JPH0198350A - demodulator - Google Patents

Abstract

PURPOSE:To attain stable demodulation of PSK even against the fluctuation of a carrier frequency by controlling a local oscillator based on a difference signal between the pulse count and the reference count in aprescribed time of a recovered intermediate carrier. CONSTITUTION:A PSK modulation wave is mixed with a signal outputted from a voltage controlled oscillator VCO 1 and converted into an intermediate carrier angular frequency, converted into a square waveform signal by a 1st Schmitt trigger circuit 5 and demodulated into the original digital data by a phase detection circuit 6. On the other hand, part from the circuit 5 is recovered into an intermediate carrier by an intermediate frequency recovery circuit 7 and converted into square wave by an 2nd Schmitt trigger circuit 8. The square wave is counted by a pulse counter 10 for a prescribed period and the count number and the reference count stored in the memory 16 are compared and its difference signal controls the oscillation frequency of the VCO 1.

Description

[Detailed Description of the Invention] <Industrial Application Field> The present invention relates to a demodulator, and more specifically, the present invention relates to a demodulator that receives a signal that has been phase modulated so that the time average value of the frequency is equal to the carrier frequency. The present invention relates to a novel demodulation device for obtaining original digital data after converting it into an intermediate frequency signal.

<Prior Art> As a conventional method for transmitting digital signals, a method has been known in which a transmitting side sends out a PSK modulated signal and the receiving side demodulates the signal to obtain the original digital signal. ing.

This kind of PSK modulation method uses “0” of the digital signal.
This method transmits "1" in correspondence with the phase of the carrier wave, and has the excellent property of reducing C/N deterioration, so it has recently been widely adopted for digital signal transmission. It is becoming.

To explain the above PSK modulation method in more detail,
For example, as shown in FIG.
degree, and when the signal is in space, 1 of the signal
In the so-called MSK method, the phase is linearly decreased by 90 degrees throughout the time slot period, and as shown in Fig. 6, when the signal is in a mark state, the phase is divided into two by dividing one time slot period of the signal into two. The so-called DSK method is adopted, in which the phase is increased in two steps by 90 degrees, and when the signal is in a space, one time slot period of the signal is divided into two and the phase is decreased in two steps by 90 degrees. .

The MSK method has the advantage that the occupied frequency band is narrow because the phase change is continuous;
This method has the advantage of being resistant to multipath fading and suitable for wideband data transmission.

Further, as a method for demodulating a signal subjected to PSK modulation as described above, a delay detection method and a synchronous detection method are generally employed.

The delayed detection method described above divides the received signal into two, delays one signal period of modulation by a delay circuit, or 1/2 signal period, and supplies it to the phase comparator, and supplies the other signal as is to the phase comparator. By doing so, the PSK modulated signal is demodulated and the original digital signal is obtained.

To explain in more detail, in the delay detection device shown in FIG. 3A, the input voltage is VIn=eos (Ωt + θ (t)) (
However, Ω is the angular frequency of the carrier wave, t is the time, and θ
(1) is a phase modulation function. ), the input voltage Vln is divided into two, one is supplied to one terminal of the phase comparator (22), the other is delayed by a predetermined time TR by the delay circuit (21), and the phase comparator ( 22), the signal Vc supplied to the above one terminal is Vc - Vln - cos (Ω = 10 (t)), and the signal Vc supplied to the above other terminal is Vc - Vln - cos (Ω = 10 (t)). Vd is Vd −e
os lΩ (t-TR) 10 (t-TI?)).

If, for example, the configuration shown in FIG. 3B is adopted as the phase comparator (22), and the output is proportional to the phase difference as shown in FIG. 3C, then the phase difference Δθ is Δθ−ΩTR+θ(1)−〇(t−TR). However, the delay time TR needs to be set to TR-T/2 (where T is one time slot of the signal) in the MSK system and the DSK system.

Also, ΩTR-(2n-1)π, or Ω-yr/
If TR-<2n-1)2 yr/T, the reference point for phase comparison can be brought to the center of the operating range of the phase shifter.

The following will explain the case of the DSK method as an example.
The same can be applied to the K method as well.

First, in the case of θ(1)-θ(t-TR)-0, since ΔθtanΩTR= (2n-1)π, this point becomes the phase reference point in a state with no phase displacement, Based on this point, θ(t)-θ(t-TR)
The output corresponding to the point that has changed by 2' is obtained.

