JPS6134303B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an amplitude fluctuation detector for detecting amplitude fluctuations used for stabilizing signal amplitude in a receiving and demodulating device of a digital multiphase multilevel carrier wave transmission system.

Among the methods for transmitting digital signals using a carrier band, the multi-phase multi-level modulation method, in which 16-value signals are arranged in a grid pattern on the signal space as shown in Figure 1, has high transmission efficiency and has the advantages of a hardware configuration. This is an advantageous method. This type of demodulator is composed of a carrier wave regeneration circuit for performing synchronous detection, a signal regeneration circuit for regenerating a binary four-series signal from a multilevel signal, a clock synchronization circuit, and the like.

In this demodulator, it is generally necessary that the amplitude of the input signal to the demodulator be constant in order to maintain the discrimination threshold at an optimum value. If it is incomplete to keep this amplitude constant, this will equivalently result in a fluctuation of the identification threshold, which will eventually cause a deterioration of the bit error rate. Conventionally, the input signal of the demodulator is detected to detect the signal amplitude, and this is used as a control signal to keep the output amplitude of an amplifier connected to the input side of the demodulator constant. However, when this received modulated signal transitions through the 16 possible states on the amplitude and phase planes, the signal wave states are not uniformly distributed on the amplitude and phase planes due to fluctuations in the signal amplitude. If there is no such signal, if the signal amplitude is detected by the conventional means described above, the detection level of the amplitude will fluctuate in response to the transmitted pulse pattern.

SUMMARY OF THE INVENTION An object of the present invention is to provide an amplitude fluctuation detector for detecting fluctuations in the input amplitude of a demodulator in a digital multiphase multilevel carrier wave transmission system, which eliminates such drawbacks.

Signal regeneration methods in demodulators for digital multi-phase multi-level carrier wave transmission systems mainly include a carrier band processing method and a baseband signal processing method. Here, we will discuss the case where the present invention is applied to the latter method. I will explain about it.

To give an overview of this latter playback method, the first
The detected signal obtained by detecting the signal shown in the figure with two mutually orthogonal regenerated carrier waves has four values -L ₂ , -L ₁ , L ₁ , and L _2, respectively, as shown in Figure 2A. .
The positive and negative polarities of this four-level signal are determined by first and second code discriminators, and two series of first pass signals as shown in FIG. 2B are reproduced. Next, by subtracting this reproduced output signal from the above-mentioned detected signal,
A difference signal as shown in is obtained, the positive and negative polarities of this difference signal are determined by third and fourth code discriminators, and two series of second pass signals are further reproduced.

FIG. 3 shows an example of a baseband signal processing type receiving demodulator. A multi-phase multi-level modulated input signal 51 having a carrier frequency ω _c is applied to two branched detectors 2 and 3. On the other hand, the signal output 54 of the voltage controlled oscillator 17
and a signal 53 which has passed through the π/2 phase shifter 4 are applied to the detectors 2 and 3, respectively, to obtain four-level detection output signals 55 and 56. code discriminator 5,
6 inputs these detection output signals 55 and 56 respectively, determines whether the input signal is positive or negative, and outputs a constant voltage of positive polarity if positive, and a constant voltage of negative polarity if negative, for example +1V or -1V. do. This output signal is the first demodulated data signal sequence. Detector 2, 3
4-value output signals 55, 56 and code discriminators 5, 6
The binary output signals 57 and 58 are added to subtracters 7 and 8, respectively, and the signals 55 and 56 are
Signals 59 and 60 are obtained by subtracting 7 and 58.

These signals 59 and 60 are binary as shown in FIG.
becomes the demodulated data signal. The operation of the code discriminator and the form of the output signal are the same as those of the code discriminators 5 and 6 described above. The output signals 59, 60 of the subtracters 7, 8 and the output signals 61, 62 of the sign discriminators 9, 10 are applied to the subtracters 11, 12,
Signals 63 and 64 are obtained by subtracting the latter from the former. The output signals 58, 57 of the sign discriminators 6, 5 and the output signals 63, 64 of the subtracters 11, 12 are added to the multipliers 13, 14, and output signals 65, 66 proportional to the product of both are obtained. . The output signals 65 and 66 of the multipliers 13 and 14 are input to the difference circuit 15,
An output signal 67 is obtained by subtracting the latter from the former. The output signal 67 of the difference circuit 15 passes through the loop filter 16 and is applied to the voltage controlled oscillator 17 as a control signal.

Now, the transmitted multiphase multilevel modulation signal S(t) can be expressed as follows, with its carrier wave angular frequency being ω _c .

