JPS6149860B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention relates to a compression/expansion method for digitally modulated waves, in which the analog equivalent value of the digitally modulated wave is level-compressed and transmitted on the transmitting side, and the analog equivalent value of the transmitted digitally modulated wave is compressed by the amount of level compression on the receiving side. By level expansion, it is possible to transmit digitally modulated waves with a quality higher than the conversion accuracy of AD converters and DA converters, and by using a signal prediction circuit, there is no need to transmit control signals indicating level compression information. It is an object of the present invention to provide a compression/expansion method that can transmit digital modulated waves using a smaller number of bits than conventional methods. FIG. 1 shows a block system diagram of an example of a conventional digital modulated wave compression/expansion method. In the figure, the analog signal entering input terminal 1 is passed through low-pass filter 2.
After removing unnecessary high-frequency components, the signal is supplied to the sampling and holding circuit 3, where it is sampled and held. The temporally discretized sampled signal taken out from the sampling hold circuit 3 is passed through a variable gain device 7, which will be described later.
While being applied to the AD converter 8, it is also supplied to the absolute value circuit 4, where the absolute value is taken. The output signal of this absolute value circuit 4 is supplied to a voltage comparator 6,
Here, it is compared with the reference voltage from the reference voltage setter 5, and when it exceeds the reference voltage, the gain of the variable gain unit 7 is varied. FIG. 2 shows the input voltage versus output voltage characteristics of the variable gain device 7 in which the above gain control is performed. As can be seen from the figure, the absolute value of the input voltage increases,
When the input voltage reaches the maximum or minimum that can be transmitted by the AD converter 8, the gain is varied by the gain control signal from the voltage comparator 6, and as a result, the absolute value of the output voltage is attenuated. I am forced to do it.
As a result, the output voltage of the variable gain unit 7 is always within a predetermined voltage range (indicated by ±1 in Figure 2) determined by the number of bits of the AD converter 8, and a level compression waveform as shown in Figure 3 is generated. Become. The output voltage of the variable gain device 7 is supplied to the AD converter 8, where it is analog-to-digital converted (specifically, quantized and encoded) into a digital modulated wave such as a pulse code modulation (PCM) signal. It is said that This digitally modulated wave is supplied to the DA converter 9 through a predetermined transmission path, where it is converted into a digital signal.
After analog conversion, the level attenuation or level enhancement in the transmission system is restored by level expansion by the variable gain unit 10, and then restored to the original analog signal by the low-pass filter 11 and sent to the output terminal 12.
It is output from However, in the conventional digital modulated wave compression/expansion method described above, the variable gain unit 10 on the receiving side controls the gain in the opposite direction and the same amount as the gain controlled by the variable gain unit 7 on the transmitting side. Since the variable gain control signal also had to be transmitted at the same time as the transmission signal, the demand for reducing the number of transmission bits was limited. The present invention eliminates the above drawbacks, and the fourth
Each embodiment will be described with reference to the drawings below. FIG. 4 shows a block system diagram of a first embodiment of the digital modulated wave compression/expansion method according to the present invention.
In the figure, the same components as those in FIG. 1 are given the same numbers, and their explanations will be omitted. In Figure 4, AD
The digital modulated wave taken out from the converter 8 is
The signal is supplied to a variable gain unit 13 (which handles digital signals and is composed of a shift register or a multiplier), and its analog conversion level is varied by a control signal from a gain control unit 14, which will be described later. The signal is then transmitted and simultaneously supplied to the signal prediction circuit 15 in the gain control section 14. Now, the analog conversion level of the input digital modulated wave at time nT of the variable gain unit 13 (T is the sampling period of the sampling and holding circuit 3) is X _o , the gain of the variable gain unit 13 at that time is G ₁ , and the variable gain unit 13 time nT
Letting the analog conversion level of the output digital modulated wave at y _o be the following equation. _{yo = _} It is said that the structure consists of The above sampling frequency converter includes an interpolator to which an input signal y _o is supplied, and an output signal of this interpolator to which the input signal y o is supplied;
The filter includes at least a first digital filter that operates at a sampling frequency that is L/M times higher. The above interpolator is a circuit that sequentially inserts L-1 atmosphere points every period k within the time interval T of each sample value of the input signal y _o , and its output signal V _(oL/M)+k is It is expressed by the formula.