Also, the phase function θ when the signal is mark-space
(1) is as shown in Figure 4A, and θ(t-T/2
) is as shown in Figure B.

Therefore, as shown in Figure C, θ(1)-〇(t-T/2) becomes π/2 during the mark period and -π/2 during the space period, resulting in the output characteristics shown in the same figure. Based on this, the output waveform shown in E of the same figure is obtained. That is, an output of 3VO/4 is obtained during the mark period, and an output of VO/4 is obtained during the space period.

As a result, if the output of the phase comparator (22) exceeds vO/2, it can be determined that it is a mark, and if it is less than or equal to VO/2, it can be determined that it is a space.

The above-mentioned synchronous detection circuit divides the received signal into two and supplies each to a phase comparator, and also outputs an output signal (signal with a frequency matching the carrier frequency of the received signal) from the voltage controlled oscillator incorporated in the phase-locked loop. is supplied as is to one phase comparator, the phase of the above output signal is shifted by 90 degrees and then supplied to the other phase comparator, and finally the original signal is output based on the output signals from both phase comparators. It is used to obtain digital signals (“Transmission characteristics of GMsK modulation method” Kazuaki Murota, Kenkichi Hiraide, Journal of the Institute of Electronics and Communication Engineers 81/10 Vo.
1. (See JO4-B Nll0).

<Problems to be Solved by the Invention> When demodulating the PSK modulated signal using the delayed detection method, the configuration is simplified because it is sufficient to divide the received signal into two and simply delay one of them. However, when applied to digital signal transmission in a high frequency band, there is a problem that reliability of demodulation decreases.

To explain this point in detail, the operating reference point of the differential detection method is Δθ−ΩT/2. Therefore, if the carrier wave angular frequency changes by ΔΩ due to temperature fluctuations, the operating reference point also changes by ΔΩT/2. If this variation is large, mark and space judgments are
This becomes impossible depending on whether the output level of the phase comparator exceeds vO/2. For example, the carrier frequency is 2.5 GHz, and the oscillator (here S'A
Taking the case of a W oscillator as an example), the degree of temperature fluctuation is ±
3X10-', the frequency fluctuation rate is ±750
It becomes KHz. If the data transmission rate in this case is set to 32 Kbps, T-325se
c, and ΔΩT/2−23.44π, that is, the fluctuation of the operating reference point is approximately 23.44π.

In reality, in addition to temperature fluctuations, the operating reference point is also affected by noise, interference waves due to multipath, etc., and the operating reference point fluctuates further. It becomes completely impossible to determine whether it is a space or a space.

The above-mentioned synchronous detection method is based on the regeneration of the carrier frequency using a Costas loop, and it does not have the disadvantages caused by frequency fluctuations unlike the phase detection method, and can perform signal demodulation with high precision. be.

However, the above synchronous detection method has the following problems.

That is, in order to obtain a signal with a frequency equal to the carrier frequency of the received signal, a voltage controlled oscillator as a local oscillator and a phase locked loop are required, which complicates the configuration and causes an increase in cost. In particular, the above problem is fatal since there is a strong demand for miniaturization, simplification, and cost reduction in radio equipment mounted on vehicles.

Furthermore, the inventor set the delay time equal to the total time of the reference phase section based on the demodulated NRZ signal, thereby reducing the total time of the reference phase section and reducing the amount of fluctuation ΔΩΔT of the operating reference point. The idea was to make it smaller and improve stability.

However, even if the above-mentioned measures are taken, there is a limit to how much stability can be improved in the current situation where carrier frequencies are becoming increasingly common, and technical difficulties may also increase. There is a problem with becoming.

To explain in more detail, if the total time of the reference phase section is made smaller, the time change rate of the phase increases and the occupied frequency band of the modulated wave becomes larger.

In addition, the video stage after detection also handles sharp pulse waves and contains high frequency components, which increases technical difficulty and worsens economic efficiency.

In addition, two types of PSK modulated waves invented by the present inventor (the sum of which forms a reference phase part of a predetermined time at the front and/or rear part of one time slot of a digital pulse signal, and in the first half of the remaining time slots) changes the phase in a predetermined direction in response to either the mark state or the space state of the transmission signal, and in the second half, restores it to the reference phase with a change opposite to the above phase change, and then changes the phase to the other state of the transmission signal. PSK that corresponds and changes the phase in the opposite direction
P which performs phase modulation to form only the u quasi-phase part over the entire range of one time slot of the digital pulse signal in correspondence with the modulated wave and the other state of the transmission signal.
Based on the property that the time average value of the instantaneous angular frequency of the SK modulated wave is equal to the carrier wave angular frequency, the local oscillation frequency at which the received signal is mixed is based on the signal obtained by frequency detection of the intermediate frequency signal. We also considered a method in which the carrier angular frequency of the intermediate frequency signal is stabilized by feedback control, and restoration to a predetermined signal is performed based on this stabilized signal.

