SU637971A1 - Electric signal digital transmitting-receiving method - Google Patents

Description

one

The invention relates to radio engineering and can be used to digitally transmit analog electric signals over digital lines of communication, in particular for the transmission of television signals.

There is a method of digital transmission and reception of electrical signals, in which on the transmitter side an analog signal is differentiated, sampled by time, quantized by level, CODIED.T and transmitted over a communication channel, and decoded on the other side, converted into an analog signal and integrated 1.

However, the known method is characterized by low noise immunity.

The aim of the invention is to improve noise immunity.

For this, in the method of digital transmission and reception of electrical signals, in which on the transmitting side the analog signal is differentiated, sampled by time, quantized by level, encoded and transmitted over the communication channel, and at the receiving side is decoded, converted into analog signal and integrated into to the transmitting side, the signal is differentiated, transformed according to Hilbert at the finite specified time interval, and at the receiving side, after integration, the received signal is transformed according to Hilbert at the finite time interval equal to or more than the specified final interval for the Hilbert transform on the transmitting side, and compensate for the echo-signals that appear when the signal is converted by the Hilbert, while the Gilbert transform is differentiated and transformed on the transmit side of the H side of the integration and transformed by the Gilbert side produced in any sequence over an analog or digital signal.

FIG. 1 and 2 are block diagrams of a device for implementing a method for digitally transmitting and receiving electrical signals using analog circuits, and in digital form; in fig. 3 rd diagram, explanatory

5 way qi (s) of the transmission and reception of electrical signals.

The device shown in FIG. 1 contains —transmit delay line 2 connected to input 1,

0 providing a signal delay on

time multiple to the Kotelnikov interchange. The central branch of this line is connected directly to the matrix 3, and the remaining branches are connected to the matrix via inverters 4 and 5. Matrix 3 is connected via analog-to-digital converter 6 JK to the equipment of communication channel 7, from the output of which the digital signal is fed to the digital-to-analog interface. 8. A second delay line 9 is connected to the output of this converter; it has several taps, which also provide a signal delay for a time multiple of the Kotelnikov interval. The outputs of this delay line are connected to the matrix 10, connected to the input of the delay line 11, the central outlet of which is connected directly to the matrix 12, and its two outermost leads are connected to the same matrix via inverters 13 and 14. The output 15 of the matrix 12 is the output of the device .

The device shown in Fig. 2 contains, at the input 16, an angular digital converter 17 connected via memory unit 18 and a computing unit 19 to the communication channel equipment 20, the output of which is fed through a second memory unit 21 and a second computing unit 22 to digital-to-analog converter 23. This converter is connected to an echo equalizer 24, its output 25 is the output of the device

The technical essence of the method of digital transmission and reception of electrically signals (let's call this method of Hilbert-differential pulse code modulation - CDMA) can be explained by the example of the transmission of an elementary signal with limited spectrum.

Si4 ft

Ш1тс 4

where T is the Kotelnikov interval for the signal f (t).

This signal, in accordance with Kotelnikov's theorem, can be determined by one sample a 1 at time t o. All the remaining samples Oj of this signal, taken over the Kotelnikov interval, are equal to zero. The elementary signal, transformed according to Hilbert and differentiated, can be written in the form:

009W 4C-i-V T)

1C

5 j

where OL is catfish

V - fib

214

one

g

to) and b

S211-1G

mc

- discrete counts, determined by the signal S (t).

At the same time, if we limit the interval of time at which the indicated transformation is carried out, then, without taking into account the constant multiplier ot, the waveform can be represented as

. - 1s

x.a) -ft

151n Y-t42n-m

(2h-ir: к.г,, 1

 t-wn-otj

in this case, the specified transformation is performed on the interval t 2 (2M-1), T, N is the number that determines the duration of the transformation interval.

With a conversion interval duration, only three samples are used;

Vir- -i--o.

These samples are digitally transmitted over the communication channel.

