JPS6041888B2 - control circuit - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a control circuit that performs compensation for returning signals that have been modified for various reasons back to the original signals.

Conventionally, a feedback control system has been widely used as a method for correcting a transformed signal based on an error from the expected original signal shape.

In other words, among the elements that caused deformation of the signal, the main M
The deformation is eliminated as much as possible by compensating the individual elements. For example, when considering automatic signal amplitude control, fluctuations in the signal level correspond to the deformation of the signal referred to here, and the factor causing this deformation corresponds to fluctuations in the gain of the amplifier through which the signal passes.

An even more complex example is automatic waveform equalization using transversal filters in waveform transmission.

This is (2N+1) of the transversal filter.
Control is performed to minimize intersymbol interference between waveforms by adjusting the tap gains of each waveform. At this time, the taps other than the center tap of the transversal filter and the gain are controlled so as to completely complete the zero crossings before and after the peak of the waveform. Not limited to these two examples, in order to perform compensation in such a feedback control system, a compensation circuit is required that can compensate for the M elements that cause signal deformation, as described above.

Next, an algorithm is required for how to change the M compensation parameters of this compensation circuit. In general, this algorithm uses an element Mi as the amount of deformation I of the signal.
The partial differential of Mi with respect to I, which represents how the parameter I influences, is determined, and the parameter Mi is gradually changed in the direction opposite to this partial differential.

At this time, if the amount of compensation for the M elements of the compensation circuit (on the contrary, the amount of change in the signal I) can be expressed linearly by M parameters, the above-mentioned partial differential can be obtained relatively easily.

In other words, regarding the automatic equalizer mentioned above, the difference between the equalizer output and its estimated value and the E equalizer outputs before and after that time are used to calculate the transversal coefficient for the amount of intersymbol interference.
The partial differentials of the (2N+1) tap gains of the filter are found.

On the other hand, the amount of compensation for M elements of the compensation circuit is
In cases where it cannot be expressed linearly with M parameters,
The above partial differential cannot be determined by simple calculations.

SUMMARY OF THE INVENTION An object of the present invention is to provide a compensation circuit that calculates partial differentials of each of M parameters with respect to the amount of change in the signal and compensates the signal based on the partial differentials of each M parameter in order to smoothly perform signal compensation under such circumstances. It is in.

The compensation circuit of the present invention compensates for signal distortion by controlling M parameters, and compensates for the distortion by adjusting the M parameters in the control to minimize distortion with respect to a normal signal. Two or more compensation circuits capable of
Connect the same parameter control terminals, select one of the compensation circuits for residual distortion observation, and select a control terminal for an arbitrary parameter Mi among the parameter control terminals of the selected compensation circuit for residual distortion observation. The difference between the residual distortion signals of this compensation circuit for residual distortion observation is calculated between the case where a minute signal is added only to and the case where the minute signal is removed, and the parameter Mi is successively changed according to this difference. This is a control circuit characterized by the following. According to the present invention, a stable control circuit can be constructed even if any type of compensation circuit is employed to compensate for signal distortion. Next, the present invention will be explained in detail with reference to the drawings.

FIG. 1 shows an embodiment of a compensation circuit used in the case of single frequency selective fusing due to multi-bus, which is a problem in wireless communication.

In the figure, a subtracter 1000, FETlO working as a variable resistor
The variable delay line 1002 reversely subtracts the multi-bus components in the transmission path. That is, if the strength and delay amount of the refracted wave relative to the direct wave in the transmission path are accurately known, then this refracted wave can be reproduced and subtracted from the direct wave.

Here, the strength of the refracted wave is changed by a control signal from a terminal 530, and the amount of delay is changed by a control signal from a terminal 540. Bandpass filter 1010, 1011, 10
12 is tuned to three pilot signals set within the transmission band. These outputs are sent to detectors 1020 and 1020, respectively.
1021, 1022, low pass filter 1030, 1
031, 1032 and subtracters 1040, 1041,
At step 1042, the difference between each pilot signal and the reference reception level is detected. This difference is squared circuit 1050, 1051
, 1052, and these three outputs are sent to adder 1
060 is added. Therefore, from terminal 580, the sum of squares of the differences between the three pilot signal levels from the reference level is obtained. A compensated signal is available at terminal 200. FIG. 2 is a diagram showing an embodiment of the present invention.