In this method, a linear detector whose frequency response is linear with respect to frequency is generally used as a frequency detector to detect the frequency of the intermediate frequency signal.
Moreover, since the linear detector is constructed from an analog circuit, the variation in characteristics becomes quite large. therefore,
In order to achieve the desired demodulation performance, there is a problem in that it is necessary to adjust the demodulation characteristics for each demodulator, and especially when mass-producing demodulators, the time required for the above adjustment work, There is also the problem that the cost increases due to the labor involved.

Therefore, the present applicant further proposed that in a demodulation device, which was filed on September 10, 1985, the above-mentioned intermediate frequency signal is converted into a square wave, the above-mentioned local oscillation signal is generated based on this square wave signal, and the above-mentioned local oscillation signal is generated based on this square wave signal. By mixing the oscillation signal and the modulated wave to stabilize the carrier wave of the intermediate frequency signal, in other words, by configuring a circuit that digitizes the local oscillation signal, variation in characteristics is almost eliminated and the adjustment of demodulation characteristics is easy. It greatly reduced the effort.

However, even in this digital system, as shown in Figure 7, even if the time average value is phase modulated to be equal to the carrier wave (Figure 7B) frequency (Figure 7A), the binary The number of pulses within one slot time between the digitalized signal (FIG. 7C) and the carrier wave binary signal (FIG. 7D) may differ by one pulse. If a local oscillation signal is generated as is despite this difference in the number of pulses, there is a risk that data may not be reproduced accurately.

Furthermore, by setting the intermediate frequency sufficiently high, this level of error can be ignored, but increasing the intermediate frequency requires increasing the number of intermediate frequency amplification stages, which poses the problem of increased costs. be.

<Object of the invention> This invention was made in view of the above problems,
It is possible to demodulate a signal subjected to PSK modulation using the above method invented by the inventor of the present invention easily and with high precision, and the configuration can be simplified.
It is an object of the present invention to provide a demodulator that can eliminate the need for adjustment work.

Means for Solving the Problems> To achieve the above object, the demodulator of the present invention has the following features:
A signal that has been phase modulated so that the time average value of the frequency is equal to the carrier frequency, a mixer that uses the output signal from a variable frequency local oscillator, and a waveform conversion that is applied to the intermediate frequency signal output from the mixer. a first waveform conversion circuit that outputs a square wave signal using a digital pulse signal; an intermediate carrier regeneration circuit that reproduces an intermediate carrier wave from the square wave signal; a second conversion circuit that converts the intermediate carrier wave into a square wave; A pulse counter that counts square waves every n times the time slot (where n is an integer or a sufficiently large number other than an integer), and a reference count number nfOT (however, the count number from the pulse counter is input). fO is the reference intermediate frequency band carrier frequency and T is the time slot of the signal); and D converts the difference signal from the subtraction circuit into an analog signal and supplies it to the variable frequency local oscillator. The second converter includes a /A converter and a restoration circuit which receives the square wave signal from the first conversion circuit and restores it to a predetermined signal.

However, as a signal subjected to phase modulation so that the time average value of the above frequency is equal to the carrier frequency, there is a reference phase part whose sum is a predetermined time at the front and/or rear part within one time slot of the digital pulse signal. In the first half of the remaining time slot, the phase is changed in a predetermined direction, and in the second half, the phase is changed in the opposite direction to the above phase change, corresponding to either the mark state or the space state of the transmission signal. Phase modulation is applied to restore the fundamental phase by a change, and to correspond to the other state of the transmitted signal, to change the phase in the opposite direction to the above case, or to prevent the phase from changing at all. It may be.

Furthermore, although a phase detection circuit can be used as the restoration circuit, for example, it is not limited to a phase detection circuit.

With the demodulator having the above configuration, the signal that has been phase modulated so that the time average value of the frequency is equal to the carrier frequency and the output signal from the variable frequency local oscillator are supplied to the mixer. Obtain an intermediate frequency signal, obtain a square wave signal by supplying the intermediate frequency signal output from the mixer to a first waveform conversion circuit, and obtain a predetermined demodulated signal by supplying the square wave signal to a restoration circuit. be able to.

