transmission characteristics and frequency utilization efficiency of digital transmission. Can be considered. By the way, it is most important to realize an inexpensive receiving device as a consumer system, and the following requirements must be met. 1) Simple configuration and small hardware scale. 2) There are few initial adjustment points. 3) Stable aging and temperature drift. 4) Must be suitable for IC. 5) Must meet the required performance.

A conventional demodulation system for a terrestrial microwave circuit is disclosed in, for example, NTT Research Practical Report Vol. 24, No. 10 (1975) p. As shown in Yamamoto et al., "Design and characteristics of carrier synchronization system for high-speed QPSK," 253-272, a demodulation device of an inverse modulation system was used. However, such a demodulation device has a complicated structure, and there are many initial adjustment points and the cost is high. Also, there are stability problems due to the large number of analog circuits.
It was not suitable for C conversion and had many problems when used as a consumer demodulator.

As one demodulation method for solving the above problems, a demodulation device to which a digital signal processing technique, which has been remarkably developed in recent years, has been known. An example of this is found in Yagi et al. "Development of digital demodulation LSI for satellite communication" at the Autumn National Conference of the Institute of Electronics, Information and Communication Engineers (1990). This demodulator demodulates a BPSK or QPSK modulated wave (coherent detection)
The carrier reproduction at the time of performing is performed by a digitalized phase-locked loop (PLL), which satisfies the requirement of an inexpensive receiver by no adjustment and LSI.

FIG. 5 shows a conventional receiver circuit block using digital PLL carrier reproduction. The QPSK modulated wave introduced to the input terminal 1 is distributed and supplied to the in-phase detector 2 and the quadrature detector 3. The local oscillator signal (hereinafter abbreviated as “local oscillator”) supplied to the detectors 2 and 3 is divided into a local oscillator 5 having a fixed frequency and a local oscillator 5 having a 0 ° phase and a 90 ° phase. It was done. Detector 2,
The outputs of 3 are input to A / D converters 6 and 7, respectively,
Converted to digital value. Further, the digitized detection outputs are respectively input to digital low-pass filters (LPF) 8 and 9 having the same frequency transfer characteristic, and spectrum-shaped. These digital low-pass filters 8 and 9 are filters that form transmission characteristics required to prevent intersymbol interference in digital data transmission, and when combined with filter characteristics on the transmission side, so-called roll-off characteristics are generally provided. Designed to be obtained. Therefore, at the outputs of the digital low pass filters 8 and 9,
Each detection output is spectrally shaped so that the eye opening ratio is sufficiently large. The respective outputs of the digital low-pass filters 8 and 9 are branched, one of which is the clock recovery circuit 1.
0, the symbol timing component in the signal is extracted and fed back to the conversion clock inputs of the A / D converters 6 and 7. The outputs of the branched digital low-pass filters 8 and 9 are input to the complex multiplier 11, respectively.

The complex multiplier 11 can realize exactly the same operation as the frequency converter or mixer in the intermediate frequency band in the base band. (A real number-type multiplier that does not use a complex number can perform a detection operation, but cannot represent a negative frequency component and therefore cannot be a general frequency converter.) The output of the complex multiplier 11 is a phase detector. The phase difference between the input signal and the numerically controlled oscillator (NCO) 15 described later is detected. The output of the phase detector 12 (phase difference information) is input to the data determination circuit 13. The data determination circuit 13 determines the QPSK data from the phase difference information, that is, demodulates and outputs the demodulated data.

The phase difference information from the phase detector 12 is input to the frequency control terminal of the numerically controlled oscillator 15 via the loop filter 14 for carrier regeneration. The numerically controlled oscillator 15 is a cumulative addition circuit that does not prohibit overflow. Since it performs an addition operation up to its dynamic range according to the value of the signal input to the frequency control terminal, it is in an oscillating state and its oscillation frequency is equal to that of the control signal. It changes with the value. That is, it operates exactly the same as the voltage controlled oscillator (VCO) in the analog circuit. General VCO
2 is different in that the oscillation frequency is very stable, and is characterized by having stability higher than that of a so-called VCO (VCXO) using a crystal and having a wide frequency variable range that cannot be realized by a VCXO. The output of the numerically controlled oscillator 15 is divided into two and is returned to the complex multiplier 11 via the data conversion circuits 16 and 17 having sine and cosine characteristics, respectively.
Is. If the loop filter 14 includes a circuit having a complete integration system, the frequency pull-in range of the PLL is infinite in principle, and ideal operation as the PLL can be expected. Further, since the A / D converters 6 and 7 are all digital signal processing, if they are integrated into an IC, a demodulation device can be realized without adjustment and a very compact device can be realized.

