JPH02237207A - Digital agc control system - Google Patents

Abstract

PURPOSE:To shorten the converging time of an output signal level in the wide range of an input signal level by installing multipliers for multiplying an error value between the square mean value of an output signal and a desired set value, and the absolute value of the error value. CONSTITUTION:The error value D between the desired set value of the output signal level and the square mean value of the output signal 210 is outputted from a subtracter 107. In a correction part 150 consisting of an absolute value part 16 and the multiplier 105, the output M of the multiplier 105 is squared by preserving the code of the error value D as it is. The output M of the multiplier 105 is multiplied by alpha (prescribed constant) in the multiplier 104, is added by the output of a register 102, namely, a gain before one sample, by an adder 103 and is supplied to the multiplier 101 as the gain for the input signal. Thus, the converging time of the output signal level can be shortened in the wide range of the input signal level.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a digital AGC control method used in a digital signal processing device.

[Prior Art] A block diagram of a conventional digital AGC control circuit is shown in FIG. 3, and its operation will be explained.

A conventional digital AGC control circuit as shown in FIG.
A multiplier 101 multiplies the input signal by the output value of the adder 103 to obtain an output signal.

At this time, the output signals are simultaneously output from the squarer 109 and the averager 1.
08, the root mean square is calculated.

This root mean square value is subtracted from the set value by a subtracter 107 to determine the error from the set value. This error is sent to the multiplier 104, which multiplies the error by a predetermined coefficient "α"', adds this value to the value of the register 102 in the adder 103, and outputs it to the multiplier 101.In this way, the following The gain for the input signal is controlled and corrected so that the error in the output value of the subtracter 107 becomes zero, and as a result, the output signal is maintained at a constant level.

[Problems to be Solved by the Invention] However, in the conventional digital AGC control circuit described above, as the difference between the input signal level and the desired output signal level increases, the time required to converge the output signal to the desired level increases. becomes longer.

This convergence time can be shortened in almost inverse proportion to the value of the coefficient "α" to the subtracter 107 by increasing the value.

However, increasing the value of "α" beyond a certain point is good when the difference between the input signal level and the desired output signal level is large, but when the difference between the input signal level and the desired output signal level is small, this will result in overcontrol. In some cases, fluctuations in the output signal level become large, or even the output signal oscillates.

That is, if an attempt is made to suppress the convergence time of the output signal to within a certain time for all input signals, there is a drawback that the dynamic range of the input signal level is limited.

[Means for Solving the Problems] The present invention has been made for the purpose of solving the above-mentioned problems, and includes the following configuration as one means for solving the above-mentioned problems.

That is, an addition means for determining a gain for an input signal, a storage means for storing the gain in the addition means, a multiplication means for multiplying the storage means by the input signal, and an input signal from the multiplication means. Modifying means for determining the root mean square of the output signal, calculating the error with respect to a predetermined set value, raising the error value to at least the second power while maintaining the sign, and modifying the gain of the addition means so that the error value becomes zero. and flifiλ.

[Function] In the above configuration, a predetermined constant is further added to the gain for the input signal at least one sample before, the error value between the root mean square value of the output signal and the desired set value to the power of at least the square while keeping the sign. By adding the multiplied value and controlling the error to zero as a gain for the current input signal to keep the output signal level at a desired constant value, the convergence time of the output signal level can be reduced over a wide range of input signal levels. can be shortened.

[Example] Hereinafter, an example according to the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram for realizing a digital AGC control method according to an embodiment of the present invention.

In FIG. 1, the same components as in FIG. 3 are given the same numbers.

The input signal 200 is input to one input of the multiplier 101, and the output from the adder 103 is connected to the other input of the multiplier 101. The output of the multiplier 101 is output as an output signal 210 after level control.

At the same time, this output signal 210 is connected to the input of the squarer 109. Then, it is squared by this squarer 109, and the output is connected to the input of an averager 108 and averaged there. The output of this averager 108 is a value obtained by averaging the output signal 210 into squares. The output from this averager 108 is connected to one of the subtracters 107, and the subtracter 10
A predetermined setting value (output level specific value of the output signal 210) is input to the other human power of 7. Therefore, the subtracter 107 outputs a value obtained by subtracting the root mean square of the output signal from a predetermined setting value, that is, an error value between the desired output signal level setting value and the root mean square value of the actual output signal 210. be done.

Here, the feature of this embodiment is that a correction section 1 consisting of an absolute value section 106 and a multiplication section 105 surrounded by a broken line in FIG.
It is 50.

In this correction section 150, the error value that is the output of the subtracter 107 is connected to one input of the multiplier 105, and is also connected to the input of the absolute value section 106. The output of the absolute value section 106 is then connected to the other input of the multiplier 105.

The output of the multiplier 105 is connected to one input of the multiplier 104, and the other input of the multiplier 104 is connected to a predetermined constant ゛'α
゜′ is input.

The output of multiplier 104 is connected to one input of adder 103, and the other input of adder 103 is connected to the output of register 102. The output of adder 103 is sent to register 10
2, and is further connected to the input of multiplier 101. As a result, the output of the adder 103 is controlled so that the error value between the desired set value and the root mean square value of the output signal becomes zero.