In the above case, a part of the square wave signal is supplied to the intermediate carrier regeneration circuit, converted into a square wave again by the second waveform conversion circuit, and a pulse counter converts the signal to an integral multiple of the time slot of the digital pulse signal. Count square waves (intermediate carrier waves) for each period, and obtain the number of counts nf
By supplying T (where f is the intermediate carrier frequency at that point) to the subtraction circuit, the reference count number n
The carrier frequency of the intermediate frequency signal can be stabilized by obtaining the difference from fOT, converting this difference signal into an analog signal using a D/A converter, and supplying the analog signal to a variable frequency local oscillator.

Further, a signal subjected to phase modulation such that the time average value of the frequency is equal to the carrier frequency has a reference phase portion whose sum total is a predetermined time at the front and/or rear of one time slot of the digital pulse signal. In the first half of the remaining time slots, the phase is changed in a predetermined direction, and in the second half, the phase is changed in the opposite direction to the above phase change, corresponding to either the mark state or space state of the transmission signal. phase modulation is applied to restore the reference phase to the reference phase, and to correspond to the other state of the transmitted signal, to change the phase in the opposite direction to the above case, or to prevent the phase from changing at all. In this case, a square wave signal can be obtained by similarly stabilizing the carrier frequency of an intermediate frequency signal and performing waveform conversion, and then a predetermined demodulated signal can be obtained.

More specifically, an intermediate frequency signal is converted into a square wave, an intermediate carrier wave is regenerated from this square wave, and the number of pulses of the intermediate carrier wave within a predetermined time is counted while the intermediate carrier wave is converted into a square wave. However, it can be obtained by dividing the count number by the count time, so instead of using an analog frequency detector, the difference between the above count number and the reference count number is obtained, and the variable frequency local signal is calculated based on the difference signal. By controlling the oscillator, the carrier frequency of the intermediate frequency signal can be perfectly stabilized. With such a configuration, it is only necessary to set the reference count number, so that adjustment work such as that of an analog frequency detector can be omitted.

<Example> Hereinafter, embodiments will be described in detail with reference to the accompanying drawings showing examples.

FIG. 2 is a diagram illustrating a modulation method, and A1B in the same figure shows that reference phase portions of time tl and t2 are formed before and after the phase change section, respectively. However, the above-mentioned times tl and t2 are set so that tl + t2 - 8T (8T is a predetermined time shorter than one time slot T of the digital signal). Then, in the remaining period of the time slot, in the case of a mark, as shown in FIG.
The phase is changed to π), and the phase is changed to O at the final stage. Conversely, in the case of space,
As shown in the figure, the phase is changed in the opposite direction to that of the mark, that is, from 10 to 0.

Figure 2 C and D are the above times tl and t2 as tl - ΔTS
This shows the case where the phase is set to t2-0, and the phase change in the remaining portion is the same as in the case of FIGS. 2A and B above.

However, the characteristics do not change between the cases shown in FIGS. 2A and 2B and the cases shown in FIGS. 2C and D. That is, as long as the relationship tL+t2-ΔT is satisfied, the characteristics do not change, as will be clear from the explanation that will be given later.The cases shown in FIGS. 2C and 2D will be described below. However, the delay time TR is set to ΔT.

FIG. 2C is a diagram showing a mark signal, and the phase θ(1) is θ0 between 0≦t and △T (however, θ0-0 is shown in the figure), △T≦t and △ T+T”/2
Between 00g(t-△T), △T1T”/2≦
During t<T, it is expressed as θO×g (T−t).

In addition, the second diagram is a diagram showing the space, and the phase θ(1
) is θ0 between 0≦t and ΔT, and ΔT≦t and ΔT
Between +T-/2, θ〇-g (-△T), △T+
It is expressed as 00-g(Tt) between T″/7!≦t<T.

However, the above g (t) is a phase change function, and T'/
2 is 1/2 time of the portion other than the reference phase portion.