On the receiving side, the inverse transform should be performed. In order to reduce the impact of digital signals in a communication channel, this inverse transformation must also be carried out on a finite interval. In this case, the signal samples after the inverse transformation should represent a linear combination of samples transmitted over the communication line:

.K.Where Hc are constant coefficients.

The determination of XK coefficients can be carried out in various ways. As an example, consider the option of accurately reconstructing the input signal f (t) over the interval t 2MT.

In this case, the ratios and where must be satisfied. Then the coefficients X are solutions of the system equations:

0.5 X - 0.2 X. ,, 0;

0.5 - 0.2 (X. +) O - M +1 M $ -1;

0.5 Ho - 0.2 (X., + X) 1;

(Hl

(V-i mm) 0

0.5 0.2;

0.5 Hd ;, - 0,2 X

It is obvious that X „X- ,,

This inverse transform provides a significant reduction in the effect of digital errors arising in the channel on the output signal as compared to systems using DPCM. As stated above, for example, in digital television systems with a DPCM, a single digital error appears as a pedestal distortion of the entire subsequent portion of the string. In the considered variant, a single digital error in the communication channel will lead to the creation of a signal interference, the value of which is determined by the coefficients Xm and the duration by the selected inverse transform interval.

When transmitting the signal f (t) at the indicated transformations, a signal is formed at the output, defined by three samples and characterized by the appearance, in addition to the main signal, of two echo signals, whose span is equal to C C. x .. So, for example, if the interval is transformed to be equal, then Xd 3,231; X XI l, 538; X.j X-j 0.615 and the echo span is C 2. 0.123.

In order to isolate the undistorted signal f (t), it is necessary to carry out the correction of the resulting echoes. If the interval to convert is equal, then X is 3.334; X X. 1,667; Xj X.2 0.834; X X 0.417; X4 X-4- 0.208; Xj Xly 0.103; X4. X 0.049; X-, X.t 0.020 and the echo swing is C p. 0.004.

In this case, the amount of echoes is so small that their correction may not be required.

Taking into account that an electrical signal with a limited spectrum is a linear combination of elementary signals f (t) (Fig. 3a), shifted in time by intervals that are multiples of the Kotelnikov interval, for understanding the reasoning we consider the conversion of this signal in a device whose structure is shown in figure 1. The elementary signal f (t) is determined by only one sample a0 1 (Fig. 36).

From the output 1 of the device (Fig. 1}, the electric signal is fed to a delay line 2 having several taps, which provide a delay of the input signal for a time multiple of Kotelnikov’s interval. The central taps of this line are connected directly to the matrix 3. One or more taps of line 2, providing the delay of the signal relative to the central retracement by the time t (2ts-1) T, n 1, 2, ..., N, multiple of an odd number

(Katelnikov’s intervals, connected to matrix 3 via inverters 4. The number N determines the interval at which the input signal is converted. One or several 5 taps of line 2, which provide an advance of signals relative to the center taper for time C (2, -1)

T, p 1,2,, N, are connected to Matria 3 through inverters 5. Matrix Q 3 provides the summation of the signals arriving at it with the corresponding weights: the signal coming from the central tap is summed with the weight b 0.5, delayed and. which is equal to the Kotelnikov interval, the inverted signals are summed with BecaNM | b. |

2 .0,2, delayed and advanced

for a time equal to Kotelnikov's threefold Q interval, the inverted signals are summed with the weights of 1 0.023, etc.

Thus, the converted by is formed at the output of the matrix 5 3.

Gilbert and differentiated input signal on a finite interval. His readings are shown in FIG. Sound This signal is then fed to an analog-to-digital converter.