[0009]

Even in the demodulator using the above digital PLL, there still remains a problem regarding frequency detuning. In satellite communication and satellite broadcasting, it is difficult to improve the stability of the frequency converter (used to make the uplink frequency and the downlink frequency different) inside the repeater mounted on the satellite, and generally large frequency detuning is performed. Have. Also in the receiver, considering a receiver for consumer use, a frequency synthesizer type down converter having a very stable frequency becomes expensive, which hinders its popularization. Therefore, frequency conversion is generally performed by a cheaper circuit (for example, a circuit using a dielectric resonator for local oscillation). In such a frequency converter, frequency detuning will still occur. For example, this frequency detuning is a frequency conversion in the 12 GHz band and extends to 1 MHz or more, and this frequency detuning appears at the demodulator input.

As described above, the frequency pull-in range (PLL pull-in range) of the demodulator having the digital PLL can be infinite as described above. So there is no problem with this. However, there are the following problems.

It is assumed that the input frequency is detuned and the center frequency (carrier frequency) of the modulated wave spectrum becomes fc as shown as the band characteristic in the figure. Local fixed frequency 5
When the frequency of is set to fL, the spectrum of each detection output is not a target spectrum for frequency 0 (direct current) due to the frequency shift. This spectrum is digitized and then spectrally shaped by the digital PLL. The characteristic of this low-pass filter is symmetrical with respect to direct current as shown in the figure. The spectrum of the signal is partially scraped. This means that the transmission characteristic for preventing intersymbol interference is not satisfied, and as a result, there is a problem that the eye opening ratio becomes small and the code error rate deteriorates. Next, in order to deal with the above problem, a demodulation device in which a digital low-pass filter for spectrum shaping is moved after a complex multiplier can be considered.

FIG. 6 shows the assumed device. The configuration of FIG. 6 is simply a configuration in which the digital low-pass filters 8 and 9 of FIG. 5 are moved to the subsequent stage of the complex multiplier 11. Other configurations and operations are exactly the same as those in FIG. In this case, since the above frequency detuning is removed by the complex multiplier, the spectrum of the digital modulation wave is not partially shaved off. However, the following new problems arise.

A digital low-pass filter for spectrum shaping is generally constructed by using a transversal filter, and if an ideal transfer characteristic is to be realized, it will be expressed by the formula (10).
It becomes a filter with taps. If such a delay element is present in the PLL loop, the feedback control will be close to an unstable state, and the jitter characteristics and pull-in range will be deteriorated.

Therefore, the present invention provides a digital modulation wave that does not cause the above-described deterioration of the receiver performance even when there is frequency detuning of a satellite or the like or frequency detuning in a frequency converter of an inexpensive receiver. It is an object of the present invention to provide a receiving device of.

[0015]

According to the present invention, there is provided means for obtaining a detection output by multiplying a modulated wave input by a local oscillation frequency signal from a first local oscillator, and an analog-digital conversion for digitally converting the detection output of this means. Means, frequency conversion means for multiplying the output of this means by a frequency conversion carrier from the second local oscillator to perform complex multiplication to obtain frequency conversion, and a digital low frequency converter for supplying spectrum shaping to the output of the frequency conversion means. The low-pass filter and the output of this low-pass filter are multiplied by the reproduction carrier to perform complex multiplication,
A means for obtaining phase information by subjecting this multiplication output to phase detection, a means for obtaining a multiphase demodulated output from this phase information, and a phase for smoothing the phase information and supplying it to a numerically controlled oscillator to obtain the reproduction carrier. Lock loop means, means for detecting the frequency error of a predetermined relationship between the frequency of the modulated wave input and the local oscillation frequency for the quadrature quasi-coherent detection, and the frequency error from this means. A means for smoothing the output and a means for controlling the oscillation frequency of the second local oscillator by the smoothed output obtained from the means are provided, and the digital signal is provided between the phase locked loop means and the frequency conversion means. A low pass filter is arranged.