[Table] This signal V _(oL/M)+k is supplied to the first digital filter, and only the signal components (frequency components below _s /2) of the input signal y _o are separated. Below, we will explain the case of M=1 as an example.
This state is shown together with FIGS. 9A to 9C and 10A to 10C. 9A and 10A respectively show the waveform and frequency spectrum of the input signal _yo , and FIGS. 9B and 10B show the waveform and frequency spectrum of the interpolated signal, respectively. FIG. 9C shows the analog converted waveform of the output signal of the digital filter operating at the sampling frequency L times the above, and its frequency spectrum is the first
The result will be as shown in Figure 0C. The analog conversion level of the output signal of the digital filter obtained in this way is Z _(oL/M)+i ,
When the weighting coefficient assigned in the signal prediction circuit 15 is a _i (i=0 to N), the signal prediction circuit 15
teeth A digital signal expressed by the following formula is output.
Equation (2-2) is also an equation representing the output digital signal of a kind of FIR type digital filter. Therefore, the signal prediction circuit 15 described above includes a sampling frequency converter, an absolute value circuit to which the output signal V _(oL/M)+ik of the sampling frequency converter is supplied, and an output signal of the absolute value circuit to which the output signal of the absolute value circuit is supplied. It can be configured with an FIR type digital filter (second digital filter). Note that in formula (2-2) |
V〓 _+ik | represents a signal obtained by passing the digital modulated wave V〓 _+ik through an absolute value circuit. As the simplest example of the signal prediction circuit 15, M=
1, a ₀ = 2, a ₁ = -1, a ₂ = a ₃ =...= a _o = 0, it can be configured as shown in Fig. 5, and the digital modulation input to the input terminal 20 After the sampling frequency of the wave y _o is multiplied by L by the sampling frequency converter 21, the absolute value is taken by the absolute value circuit 22, and then the absolute value of V _oL+i is multiplied by L by the subtractor 23. The output of the delay device 24 which is delayed by the sampling period T _L is the digital modulated wave V _oL before one sample.
By subtracting the absolute value of _+i-1 to obtain a difference signal and adding it to the absolute value of V _oL+i in adder 25, the analog predicted to arrive at time (n+1) T _L is calculated. The digital signal z _oL+i with the conversion level is used as the predicted signal at the next time (n+1)T
It is output from the output terminal 26 until _{the L} digital modulated wave V _oL+i+1 comes in. The output prediction signal z _oL+i of the signal prediction circuit 15 is supplied to a comparator 17, where the level is compared with the reference level _lo from the variable reference level generator 16 in terms of an analog conversion value. As a result, when the digital signal z _oL+i is at a higher level than the reference level l _o , the gain of the variable gainer 13 is lower than that of the comparator 17.
Lower _G1 (specifically, if variable gain unit 13 is configured with a shift register, shift in the LSB direction; if configured with a multiplier, reduce the coefficient), and A signal for changing the reference level of the level generator 16 to a new reference level lowered by the same value is output. On the other hand, when the digital signal z _oL+i is at a lower level than the reference level l _o , the comparator 17 increases the gain G ₁ of the variable gain unit 13 (specifically, the variable gain unit 13 is composed of a shift register). If it is configured with a multiplier, shift it towards the MSB, or increase the coefficient if it is configured with a multiplier),
In addition, a signal for changing the reference level of the variable reference level generator 16 to a new reference level that is higher by the same value is output. Note that the output of the comparator 17 is also supplied to the signal prediction circuit 15. However, upon obtaining the prediction signal, the gain control section 14 operates at T _L and the variable gain unit 13 operates at T. In this way, the gain G ₁ of the variable gain unit 13
The digitally modulated wave _{y o} obtained by gain control (level control of the analog conversion level) by the output signal of the gain control unit 14 based on the predicted signal, the lowest level of quantization increases when the change in the analog conversion level is large. On the contrary, when the level change is small, the lowest level of quantization is lowered, and the signal becomes equivalent to a signal that has been subjected to finer quantization. The digitally modulated wave described above is received through a desired transmission path, is supplied to a variable gain unit 18 having the same configuration as the variable gain unit 13, and at the same time is supplied to a gain control unit 19. The variable gain unit 18 has the same configuration as the variable gain unit 13, and the analog conversion level of the supplied digital signal is variably controlled by the output of the gain control unit 19. Here, the variable gain device 18
When the analog equivalent level of the input digital modulated wave at time nT is p _o , the analog equivalent level of the output digital modulated wave at time nT is q _o , and the gain of the variable gain unit 18 is G ₂ , the following equation is satisfied. q _o = p _o · G ₂ (3) On the other hand, the gain control section 19 is the same as the gain control section 14 described above.