FIG. 1 is a block diagram of main parts showing one embodiment of the demodulation device of the present invention, in which a local oscillation signal (angular frequency mixer (2) supplied with ΩL)
At the same time, the signal output from the mixer (2) is supplied to the first Schmitt trigger circuit (5) through a bandpass filter (3) and an amplifier (4). and Schmitt trigger circuit (
The output signal from 5) is supplied to a phase detection circuit (6). A part of the output signal from the Schmitt trigger circuit (4) is passed through the intermediate carrier regeneration circuit (7) to the second Schmitt trigger circuit (8) and the gate circuit (
9), the pulse counter (10), the pulse count number reading circuit (11), and the pulse count number memory (12) to the subtraction circuit (13), and the difference signal from the subtraction circuit (13) is /A converter (14) and low-pass filter (15).
(1). Note that the subtraction circuit (13) is supplied with a reference pulse number signal from a reference pulse number memory (16). Further, the gate circuit (9) is supplied with a gate control signal corresponding to a period that is an integral multiple of the time slot of the digital pulse signal, and the pulse counter 00) is supplied with a gate control signal that corresponds to a period that is an integer multiple of the time slot of the digital pulse signal. A reset signal is supplied every double period, and the pulse count number reading circuit (11) is
A read control signal corresponding to the reset signal is supplied.

The intermediate carrier regeneration circuit (7) will be described in more detail with reference to FIG. changes. This intermediate frequency signal is then supplied to a phase detection circuit (71) to extract the modulated phase component.

Next, the extracted phase component is smoothed by a low-pass filter (72). Therefore, a DC component appears from the low-pass filter (72). Based on this DC component, V C
By controlling O(73), it is possible to reproduce an intermediate carrier wave with no phase change.

The operation of the demodulator having the above configuration is as follows.

First, the PSK modulated wave having the phase change characteristic shown in FIG. ) Zen d/
dt(Ωt+θ(t))■Ω+θ' (t).

Therefore, when the signal is a mark, the instantaneous angular frequency is ω(1)-Ω in the period 0≦tΔT, and ω(1)Ω−+g−(
t-ΔT), ΔT+T″/2≦t<t during the period ω
(1) Zen Ω-g"(T-t), and when the signal is a space, the instantaneous angular frequency is ω in the period of 0≦t and ΔT
(1)-Ω, ΔT≦t, ΔT+T-/2, ω(1) Ω-g-(t-ΔT), ΔT+T-/2≦t
In the period T, ω(1) Ω 0g-(Tt).

Then, when the signal is a mark, the average value ω of the instantaneous angular frequency within the time slot T is ω-(1/T), 1'ω(t)dt8T-(1/T)(j' ω( t) dtΔT+T″/2 +j ω(t)dt ΔT +j ω(t)dt ΔT+T=/2 −(1/T)[ΩT+ (g (T−/2)−g (0
))+ (g (0)-g (T-/2))]■Ω.

Furthermore, even when the signal is a space, the average value ω of the instantaneous angular frequency within the time slot T is Ω.

In other words, the time average value of the instantaneous angular frequency of the PSK modulated wave invented by the inventor of the present invention always matches the angular frequency Ω of the carrier wave, regardless of the signal content. It can be said that the PSK modulated wave is a completely new modulation method in that there is no guarantee that the time average value of the instantaneous angular frequency matches the angular frequency of the carrier wave.

Next, the operation of the demodulator of FIG. 1 under the above conditions will be explained.

A PSK modulated wave whose carrier wave has an angular frequency of ΩS is expressed as V C
By mixing with the signal whose angular frequency is ΩL output from O(1), it is converted to an intermediate carrier angular frequency Ω, and after noise components etc. are removed by a bandpass filter (3), the amplifier (4) The signal is amplified to a predetermined level by the first Schmitt trigger circuit (5), converted into a square wave signal, and the square wave signal is supplied to the phase detection circuit (6) to obtain the original digital data.

Further, a part of the output signal from the first Schmitt trigger circuit (5) is supplied to an intermediate frequency regeneration circuit (7) to regenerate an intermediate carrier wave with no phase change, and this intermediate carrier wave is transferred to the second intermediate frequency regeneration circuit (7).
The Schmitt trigger circuit (8) converts the signal into a square wave again. Then, the number of pulses within a predetermined period is counted by supplying it to the pulse counter 00) through the gate circuit (9), and this count value is read by the pulse count number reading circuit (11) and stored in the pulse count number memory (12). Store. Then, the difference between the count number stored in the pulse count number memory (12) and the reference count number stored in the reference pulse number memory (1B) is obtained by the subtraction circuit (13), and the difference is obtained from the D/A converter. (14)
After converting it into analog data, the digital error is smoothed by a low-pass filter (15) and the data is converted to analog data.
By supplying the signal to VCO(1), the oscillation frequency of VCO(1) is controlled so that the difference between the two counts becomes zero.