[0016]

By the above means, at the first rise, frequency error detection, frequency control of the second local oscillator based on this error, that is, AFC operation is executed. As a result, the state of separation between the input signal and the local oscillation frequency of itself is detected, and the frequency error of this detection is reduced by the AFC loop filter means. The output of the filter means is then supplied to the frequency control input terminal of the second local oscillator. The frequency control means obtains the same operation as the mixer in the intermediate frequency band. After this control, carrier regeneration is performed by the phase locked loop (PLL). There is no digital low pass filter for spectral shaping in this PLL. As a result, the frequency detuning of the input is removed by the AFC operation. Therefore, the phenomenon that a part of the modulated wave spectrum is scraped off does not occur in the output of the digital LPF means for spectrally shaping the in-phase and quadrature detection outputs, and accurate spectrum shaping becomes possible. Therefore, even when there is frequency detuning in the input, the deterioration of the eye opening ratio does not become a problem. Further, since there is no large delay element in the PLL loop, the jitter characteristics and the like during the operation of the PLL do not deteriorate. When the frequency detuning amount is sufficiently reduced by the AFC, the AFC is held and the PLL is activated by the means for activating the PLL this time, and phase synchronization is achieved after drawing in a slight frequency error. ..

[0017]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 shows an embodiment of the present invention. The same parts as those of the circuit of FIG. 5 described above are designated by the same reference numerals. The QPSK modulated wave introduced to the input terminal 1 is distributed and supplied to the in-phase detector 2 and the quadrature detector 3. The local oscillation signal (hereinafter abbreviated as “local oscillation”) given to the detectors 2 and 3 is the divider 4 in which the local oscillation of the first local oscillator 5 having a fixed frequency is generated by the distributor 4. It was a local departure. The outputs of the detectors 2 and 3 are input to the A / D converters 6 and 7, respectively, and converted into digital values.

Here, the digitized detection output is input to the complex multiplier 33 which realizes frequency conversion. A local oscillator output (second local oscillator output) from an AFC loop, which will be described later, is supplied to the complex multiplier 33 as a frequency conversion carrier.
Is supplied. The outputs obtained from the complex multiplier 33 are input to the digital low-pass filters (LPF) 8 and 9 having the same frequency transfer characteristic, respectively, and the spectrum is shaped. These digital low-pass filters 8 and 9 are filters that form transmission characteristics required to prevent intersymbol interference in digital data transmission, and when combined with filter characteristics on the transmission side, so-called roll-off characteristics are generally provided. Designed to be obtained. Therefore, at the outputs of the digital low pass filters 8 and 9,
Each detection output is spectrally shaped so that the eye opening ratio is sufficiently large. The respective outputs of the digital low-pass filters 8 and 9 are branched, one of which is the clock recovery circuit 1.
0, the symbol timing component in the signal is extracted and fed back to the conversion clock inputs of the A / D converters 6 and 7. The outputs of the branched digital low-pass filters 8 and 9 are input to the complex multiplier 11, respectively.

The complex multiplier 11 can realize exactly the same operation as the frequency converter or mixer in the intermediate frequency band in the baseband. (A real number-type multiplier that does not use a complex number can perform a detection operation, but cannot represent a negative frequency component and therefore cannot be a general frequency converter.) The output of the complex multiplier 11 is a phase detector. The signal is input to 12 and the phase difference between the input signal and the numerically controlled oscillator 15 is detected.
The output of the phase detector 12 (phase difference information) is input to the data determination circuit 13. The data determination circuit 13 determines the QPSK data from the phase difference information, that is, demodulates and outputs the demodulated data.

Further, this phase difference information from the phase detector 12 is input to the frequency control terminal of the numerically controlled oscillator 15 via the loop filter 14 for carrier regeneration. The numerically controlled oscillator 15 is a cumulative addition circuit that does not prohibit overflow. Since it performs an addition operation up to its dynamic range according to the value of the signal input to the frequency control terminal, it is in an oscillating state and its oscillation frequency is equal to that of the control signal. It changes with the value. That is, it operates exactly the same as the voltage controlled oscillator (VCO) in the analog circuit. General VCO
2 is different in that the oscillation frequency is very stable, and is characterized by having stability higher than that of a so-called VCO (VCXO) using a crystal and having a wide frequency variable range that cannot be realized by a VCXO. The output of the numerically controlled oscillator 15 is divided into two and is returned to the complex multiplier 11 via the data conversion circuits 16 and 17 having sine and cosine characteristics, respectively.
Is. If the loop filter 14 includes a circuit having a complete integration system, the frequency pull-in range of the PLL is infinite in principle, and ideal operation as the PLL can be expected. Further, since the A / D converters 6 and 7 are all digital signal processing, if they are integrated into an IC, a demodulation device can be realized without adjustment and a very compact device can be realized.