Although the configuration is similar to that of the gain control section 14, the operation is opposite to that of the gain control section 14. That is, the gain control unit 19 uses a comparator (not shown) to output the digital modulated wave p.
When the analog equivalent level of the predicted signal obtained from _o is higher than the reference level, the above gain G ₂ is increased and the reference level is increased by the same value. In the opposite case, the gain G ₂ of the variable gain unit 18 is lowered and the reference level is changed to a reference level lower by the same value. That is, the gain control by the gain control section 19 performs a level expansion operation complementary to the gain control by the gain control section 14, so that the relationship G ₁ ·G ₂ =1 (4) is satisfied. Also, y _o ≒ p _o (5). The digital modulated wave (analog conversion level q _o ) extracted from the variable gain unit 18 is supplied to the DA converter 9 . Here, equation (6) is established from equations (1), (3), (4), and (5). q _o = p _o・G ₂ ≒y _o・G ₂ =x _o・G ₁・G ₂ =x _o ∴ q _o ≒x _o (6) Therefore, the level-compressed digital modulated wave on the transmitting side is On the receiving side, the level is expanded by the amount of level compression, and the signal is returned to the original digitally modulated wave. According to this embodiment, the above-mentioned levels of compression and expansion are performed based on the predicted signal. For example, in equation (2-2), M = 1, a ₀ = 2, a ₁ = -
1, a ₂ =...=a _o = 0, then z _oL+i = V _oL+i + (V _oL+i − V _oL+i-1 ) = V _oL+i + dv _oL+i /dt (7), A predicted signal is obtained by differentiation. Similarly, by performing appropriate weighting using equations (2-1) and (2-2), an appropriate predicted signal can be obtained depending on the nature of the input digital modulated wave. As can be seen from equations (2-1) and (2-2), this predicted signal can be obtained from the digitally modulated wave, so there is no need to separately transmit a gain control signal to the receiving side as in the conventional system. Therefore, the digital modulated wave can be compressed and expanded using a smaller number of bits than before. FIG. 6 shows a block system diagram of a second embodiment of the system of the present invention. In the figure, the same components as those in FIG. 4 are given the same numbers, and their explanations will be omitted. In this embodiment, a gain control circuit 27 for generating a variable gain control signal for the variable gain unit 13 is configured with the output digital modulated wave of the AD converter 8 as an input signal,
A gain control circuit 28 that generates a control signal for varying the gain of the variable gain device 18 is configured to obtain the control signal from the output digital modulated wave of the variable gain device 18. The gain control circuits 27 and 28 have the same configuration as the gain control circuits 14 and 19, and this embodiment also has the same features as the first embodiment. 7 is a block system diagram of a third embodiment of the system of the present invention, and FIG. 8 is a block system diagram of a fourth embodiment of the system of the present invention. The third and fourth embodiments described above are different from the first and second embodiments in that the variable gainers 7 and 10 receive the sampled signal as input. Therefore, y _o in equation (1), (2-1), (2-2)
V _oL+k and z _oL+i in the equations represent signal levels rather than analog conversion levels. Further, in the third embodiment, the gain control unit 29 generates a predicted signal from the output signal of the variable gain unit 7 by the method described above, compares it with a reference level, and when the predicted signal level is higher than the reference level, On the one hand, the gain of the variable gain device 7 is controlled to a small value, and on the other hand, the reference level is changed to a small value, and on the other hand,
When the gain is smaller than the reference level, the gain of the variable gain unit 7 is controlled to be large, and the reference level is also changed to be large. Further, the gain control section 30 shown in FIG. 7 has the same configuration as the gain control sections 19 and 28 and performs the same operation. Furthermore, in the fourth embodiment shown in FIG. 8, the gain control section 31 uses the output sampling signal of the sampling and holding circuit 3 as an input signal, and performs the same operation as the gain control section 29 to obtain a control signal to control the variable gain control section. 