S(t)=√2{g ₁₁ (t) cosω _c t +g ₁₂ (t) sinω _c t+1/2g ₂₁ (t) cosω _c t +1/2g ₂₂ (t) sinω _c t} (1) Here gij(t) represents a signal pulse train. i,
j is 1 or 2, respectively.

gij(t)=+1 or -1 (2) For simplicity, assume that this signal is input to the receiving device without being subjected to any distortion on the transmission line. The output signal 54 of the voltage controlled oscillator 17 is sinω _c
If t+e, the output signal 55 of the detector 2 is S ₅₅ =k{g ₁₁ (t)+1/2g ₂₁ (t)}cose−k {g ₁₂ (t)+1/2g ₂₂ (t)}×sine (3 ) Here, k is the amplitude proportionality constant.

If e=0, then S ₅₅ =k{g ₁₁ (t) + 1/2g ₂₁ (t), so the output signal 57 of the sign discriminator 5 is
S ₅₇ =g ₁₁ (t) regardless of the polarity of the sign of g ₂₁ .
Similarly, for the output signal 58 of the sign discriminator 6, S ₅₈ =g ₁₂ (t). In subtractor 7 the signal S ₅₅
Subtract the signal S ₅₇ from. As a result, the output signal of the subtracter is as follows.

S ₅₉ = g ₁₁ (t) (kcose-1) + 1/2k ・g ₂₁ (t) cose-k{g ₁₂ (t) + 1/2g ₂₂ (t)}sine (4) Here, k = 1, In the case of e=0, S ₅₉ =1/2g ₂₁ (t), and when the sign discriminator 9 discriminates the polarity of this signal, the transmitted signal train can be demodulated. That is, the output signal 61 of the sign discriminator 9 is S ₆₁
=g ₂₁ (t). This means that the sign discriminator 10
The same is true for S 62 = g 22 and its output signal is S ₆₂ = g ₂₂
(t).

Now, based on the operation of each part mentioned above, subtracter 1
Calculating the output 67 of 5, we get S ₆₇ =ε(t)=−k{g ² ₁₁ (t) +g ² ₁₂ (t)+1/2g ₁₂ (t)・g ₂₂ (t)+1/2g ₁₁ (t) ×g ₂₁ (t)} sine+1/2{g ₁₂ (t) g ₂₁ (t)−g ₁₁ (t) g ₂₂ (t)} ×(kcose−1) (5) Here, equation (5) is It shows the phase comparison voltage characteristics,
The second term is considered to be noise related to the modulation pattern of the modulated wave. Since the voltage ε(t) of the output signal 67 is the control voltage of the voltage controlled oscillator 17, noise occurring between these voltages becomes jitter (hereinafter referred to as pattern jitter) in the reproduced carrier wave, causing noise in the reproduced signal. As a result, the bit error rate deteriorates. Also, the second term of equation (5) is, when e is small,
When the constant k is sufficiently close to 1, this term disappears and does not generate noise. Focusing on the constant k, we continued the explanation assuming k = 1 in both equations (4) and (5), but if we consider the case where k is far from 1, the signal
It is clear that the noise component increases in S ₅₉ , which has a serious negative effect on signal reproduction, and that noise occurs in S ₆₇ = ε(t), increasing pattern jitter. It is important in the demodulator shown in FIG. Specifically speaking, keeping k at 1 means that the output signals 5 of the detectors 2 and 3
5, 56 and the amplitude of the output signals 57, 58, 61, 62 of the sign discriminators 5, 6, 9, 10, that is, the amplitude V ₁ of the detected signal shown in FIG. 2A and FIG. 2B The purpose is to maintain a constant ratio with the sign discriminator output amplitude V ₂ shown in . If this ratio is not appropriate, that is, if there is amplitude fluctuation, the waveform of the difference signal will be distorted, as shown in FIG. 2D, compared to the normal waveform shown in FIG. 2C.

However, in conventional demodulators, the amplitude of the input signal is controlled mainly by using an input signal amplifier 1 with automatic amplitude control. With this alone, it is impossible to perform accurate control of fluctuations caused by pulse patterns, and it is impossible to absorb fluctuations in amplitude that occur in subsequent circuits.

The present invention provides an amplitude fluctuation detector that detects fluctuations in input amplitude and fluctuations in output amplitude of a sign discriminator, etc., in order to eliminate the above-mentioned drawbacks of conventional demodulators.

FIG. 4 shows an embodiment of the amplitude fluctuation detector according to the present invention, in which signals are taken out from each part of FIG. 3 and connected. Output signals 57, 58 of code discriminators 5, 6 and output signals 6 of code discriminators 9, 10 in FIG.
1 and 62 are added to adders 18 and 19, respectively, and output signals 69 and 70 of adders 18 and 19 are signals proportional to the sum of both signals.