Next, an AFC loop is formed in this system. That is, the phase error signal output from the phase detector 12 is supplied to the frequency error detection circuit 19.
The frequency error detection circuit 19 detects the frequency error between the input and the local oscillator (first local oscillator and second local oscillator).
This frequency error component is smoothed by the AFC loop filter 20, passed through the latch circuit 21, and supplied to the frequency control terminal of the numerically controlled oscillator 22. Since the oscillation output of the numerically controlled oscillator 22 is a sawtooth wave signal, it is input to the data converter 31 having a sine conversion characteristic and the data converter 32 having a cosine conversion characteristic. Data converter 31,
The output of 32 is supplied to the complex multiplier 33 as the second local oscillator. As a result, an AFC loop is formed.

When the AFC operation is performed and the frequency detuning of the predetermined relationship between the input signal and the local oscillation becomes sufficiently small, the frequency error detection output of the frequency error detection circuit 19 changes. As a result, the frequency error detection circuit 19 outputs the loop switching signal and the AFC hold signal.
This signal is substantially the same. The loop switching signal is
The loop filter 14 in the PLL is switched to the operating state. This starts the PLL operation, and the AFC
The error information in the loop is held in the best AFC state. The PLL operation starts the pull-in operation so as to be in carrier synchronization. The frequency conversion operation by the complex multiplier 33 will be further described below.

Frequency conversion in the intermediate frequency (IF) band is generally performed by an analog mixer, and is realized by obtaining the sum frequency component or the difference frequency component of the local oscillator and the input. However, such an operation is premised on that each spectrum does not fall below DC in principle. If a part of the frequency conversion output has a frequency component below the direct current, this part is folded back at the direct current, resulting in detection. In this way, complex frequencies can be used to represent negative frequencies below DC. The mixer operation may be performed in the complex domain. Since the mixer operation is nothing but multiplication, if a complex multiplier is used, frequency conversion in the vicinity of direct current (that is, in the baseband) can be freely realized without considering the aliasing of the spectrum.

In this embodiment, frequency conversion for AFC is realized using the complex multiplier 33 described above. The complex multiplier in the digital PLL operates exactly the same.

Unlike the conventional demodulator which uses a fixed frequency local oscillator, since the local oscillator in this embodiment is feedback-controlled by the frequency error output, the quadrature synchronous detection is performed with almost no frequency detuning. To be done. Therefore, the spectrum is not scraped when the spectrum is shaped by the digital low-pass filter, and demodulation can be performed in an almost ideal state.

Further, as described above, since a large delay element is not included in the carrier reproduction PLL, the carrier reproduction operation having a good jitter characteristic can be performed. Further, since the frequency pull-in range of the PLL can be widened, very strict characteristics are not required for the AFC operation. That is, even if the frequency error within the pull-in range of this PLL circuit remains, the carrier reproducing operation is possible.

FIG. 2 shows a concrete example of the numerically controlled oscillator 22 for the AFC loop. This numerically controlled oscillator 22 is
The contents are the same as the numerically controlled oscillator 15 in L. The signal input to the frequency control terminal 200 is added by the adder 201.
To one input and the other input to the adder 201
Is output by the latch circuit 202 after being delayed by one clock. As a result, the output of the latch circuit 202 operates as an accumulator and outputs a sawtooth wave as shown in FIG. This oscillation frequency can be controlled by a numerical value applied to the control terminal 200 if the clock has a fixed frequency.

FIG. 3 shows a configuration example of the frequency error detection circuit 19 in the above embodiment. It is assumed that the phase error signal is input to the input terminal 300. Latch circuit 301 that uses the symbol timing of a digital modulated wave as a clock
And a subtractor 302 detects a phase change between the symbols of the digital modulated wave. This is nothing but obtaining the frequency error. Therefore, this output is also output to the output terminal 303 as a frequency error detection output. The frequency error detection output is also branched to a low-pass filter 304, where it is sufficiently smoothed, and then passes through an absolute value circuit 305 that removes positive and negative signs, and a binarization circuit 3
It is binarized at 06. This binary output is AFC or PL
A signal for controlling which operation of L should be performed (A above)
FC hold and loop switching signal).