This embodiment uses the compensation circuit shown in FIG. 1 to automatically equalize frequency selective phasing. 10 in the figure,
20 and the compensation circuit, the control terminals for the strength of the refracted wave are 510 and 530, and the control terminals for the amount of delay are 520 and 54.
It is 0. Control terminals 510 and 530 are connected and manually controlled by a control signal from terminal 300. On the other hand, the terminal 520 is connected to the terminal 530 via the adder 30. Therefore, while the control signal of the terminal 400 is applied to the terminal 540, the "° control signal" is applied to the terminal 520 by closing the switch 600. The value of α゛ is battery 70
Add from 0. In this manner, due to the difference in the input signals to the terminals 520 and 540, different values are output to the output terminal 590 of the control circuit 10 depending on whether the switch 600 is closed or opened. This output difference is calculated by the sample and hold circuit 601.
, 602, are detected by the subtractor 40. That is, when the switch 600 is closed, a hold signal is applied to the sample and hold circuit 601, when the switch 600 is opened, a hold signal is applied to the sample and hold circuit 602, and the difference between the outputs of both sample and hold circuits is subtracted. detected by the device 40. The obtained output difference is held by another sample and hold circuit 603 and multiplied by 1/α by the amplifier 50. The output of this amplifier 50 clearly represents the partial differential coefficient of the error signal at terminal 590 with respect to the amount of delay. That is, the amount of change in the error signal when the intensity of the refracted wave is fixed in the control circuit 10 and the amount of delay is changed by +α is obtained from the output of the sample-and-hold circuit 602. The obtained partial differential coefficient is sent to the integrator 70
added to. Since the integrator output uses an integrator whose sign is reversed between the input and output, the integrated value of the value according to the output of the amplifier 50 is subtracted from the initial value of the integrator 70 over time, and the aforementioned partial difference is When the coefficient becomes zero, this integral operation also stops, and the frequency amplitude characteristics are equalized. The equalized signal is obtained from the output terminal 200. FIG. 3 is a diagram showing another embodiment of the present invention.

In the embodiment shown in FIG. 2, the strength of the refracted wave was manually adjusted to compensate for the frequency amplitude characteristics. In the embodiment shown in FIG. 3, the strength of the refracted waves is also automatically controlled at the same time. In the figure, compensation circuits 10, 20, adder 30, subtracter 40, amplifier 5
0, integrator 70, sample/hold circuit 601, 60
2,603, switch 600, and battery 700 are exactly the same as those in the embodiment shown in FIG.

In order to control the strength of the refracted wave, an adder 31 and an integrator 71 are used.
, sample and hold 604 switches 605 each have 30
, 70, 603, and 600 to have exactly the same function. Therefore, by opening and closing the switch 605, the sample/hold 604 obtains the partial differential coefficient of the error signal at the terminal 590 with respect to the strength of the refracted wave. Therefore, the output of the integrator 71 passes through the terminal 300 to the control terminal 5.
The voltage at the control terminal 530 is varied until the partial differential coefficient of the strength of the refracted wave with respect to the error applied to the refracted wave 30 becomes zero. According to the control circuit of the present invention, stable signal compensation can be performed even when the amount of compensation for a signal and each parameter are not linear or unclear.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an example of a compensation circuit, and FIGS. 2 and 3 are block diagrams each showing an embodiment of the present invention. In the figure, 10, 20, 11 are compensation circuits, 30, 31 are adders, 40, 41 are subtracters, 50, 51 are amplifiers, 7
0, 71 are integrators, 601, 602, 603, 604 are sample and hold circuits, 600, 605 are switches,
700 is a battery.

Claims (1)

[Claims] 1. In a control circuit that compensates for signal distortion by controlling M parameters and converges the distortion to a normal signal to a minimum, by adjusting the M parameters,
Two or more compensation circuits capable of compensating for the distortion are independently provided, one is used exclusively for actual signal compensation, the other is used for residual distortion observation, and the M parameter control terminals of these compensation circuits are the same. select one of the compensation circuits for residual distortion observation, and select an arbitrary parameter M among the parameter control terminals of the selected compensation circuit for residual distortion observation.
means for calculating the difference between the residual distortion signal of the residual distortion observation compensation circuit when a minute signal is added only to the control terminal of i and when the minute signal is removed; A control circuit characterized in that the parameter Mi is successively changed.