S ₆₉ = g ₁₁ (t) + 1/2g ₂₁ (t), S ₇₀ = g ₁₂ (t) + 1/2g ₂₂ (t) These signals 69 and 70 and the output signals 63 and 64 of the subtractors 11 and 12 is added to the multipliers 20 and 21, and S ₆₃ = {g ₁₁ (t) + 1/2g ₂₁ (t)} × (kcose-1) - k {g ₁₂ (t) + 1/2g ₂₂ (t)}sine S ₆₄ = {g ₁₂ (t) + 1/2g ₂₂ (t)} × (kcose-1) + k {g ₁₁ (t) + 1/2g ₂₁ (t)} sine(6) Proportional to the product of both signals Output signals 71 and 72 are obtained. The output signals 71 and 72 are as follows: S ₇₁ = S ₆₉ ×S ₆₃ = {g ² ₁₁ (t) + g ₁₁ (t)・g ₂₁ (t) + 1/4g ² ₂₁ (t)} (kcose−1) − _{k _ _ _ _ _ _ _} t)}sine S ₇₂ = S ₇₀ ×S ₆₄ = {g ² ₁₂ (t) + g ₂₁ (t)・g ₂₂ (t) + 1/4g ² ₂₂ (t)} (kcose−1) +k{g ₁₂ ( t)・g ₁₁ (t)+1/2g ₁₁ (t)・g ₂₂ (t)+1/2g ₂₁ (t)・g ₁₂ (t) +1/4g ₂₁ (t)・g ₂₂ (t)}sine ( 7) Added to the sum circuit 22 to obtain an output signal 73 proportional to the sum of both.

S ₇₃ = S ₇₁ + S ₇₂ = {g ² ₁₁ (t) + g ² ₁₂ (t) + 1/4g ² _2
1 (t) +1/4g ² ₂₂ (t)} (kcose-1) +{g ₁₁ (t)・g ₂₁ (t)+g ₁₂ (t)・g ₂₂ (t)} (kcose-1) (7 ) where the signals g ₁₁ (t), g ₂₁ (t), g ₁₂ (t) and
g ₂₂ (t) has values of +1 and -1, and there is no correlation between the respective signals and the signals. By this, the second term of equation (7) {g ₁₁ (t)・g ₂₁ (t) + g ₁₂
(t)·g ₂₂ (t)} becomes zero when averaged over time. That is, the second term component is removed by the low frequency filter, and the output signal 74 becomes as shown in equation (8).

S ₇₄ = [g ² ₁₁ (t) + 1/4g ² ₂₁ (t) + g ²
₁₂ (t) +1/4g ² ₂₂ (t)〕×(kcose−1) (8) Since the value in brackets [ ] in equation (8) is a constant value, e〓0
Then, the signal S ₇₄ will have a value proportional to (k-1), and if the amplitude of the input signal fluctuates or the output amplitude of the sign discriminators 9 and 10 fluctuates, an output signal proportional to the fluctuation will be obtained. I know that it will happen. A negative feedback loop is formed using the output signal obtained in this way, that is, the amplitude error signal as a control signal, and the signal
By controlling _S74 to zero, it becomes possible to control the demodulator shown in FIG. 3 in the best condition.

FIG. 5 shows a second embodiment of the invention, which is a simplified version of FIG. Multiplier 2 in Figure 4
An output signal 75 is obtained by removing high frequency components from the signal output 71 or 72 of 0 or 21 using the low frequency filter 24.

S ₇₅ = [g ² ₁₁ (t) + 1/4g ² ₂₁ (t)] × (kcose-1) or S ₇₅ = [g ² ₁₂ (t) + 1/4g ² ₂₂ (t)] (kcose-1) (9) Since the value in brackets [ ] in equation (9) is a constant value, e〓0
When , S ₇₅ becomes a value proportional to (k-1), and a voltage proportional to the amplitude error is detected as in the application example in FIG. 4.

FIG. 6 shows a third embodiment of the present invention.
The output signals 57, 58 of the sign discriminators 5, 6 and the output signals 63, 64 of the subtracters 11, 12 in the figure are added to multipliers 25, 26, respectively, and output signals 76, 77 proportional to the product of both are added. obtain. Signals 76, 77
is added to adder 27 to obtain an output signal 78 proportional to the sum of both. When the signal 78 is guided to the low frequency filter 28, the output signal 79 is as follows.

S ₇₉ = [g ² ₁₁ (t) + g ² ₁₂ (t)] (kcose
-1) (10) Here, in the case of e=0, since _S79 is proportional to (k-1), the same effect as in the case of FIG. 4 can be obtained.

FIG. 7 shows a fourth embodiment of the invention, which is a simplified version of the case shown in FIG. The output signal 76 or 77 shown in FIG.
get.