In the above embodiment, the numerically controlled oscillator 22
A data converter 24 for converting to a sine wave is used as a local oscillation means for quadrature detection whose frequency is controlled as an output of
The configuration is shown in which the output of the data converter 24 is supplied to the D / A converter 25 and the analog conversion output thereof is supplied to the frequency multiplication circuit. However, this local oscillation means can also be realized by using another frequency conversion circuit. Figure 4
(A) is another embodiment of the frequency conversion circuit using a mixer.

The same parts as those of the circuit of FIG. 1 are designated by the same reference numerals. That is, the output of the numerically controlled oscillator 22 is converted into a sine wave in the data converter 24 having a sine conversion characteristic, and converted into an analog sine wave in the D / A converter 25. This output is supplied to the RF input terminal of the mixer circuit 401. The output of the fixed frequency oscillator 402 is supplied to the local oscillator input of the mixer circuit 401. Local oscillator 402
If the sine wave frequency is 4 MHz, the oscillating frequency is set to 144 MHz or 136 MHz to obtain a converted output of 140 MHz. The converted output is supplied to the distributor 4 after the local leak component and the image component are removed by the band pass filter (BPF).

In the method using this mixer, the frequency change of the output of the numerically controlled oscillator 22 and the frequency change of the output of the band pass filter 403 are completely the same as in the frequency conversion method using the frequency multiplication circuit, and the frequency multiplication operation is performed. There is a feature that the increase of the phase jitter that occurs does not occur. FIG. 3B shows another embodiment of the frequency converter using the quadrature modulator.

The output of the numerically controlled oscillator 22 is converted into a sine wave by the data converter 24 having a sine conversion characteristic, and converted into an analog sine wave by the D / A converter 25. This output is supplied to the RF input terminal of the mixer 501. Further, the output of the numerically controlled oscillator 22 is supplied to a data converter 501 having a cosine conversion characteristic and converted into an analog cosine wave by a D / A converter 502. This output is supplied to the RF input terminal of the mixer 504.

Reference numeral 506 is a fixed frequency local oscillator, and its oscillation output is distributed by a distributor 505 to outputs of 0 ° phase and 90 ° phase. And the oscillation output of 0 degree phase is
The 90 ° -phase oscillation output is supplied to the mixer 504 and is supplied to the mixer 503. The outputs of the mixers 503 and 504 are combined by a combiner 507 and a band pass filter (BP
F) is supplied to the distributor 4 via 508. This circuit is characterized in that the image components appearing at the outputs of the two mixers 503 and 504 are canceled at the time of synthesis, so that the characteristics of the bandpass filter may be relatively gentle.

Further, in the embodiment of the present invention, when switching between the AFC operation and the PLL operation, the latch circuit 21 and P
This is realized by controlling the LL loop filter 14, but is executed when a control signal appears that the frequency error has become sufficiently small in the frequency error detection circuit 19. However, the present invention is not limited to this, and other embodiments are possible.

FIG. 6 shows another embodiment of the AFC operation / PLL operation switching means. The output phase error signal from the phase detector 12 is supplied to the frequency error detection circuit 19. The operation of the frequency error detection circuit 19 is as described above, and outputs the control signal when the frequency error becomes sufficiently small. The output of the phase detector 12 is supplied to the switch 6 before being supplied to the loop filter 14 on the PLL side.
It is designed to be supplied via 01. The frequency error detection signal obtained by the frequency error detection circuit 19 is A
Before being supplied to the FC loop filter 20, it is supplied via the switch 602. The switches 601 and 602 are provided in the frequency error detection circuit 19
The control signal from the control means controls so that when one of them is on, the other is off.

Also with this configuration, when the frequency error of the output of the phase detector 12 is large, the switch 602 is turned on to execute the AFC operation, and when the frequency error is sufficiently small, the PLL operation is executed. Loop switching can be realized.

As described above, in demodulating a digital modulated wave, even if there is frequency detuning in the transmission system, deterioration of the eye opening rate due to the frequency shift of the digital low pass filter hardly occurs, and the demodulator is fixed. It enables demodulation without data transmission error. In addition, the digital P inside the PLL
Since LL is not included, phase jitter due to the influence of delay in the loop does not occur, and extremely good demodulation of a digital modulated wave becomes possible.