D 6, from the output of which the digital codes enter the equipment of the communication channel 7. Dedicated codzes at the output of the channel are converted again to an analog signal using an analog converter 8. Next, the signal arrives at a delay line 9 having a central tap and taps providing delay or advance of signals relative to the central tap for a time multiple of Kotelnikov interval. The number of taps of this line determines the interval at which the inverse transform is performed. From the outputs of the delay line, signals are fed to the matrix 10, which provides their summation with certain coefficients. Moreover, the coefficient of the matrix signal coming from the central branch, must be equal to X. Signals

The 0 lagging and leading signals taken from the central branch, for the time tr tT, are summed up respectively with the weight X (). These coefficients determine the counts of 5 mechs (Fig. 3d), arising from a single and digital error in the communication channel. The signal from the output of the matrix with samples of FIG. The rear differs from the original transmitted signal and contains echoes. To eliminate

Claims (2)

These echoes use an echo-corrector, in the simplest form made in the form of a delay line 11 with three taps — the central of which is connected to matrix 12 not mediated, but the other two are connected to this matrix via inverters 13 and 14. Signal delay time at the inputs of the respective inverters 13 and 14, relative to the signal at the center tap, is chosen equal to T CT, where M is the coefficient determining the inverse transform interval. Matrix 12 provides a summation of the center-out signal with a matrixization factor of 1 with the signals received by the G inputs of inverters .13 and 14. Coeff (1x1 matrixes of matrixing these signals are X lbwvl. At output 15 of matrix 12, the desired signal is selected, defined by from the report of Fig. Ze. The implementation of the method for digital conversion of a signal is explained using the block diagram of the device shown in FIG. 2. The electrical signal (Fig. 3a) to be transmitted over the digital communication channel from the input 16 is fed to the analog-digital converter 17 which provides for the selection of signal samples through the Kotelnikov interval (Fig. 3b) and their conversion into a digital form. The converted signal samples are successively fed to memory block 18, which stores 4N-1 current samples. Thus, the current signal samples are stored in the interval t 2 (2N-1) T. From the outputs of the memory block, digital signals are supplied to the subtractor 19, which provides the summation of the magnitude of the central sample with a factor of 5 and the values of the signal samples shifted relative to the price / neutral sample by time | t | (2f, -1) T, with coefficients b, b, n2 stDug nl, 2, ..., N (fig. 3 communication, from the output of which digital codes are sent to another block of memory 21 on 2M + 1 current bills. Thus, the current digital samples of the signal are stored on the interval T 2MT. From the outputs of block 21 of memory, digital samples are sent to the block 22 calculations, providing the summation of the magnitude of the central reference with the coefficient Ho and in signal counts offset from the central reference by time t tT, with coefficients m 1,2, M (Fig. 3g). At the output of block 22, the signal is numerically digitized (Fig. Back), which is again converted to analogue signal With the help of a digital-to-flash converter 23. If necessary, correction of the echo signal is performed in block 24, at the output 25 of which the original electrical signal is extracted (Fig. Ze). The proposed method of transmitting and receiving signals allows to significantly reduce the effect of digital errors in the communication channel. The use of the proposed method, for example, in the transmission of television signals, reduces the requirements for transmission reliability in a communication channel by about an order of magnitude to differential PULSE-CODE modulation. This FIRE will use the proposed method on communication lines with energized energies, for example, on communication lines and satellites. A method of digital transmission and reception of electrical signals in which an analog signal is differentiated on the transmitter side, time sampled, a quantum is level, encoded, and transmitted over a communication channel, and decoded on the receiver side, converted into an analog signal and integrated , characterized in that, in order to improve the noise immunity, after transmitting the analog signal on the transmitting side is transformed according to Hilbert at the final specified time interval, and on the receiving side after integrating The received signal is transformed according to Hilbert at a finite time interval equal to the final specified time interval for conversion according to Hilbert on the transmitting side, and co-accumulate the echo signals arising from the conversion of the signal according to Hilbert. Sources of information taken into account during the examination 1. Katermol K.V. Principles of pulse code modulation. M., Holy Hour, 1974, p. 189 (prototype). Well “A t g T but About mg.5 at 04 IS nineteen ZS 2