7, and the gain control section 32 uses the output analog signal of the variable gain device 10 as an input signal and performs the same operation as the gain control section 30 to obtain a control signal. The digital modulated wave is compressed and expanded by variable control of the gain. The third and fourth embodiments described above also have the same features as the first and second embodiments described above. In addition, in the first and second embodiments, the analog conversion level of the digital modulation wave is level-compressed and expanded, and in the third and fourth embodiments, the modulation signal (analog signal) of the digital modulation wave is level-compressed and demodulated. Although the signal is level-expanded, the desired purpose can be achieved by level-compressing and expanding the differential signal (m _o - _{m o-1} ) as well (however, m _o is the digital signal at time nT). (indicates the analog equivalent level of the modulated wave). This differential signal is also included in the digital modulated wave in this specification. It goes without saying that the sampling frequency converters in the gain control units 14, 27, 29, and 31 can increase the sampling frequency not only to L times the input signal y _o but also to L/M times the input signal yo. As described above, the digital modulated wave compression/expansion method according to the present invention includes a first variable gain device provided on the output side of an AD converter that digitally modulates an analog signal to obtain a digital modulated wave; Let the analog conversion level of the digital modulated wave at time nT taken from the output side (or input side) of the variable gainer of y _o (where T is the sampling period of the digital modulated wave) and the coefficient for weighting be a _i , when the analog conversion level of the output signal of the interpolator in the sampling frequency converter to increase the sampling frequency by L/M times is V _(oL/M)+k, V〓 _+k = {y _o (k= 0)
0(k=1,2,...,L-1)} (However, L, M, N are arbitrary natural numbers, a ₀ ~ a
_N is 0 or any arbitrary number) A predicted signal with an analog conversion level z〓 _+i that satisfies the formula is generated, and the reference level from the variable reference level generator and the analog conversion level Z〓 _+i of the above predicted signal are a first gain control unit that causes a first variable gain unit to compress the analog equivalent level of the input digital modulated wave within a predetermined level range using a control signal obtained by comparing the two; A digital modulated wave taken out from the second variable gain device is supplied through a transmission line and its analog conversion level is varied according to a control signal, and the digital modulated wave is taken out from the input side (or output side) of the second variable gain device. The second variable gain unit performs an operation opposite to that of the first gain control unit for the input digital modulated wave, and adjusts the analog conversion level of the input digital modulated wave to the level corresponding to the level compression within a predetermined level range. Since it is composed of a second gain control section for expansion and a circuit for demodulating the output digital modulated wave of the second variable gain device into the original analog signal, and when a variable gain device is provided on the input side of the AD converter. was configured in the same way as above, so the AD converter and
Even if the conversion accuracy of the DA converter is low, since the input level and output level are controlled, signal transmission that exceeds their performance can be ensured.Also, a predicted signal is generated and the level is compared with the reference level. 1st
and the gain control signal of the second variable gain device, there is no need to separately transmit the gain control signal as in the conventional method, and the number of transmission bits is reduced compared to the conventional method. can,
Also, if the number of transmission bits is the same as the conventional method,
The amount of information that can be transmitted can be increased compared to conventional methods, and digital modulated waves with a predetermined bit length, that is, PCM signals and DPCM signals, can be compressed and transmitted, so the number of transmission paths can be reduced, and the signal When the level becomes small, the minimum level of quantization is lowered to perform finer quantization, so even if the signal level is low, quantization noise can be reduced compared to the conventional method, and the apparent sampling frequency is lowered to L/ Since the predicted signal is obtained by multiplying the signal by M times, it has many advantages such as being able to substantially minimize the prediction error.