S ₈₀ = g ² ₁₁ (t) (kcose-1) or S ₈₀ = g ² ₁₂ (t) (kcose-1) (11) g ² ₁₁ (t) and g ² ₁₂ (t) are constant values. ,
e〓
In the case of 0, since S ₈₀ is proportional to (k-1), the same effect as in the case of FIG. 4 can be obtained.

FIG. 8 shows a fifth embodiment of the present invention;
The output signals 57, 58 of the sign discriminators 5, 6 and the output signals 59, 60 of the subtracters 7, 8 in the figure are the same as those of the multiplier 2.
5 and 26 to obtain output signals 81 and 82 which are proportional to the product of both. Signals 81 and 82 are added to adder 2
7 and is proportional to the sum of both.
The output signal 84 is obtained by removing high frequency components by the low frequency filter 28.

S ₈₄ = [g ² ₁₁ (t) + g ² ₁₂ (t)] (kcose
−1)(12) Here, since the value in brackets [ ] in equation (12) is a constant value,
In the case of e=0, the upper S ₈ ⁴ becomes a value proportional to (k-1), and the same effect as in the case of FIG. 4 can be obtained.

Figure 9 is a simplified version of the example in Figure 8,
The output signal 81 or 82 in FIG.
8 and removes high frequency components to obtain an output 85.

S ₈₅ = g ² ₁₁ (t) (kcose-1) or S ₈₅ = g ² ₁₂ (t) (kcose-1) (13) where g ² ₁₁ (t) and g ² ₁₂ (t) are constant When e=0, S ₈₅ is proportional to (k-1), and the same effect as in the case of FIG. 4 can be obtained.

The above amplitude fluctuation detector is used as shown in FIG. 10, for example. A digital multiphase multilevel carrier signal from an input terminal 100 is amplified by a signal amplifier 101 with an output amplitude control terminal, and the amplified output is demodulated by a conventional demodulator 102 shown in FIG. 1 and output. An amplitude fluctuation is detected by the signal supplied to the terminal 103 and taken out from a part of the demodulator 102 by the amplitude fluctuation detector 104 shown in any one of FIGS. 4 to 9 described above, and its output is as follows. The signal is applied to the control terminal of amplifier 101.

As described above, according to the amplitude fluctuation detector of the present invention, the amplitude fluctuation of the input signal of the conventional multi-phase multi-level carrier transmission type receiving demodulator and the fluctuation of the output amplitude of the code discriminator can be detected by the modulation pattern of the input signal. can be detected without being affected by Therefore, by using this as a control signal for level stabilization, it is possible to significantly improve the characteristics of the demodulator.

[Brief explanation of the drawing]

Fig. 1 is a diagram showing a multi-phase multi-value signal space arrangement, Fig. 2 is a diagram showing signal waveforms of each part in a demodulator, Fig. 3 is a block diagram showing a conventional baseband processing type demodulator, and Fig. 4 is a diagram showing a conventional baseband processing type demodulator. 9 to 9 are block diagrams showing embodiments of the amplitude fluctuation detector according to the present invention, and FIG. 10 is a block diagram showing an example of a multiphase multilevel carrier demodulation device using the amplitude fluctuation detector according to the present invention. be. 1: Amplifier, 2, 3: Detector, 4: π/2 phase shifter, 5, 6: Sign discriminator, 7, 8: Subtractor, 9,
10: sign discriminator, 11, 12: subtractor, 13,
14: Multiplier, 15: Subtractor, 16: Loube filter, 17: Voltage controlled oscillator, 18, 19: Adder, 20, 21: Multiplier, 22: Adder, 23,
24: Low frequency filter, 25, 26: Multiplier, 27:
Adder, 28: Low frequency unit.

Claims (1)

[Claims] 1. An input digital multi-phase multi-level carrier signal is detected by first and second detectors using two mutually orthogonal local oscillation signals, and each output of the first and second detectors is Each of the first and second sign discriminators determines whether it is positive or negative, and the determination outputs of the first and second sign discriminators and the outputs of the first and second detectors are subjected to first and second subtraction, respectively. The outputs of the first and second subtracters are determined to be positive or negative by the third and fourth sign discriminators, respectively, and the demodulated output is output from the first, second, third, and fourth sign discriminators. In a digital multiphase multilevel carrier demodulation device for obtaining the first code discriminator output signal, and the output signal of the first subtractor or the difference signal between this output signal and the output signal of the third code discriminator, What is claimed is: 1. An amplitude fluctuation detector, comprising: multiplying by a multiplier, and detecting fluctuations in input signal amplitude or amplitude fluctuations in a sign discriminator output signal from the multiplication output.