JPS5826862B2 - automatic equalizer - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (1) Field of Application of the Invention The present invention relates to a device that automatically corrects distortion that occurs in a transmission line when transmitting data, images, etc. over a telephone line.

(2) Prior Art A device that automatically corrects distortion in a transmission path, called an automatic equalizer, has been developed together with a data transmission modem (usually called a MODEM).

Conventionally known methods transmit impulses at known times, detect the sum of absolute values or sum of squares of code amount interference in the received signal that has been distorted on the transmission path, and adjust the frequency to minimize this. The parameters of the correction means are controlled so that the characteristics are variable.

However, the method of detecting the amount of transmission path distortion as the amount of code amount interference has the drawback of requiring complicated equipment.

On the other hand, the present inventors have previously filed an application for the following method as a transmission path distortion detection method.

Here, a step signal of amplitude E as shown in Fig. 1 is transmitted, and the maximum value f(t)m based on the median value S in the received signal is
a) (and detect the minimum value f(t)min.

Furthermore, find the difference between f(t)max and f(t)min,
The parameters of the correction means are controlled so as to minimize this.

Minimizing the above difference (ideally O) means completely correcting group delay distortion in transmission line distortion, which occurs in carrier lines where group delay distortion has a greater influence than amplitude distortion. This is an extremely effective method for easily detecting distortions in telephone lines, including distortions that occur.

However, in the above method, it is necessary to accurately set the reference value S at the center of the received step signal.

Further, different types of detection circuits are required for the maximum value and the minimum value.

These factors mean that the system is easily influenced by fluctuations in the received signal level or DC fluctuations in the circuit, which poses some problems in device implementation.

(3) Object of the Invention An object of the present invention is to provide an automatic equalization device equipped with means for simply and stably detecting an amount corresponding to distortion in a transmission path.

(4) Detailed explanation of the invention The difference between the maximum value at the rising edge and the maximum value at the falling edge (or the minimum value at the rising edge and the minimum value at the falling edge) of the received waveform is detected, and this is determined as the minimum value. The correction means having variable frequency characteristics is controlled so as to

(5) Examples Hereinafter, the present invention will be explained in detail with reference to Examples.

When a pulse signal is transmitted from the transmitting side at a known time, the received waveform P(t) accompanies vibration as shown in FIG.

This vibration is the sum of vibrations caused by the inherently finite band of the transmitted pulse signal and vibrations caused by distortion of the transmission path.

Assuming a transmission line with large distortion that requires automatic equalization, the vibrations in the received waveform are much larger due to the latter distortion.

Here, the maximum value of the received waveform including vibrations is determined by distinguishing between the rising and falling times and P(t)maxllP(t)m, respectively.
Let it be ax2.

Similarly, the minimum values are P(t)minl, P(t)m
Let it be in2.

In the present invention, the difference between these maximum values or between these minimum values is detected, and the correction means having variable frequency characteristics is controlled so as to minimize this difference.

As shown in Figure 2, each maximum and minimum value is determined based on the level S', but S/ is as shown in Figure 1 (
A feature of the present invention is that this does not need to be the center of the amplitude E of the received waveform, and may be any value.

Considering the telephone line, which is the most commonly used transmission line, most of the distortion in amplitude and group delay occurs when the telephone line passes through a carrier line.

Figure 6 shows the waveform (step response waveform) of the rising part of the pulse in the receiving section when the transmitted pulse is sent through an ideal low-pass filter and the number of carrier lines is O~5. It is something.

This number of carrier line links may be considered as the number of equalized links after equalization.

When equalization becomes excessive, a waveform that is symmetrical to the above waveform occurs at the trailing edge of the pulse.

When equalization is correctly performed, the waveform corresponding to the number of carrier line links of 0 becomes symmetrical with respect to the median value S at the rising and falling parts of the pulse as shown in FIG.

Therefore, for each number of links, the difference in the maximum value P(t)
max 1- P(t)max2 or minimum value difference P(
t)minIP(t)min2 is calculated as shown in FIG.

It is assumed that the pulse width W is long enough so that vibrations at the rise and fall of the pulse do not affect each other.

As will be described in detail later, it is assumed that the above-mentioned difference is detected from the received waveform after the distortion generated in the transmission path is corrected by the correction means.

Therefore, the difference between the distortion generated in the transmission path and the amount corrected by the correction means, that is, the residual distortion, corresponds to the number of links in Fig. 3,
A positive value indicates that the correction is insufficient, and a negative value indicates that the correction is excessive.

From FIG. 3, it can be seen that the above difference is positive when the correction is insufficient, negative when it is overcorrected, and the time difference when the correction is properly performed is O.

Note that the numerical values on the vertical axis in Figure 3 are just examples, and take different values depending on line distortion, but at least when the correction for group delay distortion is appropriate, the difference is O, and when it is insufficient it is positive and when it is excessive. It is generally true that it is negative.

Therefore, if the correction means is controlled to minimize the difference by utilizing the correspondence between the residual distortion after correcting the transmission path distortion and the above-mentioned difference, the residual distortion will also be minimized.

Note that f (t)max − f (t
)min also shows the same characteristics as in FIG. 3 with respect to the number of links.