[Brief explanation of the drawing]

Figure 1 is a block system diagram showing an example of the conventional method, Figure 2 is a diagram showing the input/output characteristics of the variable gain unit on the transmitting side, and Figure 3 is an example of the output signal waveform of the variable gain unit on the transmitting side. 4, 6, 7 and 8 are block system diagrams showing respective embodiments of the system of the present invention, and FIG. 5 is a block system diagram showing an embodiment of the main part of the system of the present invention. 9A to 9C are respective signal waveform diagrams of various parts in the signal prediction circuit according to the present invention, and FIGS. 10A to 10C are diagrams showing frequency spectra of FIGS. 9A to C, respectively. 1...Analog signal input terminal, 7, 10, 13,
18... Variable gain device, 8... AD converter, 9... DA converter, 12... Analog signal output terminal, 14, 27,
29, 31... Gain control unit for level compression, 15... Signal prediction circuit, 16... Variable reference level generator, 17
...Comparator, 19, 28, 30, 32... Gain control section for level expansion, 21... Sampling frequency converter, 25... Prediction signal output terminal.

Claims (1)

[Claims] 1. A first variable gain device provided on the output side of an AD converter that digitally modulates an analog signal to obtain a digital modulated wave; The analog conversion _level of the digital modulated wave at time nT taken from
(T is the sampling period of the digital modulated wave), the coefficient for weighting is a _i , and L/M
When the analog conversion level of the output signal of the interpolator in the sampling frequency converter to double the sampling frequency is V _(oL/M)+k , [Table] (However, L, M, N are arbitrary natural numbers, a _O ~
a _N is O or any number, L>M) Analog conversion level Z that satisfies the formula _(oL/M)+
i , and the control signal obtained by comparing the reference level from the variable reference level generator with the analog equivalent level Z _(oL/M)+i of the above-mentioned predicted signal, controls the first variable a first gain control section that causes the gain unit to compress the analog equivalent level of the input digital modulated wave within a predetermined level range; and the digital modulated wave taken out from the first gain control section is supplied through a transmission line. a second variable gain device that varies the analog conversion level thereof according to a control signal; and a first gain control for a digital modulated wave taken out from the input side (or output side) of the second variable gain device. a second gain control section that performs an operation opposite to that of the second variable gain unit to expand the analog equivalent level of the input digital modulated wave by the level compression amount in a predetermined level range; A digital modulated wave compression/expansion system comprising a circuit for demodulating the output digital modulated wave of the second variable gainer into an original analog signal. 2. A first variable gain device installed on the input side of an AD converter that digitally modulates an analog signal to obtain a digital modulated wave, and the time taken out from the output side (or input side) of the first variable gain device. The sampling signal level at nT is y _o (where T is the sampling period of the digital modulation wave), the weighting coefficient is a _i , and interpolation within the sampling frequency converter is performed to increase the sampling frequency by L/M times. When the level of the output signal of the device is V _(oL/M)+k , [Table] (However, L, M, N are arbitrary natural numbers, a _O ~
a A predicted signal with a level Z _(oL/M)+i that satisfies the formula (where _N is O or any arbitrary number, L>M) is generated, and the reference level from the variable reference level generator and the above predicted signal level Z are generated. _(oL/M)+i , and the first variable gain unit compresses the level of the analog modulation signal of the input digital modulation wave within a predetermined level range. a gain control unit; a second variable gain unit in which the digitally modulated wave taken out from the first gain control unit is sequentially supplied through a transmission line and a DA converter and whose signal level is varied in accordance with a control signal; , the analog signal taken out from the input side (or output side) of the second variable gain unit is operated in the opposite manner to that of the first gain control unit, and the second variable gain unit The second gain control unit expands the level of the input analog signal by the amount of level compression within a predetermined level range, and the circuit demodulates the output signal of the second variable gainer into the original analog signal. Features a compression/expansion method for digitally modulated waves.