The operation of the digital AGC control circuit of this embodiment having the above configuration will be explained below.

If the error value between the desired setting value, which is the output of the subtracter 107, and the root mean square of the output signal is "D", and the output of the multiplier 105 is "M", then M=DX D .

In other words, "M" is squared while preserving the sign of the error value "D".

The output "M゜゜" of the multiplier 105 is multiplied by α by the multiplier 104, added to the output of the register 102, that is, the gain of one sample before by the adder 103, and is supplied to the multiplier 101 as the gain for the current input signal. be done.

In this way, the output signal level is controlled to be constant, but in this embodiment, by performing a nonlinear operation of squaring the error value, when the difference between the output signal level and the desired set value is large, that is, , D>1, a larger gain control amount can be obtained than in the conventional method described above.

Moreover, since this control amount is proportional to the square of ゛D゜゜,
To stabilize an output signal level sufficiently quickly even when the difference between a human input signal level and a desired output signal level is large.

In addition, since the operation after the level of the output signal 210 is stabilized is D<1, the squared control amount becomes smaller than in the conventional method, and fluctuations in the output signal level can be reduced.

As explained above, according to this embodiment, by providing a multiplier for multiplying the error value between the root mean square of the output signal and the desired setting value by the absolute value of this error value, the input signal level can be adjusted. The convergence time of the output signal level can be shortened over a wide range.

[Other Embodiments] The above description has been about a configuration including an error value between the root mean square of the output signal and a desired set value, and a multiplier for multiplying the absolute value of this error value. is not limited to the above configuration, and the same effect can be obtained by using the cube of the error value as the control amount.

Hereinafter, a case will be described in which the cube of the error value is used as the control amount.

In FIG. 2, the correction section 150 shown in FIG. 1 is replaced with a circuit 160 shown within the broken line, and by setting the cube of the error value as the control amount, the same effect as in FIG. 1 is obtained. FIG. 2 is a block configuration diagram for realizing a digital AGC control method when configured.

In this case as well, the error value between the desired setting value, which is the output of the subtracter 107, and the root mean square of the output signal is set as "D", the output of the multiplier 105 is set as "M", and the output of the multiplier 110 is set as "N". ′゛, then N=D2, and further M=NX
D=D3.

The output ``M'' of the multiplier 105 is converted to α' by the multiplier 104.
The signal is multiplied by .degree. and added to the output of the register 102, that is, the gain of one sample before, by the adder 103, and is supplied to the multiplier 101 as the gain for the current input signal.

In this way, the output signal level is controlled to be constant, but in this embodiment, by performing a nonlinear operation of raising the error value to the third power, the output signal level and the desired set value can be adjusted as in the above embodiment. When the difference is large, that is, (D>1
), a larger gain control amount can be obtained than in the conventional method described above, and the output signal level is stabilized sufficiently quickly even when the difference between the human input signal level and the desired output signal level is large.

In addition, since the operation after the level of the output signal 210 is stabilized is (D<1), the control amount multiplied by the cube is smaller than in the conventional method, which reduces fluctuations in the output signal level. I can do it.

As explained above, according to this embodiment, by providing the error value between the root mean square of the output signal and the desired setting value and the multiplier that raises this error value to the third power, the output signal can be output over a wide range of input signal levels. The signal level convergence time can be shortened.

[Effects of the Invention] As explained above, according to the present invention, the time from when an input signal is input until the output signal level stabilizes is shortened over a wider range of input signal levels, and the output signal level is This has the excellent effect of keeping level fluctuations to a minimum after the level has stabilized.

[Brief explanation of drawings]

FIG. 1 is a digital AG control system according to an embodiment of the present invention. FIG. 1 is a block diagram of a digital AGC control method according to an embodiment of the present invention. FIG. 2 is a block diagram of another embodiment according to the present invention. Block configuration diagram for realizing the AGC control method. FIG. 3 is a block configuration diagram of a conventional AGC control circuit. In the figure, 101, 104, 105, 110... multipliers,
l○2...Register for saving gain, 103...
Adder, 104... Multiplier, 106... Absolute value part,
107... Subtractor, 108... Averager, 109...
. . . squarer, 150 . . . correction section, 160 . . . cubic section. Patent applicant Canon Inc.

Claims (2)

[Claims] (1) A digital AGC control method used in a digital signal processing device, which maintains the sign of the gain for an input signal at least one sample before, and the error value between the root mean square value of the output signal and a desired set value. The output signal level is maintained at a desired constant value by adding a value obtained by multiplying at least a predetermined constant to a value obtained by multiplying the value obtained by at least the square of the current input signal, and controlling the gain to the current input signal so that the error becomes zero. Digital AGC control method. (2) A digital AGC control method used in a digital signal processing device, which calculates the error between the gain for the input signal at least one sample before, the root mean square of the output signal, and a predetermined setting value. The output signal level is maintained at a desired constant value by adding an error value and a value obtained by multiplying the absolute value of this error value and further multiplying by a predetermined constant, and controlling the error value to become zero. Digital AGC control method.