Furthermore, since the entire AFC loop is composed of a digital circuit and there is no control to the analog part, it is easy to make no adjustment and IC, and a cheaper demodulation can be obtained.

[0040]

As described above, according to the present invention,
Even when there is frequency detuning of a satellite or the like or frequency detuning in the frequency converter of an inexpensive receiver, there is no deterioration in receiver performance.

[Brief description of drawings]

FIG. 1 is a structural explanatory view showing an embodiment of the present invention.

FIG. 2 is a circuit diagram showing a specific configuration example of a numerically controlled oscillator and a frequency error detection circuit of FIG.

FIG. 3 is a diagram showing another embodiment of the loop switching means in FIG.

FIG. 4 is a diagram showing another embodiment of the loop switching means in FIG.

FIG. 5 is a diagram showing a conventional demodulator for digitally modulated waves.

FIG. 6 is a diagram showing a demodulator of a digital modulation wave that can be considered from a conventional circuit.

[Explanation of symbols]

2 ... In-phase detector, 3 ... Quadrature detector, 4 ... Distributor, 6, 7
... A / D converter, 8, 9 ... Digital low-pass filter,
10 ... Clock recovery circuit, 11 ... Complex multiplier, 12 ... Phase detector, 13 ... Data determination circuit, 14 ... PLL loop filter, 15 ... Numerically controlled oscillator, 16, 17 ... Data converter, 19 ... Frequency error detection circuit , 20 ... AFC loop filter, 21 ... Latch circuit, 22 ... Numerically controlled oscillator, 31, 32 ... Data converter, 33 ... Complex multiplier.

Claims (3)

[Claims] 1. A means for obtaining a detection output by multiplying a modulated wave input by a local oscillation frequency signal from a first local oscillator, and an analog-digital conversion means for converting the detection output of this means into a digital signal. A frequency conversion means for multiplying an output of the means by a frequency conversion carrier from the second local oscillator to perform a complex multiplication to obtain frequency conversion; and a digital low-pass filter for supplying spectrum shaping to the output of the frequency conversion means. Means for multiplying the output of the low-pass filter by a reproduction carrier to perform complex multiplication, and subjecting the multiplied output to phase detection to obtain phase information, and means for obtaining a multiphase demodulation output from the phase information; Is supplied to the numerically controlled oscillator after smoothing, and phase locked loop means for obtaining the reproduction carrier, the phase information is input, and the modulated wave input Means for detecting a frequency error having a predetermined relationship between the frequency and the local oscillation frequency for the quadrature quasi-coherent detection, means for smoothing the frequency error output from this means, and smoothing output obtained from this means. Means for controlling the oscillation frequency of the second local oscillator, wherein the digital low pass filter is arranged between the phase locked loop means and the frequency conversion means. Demodulator. 2. The means for detecting a frequency error having a predetermined relationship between the frequency of the modulated wave input and the local oscillation frequency for the quadrature quasi-coherent detection is a symbol included in the digitally obtained phase information. 2. The phase error is detected by detecting a phase change during the period.
A demodulator for the digitally modulated wave described. 3. A means for obtaining a detection output by multiplying a modulated wave input by a local oscillation frequency signal from a local oscillator, an analog-digital conversion means for digitally converting the detection output of this means, and an output of this means. , A frequency conversion means for multiplying the frequency conversion carrier from the second local oscillator to perform complex multiplication to obtain frequency conversion, a digital low-pass filter which is supplied with the output of the frequency conversion means and performs spectrum shaping, A means for obtaining phase information by performing phase detection on the multiplication output by multiplying the reproduction carrier by the output of the band-pass filter, and obtaining phase information from the phase information, and the phase information being input. , Detecting a frequency error of a predetermined relationship between the frequency of the modulated wave input and the local oscillation frequency for the orthogonal quasi-coherent detection, and A control signal indicating whether the error is larger or smaller than a constant error is obtained, the oscillation frequency of the second local oscillator is controlled by a smoothed output obtained by smoothing the frequency error output, and when the frequency error becomes sufficiently small, the Frequency control loop means for holding the control state of the local oscillator fixed, and smoothing the phase information and supplying it to the numerically controlled oscillator to obtain the regenerated carrier, and the operation is on / off controlled by the control signal, and the frequency error And a phase-locked loop means that operates only when is sufficiently small.