Therefore, the intermediate carrier angular frequency Ω output from mixer 0 can be stabilized at a preset angular frequency.

The above series of controls can be repeated for each repetitive operation by reading the count value of the pulse counter (10), resetting the contents of the pulse counter (10), and opening the gate circuit (9) again. Pulse count number memory (12
) can be updated to continuously stabilize the intermediate carrier angular frequency Ω.

By supplying the PSK modulated wave with a stabilized intermediate carrier angular frequency to the phase detector (6), stable phase detection can be performed without minimizing ΔT.

In the above embodiment, since phase detection is performed with the intermediate carrier angular frequency stabilized, it is possible to set 8T quite large, and the occupied frequency of the modulated wave In addition to making it possible to reduce the bandwidth, the video stage after detection can also be made free of high frequency components, increasing design tolerance and improving economic efficiency for the entire system. It will be possible.

Further, by converting the device into a digital circuit as described above, it becomes possible to convert the entire device into an IC.

It should be noted that the present invention is not limited to the above-mentioned embodiments. For example, the phase change portion is formed only in either the mark state or the space state of the transmission signal, and the phase change portion is formed in the other state of the transmission signal. It is also possible to demodulate a signal that has been subjected to phase modulation to form only a phase part, and within the scope of not changing the gist of the present invention,
Various design changes are possible.

<Effects of the Invention> As described above, the present invention receives a signal that has been phase modulated so that the time average value of the frequency is equal to the carrier frequency, then regenerates the intermediate carrier wave and performs digital feedback loop control. By mixing the signal from the variable frequency local oscillator and the received signal wave, an intermediate frequency signal with a stabilized intermediate carrier frequency is obtained. , it is possible to obtain the original digital signal with high precision, and furthermore, it has the unique effect of being able to prevent the occurrence of product errors.

[Brief explanation of the drawing]

FIG. 1 is an electric circuit diagram showing the main parts of an embodiment of a demodulator, FIG. 2 is a diagram illustrating an embodiment of a modulation method, FIG. 3 is a diagram illustrating a conventional delay detection device, and FIG. This figure is a diagram explaining the operation of the differential detection device of FIG. 3, and FIGS. 5, 6, and 7 are diagrams explaining the conventional modulation method. (1)...Voltage controlled oscillator, (2)...Mixer, (
5)...first Schmitt trigger circuit, (6)...
Phase detection circuit, (7) ... intermediate carrier regeneration circuit, (8
)...Second Schmitt trigger circuit, (10)...
Pulse counter, (13)...subtraction circuit, (14)...
...D/A converter, (16)...Basic pulse number memory Fig. 2 Fig. 3 (A) (C) 0 to 2 phase difference ΔH Fig. 5 Fig. 6

Claims (1)

[Claims] 1. Time average value of frequency is equal to carrier frequency A signal that has been phase modulated to make it more accurate. and the output from the variable frequency local oscillator A mixer that takes the signal as input, and from the mixer Waveform conversion is applied to the output intermediate frequency signal. a first waveform modification that outputs a square wave signal; converter circuit and intermediate carrier from the above square wave signal. An intermediate carrier regeneration circuit that regenerates the The second transformation converts the intercarrier into a square wave. Circuit and digital pulse signal time The period is n times the slot (where n is an integer, or a sufficiently large number other than an integer) A pulse counter that counts square waves, Enter the count from the pulse counter Subtraction to get the difference from the reference count number as circuit and the difference signal from the subtraction circuit. variable frequency local oscillator a D/A converter that supplies the first Square wave signal from conversion circuit as input A restoration circuit that restores the signal to a predetermined signal. A demodulator comprising: 2. Time average value of frequency is equal to carrier frequency A signal that has been phase modulated to make it more accurate. is one time period of digital pulse signal Total front and/or rear in lot The sum forms a reference phase part for a given time, and the transmission Mark state or space shape of transmitted signal The remaining state corresponds to one of the states. In the first half of the remaining time slot, While changing the phase in a predetermined direction, In the second half, the change in phase is opposite to the above phase change. The other side of the transmitted signal is restored to the reference phase. Corresponding to the state of or change the phase completely. In order to prevent the phase modulation from changing too much, The scope of the above claims as filed The demodulator according to item 1.