However, in the method of FIG. 1, the reference level S must be the exact median value of the received step signal, and the maximum and minimum values must be detected.
The method of the present invention shown in the figure (here, the reference level S' may be any level with respect to the received pulse signal, and it is sufficient to detect either the maximum value or the minimum value). has a great effect on simplifying and stabilizing the circuit.

FIG. 4 (This shows an example of the configuration of the present invention.

A received signal is applied to terminal 1.

If the signal transmitted to the telephone line is a signal obtained by modulating a carrier wave, terminal 1 is given a signal obtained by demodulating this signal.

2.3, 4, and 5 are correction circuits having characteristics opposite to the distortion generated in the transmission path.

The example shown in FIG. 4 has a characteristic that is the opposite of the characteristic obtained by converting the distortion generated in the transmission path into baseband.

Also, as an example, 2 is 1 link, 3 is 2 links, and 4 is 3 links.
Link: Use a suitable correction circuit.

The optimum signal among the outputs of each correction circuit is selected by the switch circuit 6.

The maximum value of these signals is held in capacitors 11 and 12 via switch circuits 7 and 8 and diodes 9 and 10.

FIG. 5 shows operating waveforms of each part, and a) is a pulse signal received at a known time ζ among the output signals of the switch circuit 6.

b) is a signal that drives the switch circuit 13 and the capacitor 1
Erase the voltage of 1 in advance.

Furthermore, C) is a signal that opens the switch 7, and the maximum value for this period (P(t)max 1 in FIG. 2) is held in the capacitor 11.

On the other hand, d) is a signal that drives the switch circuit 14, and e) is a signal that drives the switch circuit 8. Similarly, after erasing the voltage of the capacitor 12, the maximum value of the period e) (P(t)ma in FIG. 2) is
x2) is held in the capacitor 12.

A comparison circuit 15 determines the polarity of the voltage difference between the capacitors 11 and 12.

Reference numeral 16 denotes a control circuit which calculates the falling time value of the signal of the output of the comparator circuit 15 (e.g., Fig. 5e) (i.e., P(
t)maxl-P(t)max2) is held, and the switch of the switch circuit 6 is operated so that if it is positive, the correction amount becomes larger, and if it is negative, it becomes smaller.

If the switch is fixed at the point where the polarity of the comparison circuit output changes, the switch circuit 6 is finally set to the correction amount that minimizes the difference between the voltages of the capacitors 11 and 12, and the optimum correction is achieved. The processed signal is output from the terminal 17.

In the example shown in FIG. 4, the difference between the maximum values at the rising and falling times is determined, but by reversing the polarities of the diodes 9 and 10, the method can be changed to a method in which the difference between the minimum values is determined.

Furthermore, although the example in the figure corresponds to quantizing the difference signal with one bit, the accuracy can be further improved by using multiple comparison circuits to quantize the difference signal with multiple bits and controlling the switch circuit 6 with this. It is also possible.

Alternatively, the difference signal may be treated as an analog signal and controlled using this signal.

In addition, although we have described an example in which the optimal correction circuit is selected from among multiple circuits with different characteristics, it is also possible to use a system in which the parameters that determine the characteristics (for example, the values of resistors and capacitors) are varied in one correction circuit. good.

Although FIG. 4 describes an example in which the signal after demodulation is corrected, the signal before demodulation may be corrected as shown in FIG. 6.

That is, a received signal is obtained at the terminal 18, and this is sent to the correction circuit 1.
9 and then demodulated by the demodulation circuit 20 and output to the terminal 17.

Further, the characteristics of the correction circuit 19 are controlled to optimal characteristics through a circuit 21 that detects the difference between the maximum value (or minimum value) at the rising edge of the demodulated signal.

(6) Summary As explained above, in the present invention, the difference between the maximum value or minimum value at the rising edge and the falling edge of the received pulse signal is detected, and the frequency characteristics of the correction means are changed so as to minimize this detected value. I'm letting you do it.

Therefore, compared to the prior art, the detection circuit is simpler and more stable.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing the detection principle of the conventional method, FIGS. 2 and 3 are diagrams showing the detection principle of the present invention, FIGS. 4 and 6 are diagrams showing a configuration example of the present invention, and FIG. 6 is a diagram showing the waveforms of each part, and FIG. 6 is a step response waveform diagram when the number of links of the carrier line is changed.

Claims (1)

[Claims] 1. A means for correcting distortion occurring in a transmission path, and a maximum value obtained at the rising edge and a maximum value obtained at the falling edge (or a minimum value obtained at the rising edge and σ at the falling edge).
an automatic equalizer comprising: means for detecting a difference between the minimum values obtained; and means for controlling characteristics of the correcting means so as to minimize the detected value. 2. In the G described in Paragraph 1, the correction means is composed of a plurality of correction circuits installed in parallel corresponding to the number of links of the carrier line of the transmission line and a switch circuit for switching each output, and has the above-mentioned characteristics. The means for controlling is an automatic equalizer configured to connect the switch circuit to a position where the detected value is minimized. 3. In the description of item 1, the correction means is configured with a single correction circuit having variable parameters, and the means for controlling the characteristics is configured to control the parameters so as to minimize the detected value. Automatic equalizer. 4. The automatic equalization device according to item 1, wherein the means for correcting distortion is configured to correct the signal after demodulation. 5. The automatic equalization device according to item 1, wherein the means for correcting distortion is configured to correct the signal before demodulation.