JPH04343508A - Digital type automatic gain controller - Google Patents

Abstract

PURPOSE:To attain gain control so that a required signal output can be obtained by converting an analog signal into a digital signal. CONSTITUTION:An inputted audio signal is supplied to a variable gain circuit 10 and an output signal from the circuit 10 is supplied to a gain setting part 14 through a maximum value detection part 11, a logarithm compression part 12 and a reverse back-lash data adder 13. The gain setting part 14 is provided with three filter units 14A to 14C and an output signal from a variable gain circuit is controlled so as to be an optimum recording signal by independently changing the response characteristics of respective filter units.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital auto-gain controller useful, for example, in controlling the level of an audio signal.

0002

2. Description of the Related Art When recording audio signals with a tape recorder or the like, automatic gain control (AGC) is used to control the level of the input audio signal so that the recording signal level for the magnetic tape is optimal.
l) Supply through the circuit is performed.

[0003] In this case, in particular, the tape recorder is, for example, a DAT (Digital Audio Tape).
In order to make effective use of its wide dynamic range, an automatic level control circuit that has high resolution and exhibits an optimal response is required.

[0004] Therefore, a feedback circuit is provided with different level detectors and integration circuits with different time constants so that the output of the variable gain amplifier is detected and the output becomes a predetermined recording current, and fine gain settings are made. An AGC circuit has been developed to do this.

[0005]

[Problems to be Solved by the Invention] However, when applying closed-loop linear control to such an AGC circuit and trying to improve its control characteristics, that is, to make the gain of feedback control as high as possible, it is difficult to use automatic gain control. The output value of the camera repeatedly vibrates slightly due to the influence of disturbances (noise), etc. Further, when the AGC circuit is configured in an analog manner, there is a problem that the circuit configuration becomes plural and the cost increases due to the increase in the number of mounted components.

[0006]

[Means for Solving the Problems] In order to solve this problem, the present invention provides a variable gain means, a gain setting means for controlling the gain of the variable gain means, and a gain setting means based on the output level of the variable gain means. , an auto gain controller comprising level detection means for determining a setting value of the gain setting means, the gain setting means including at least three digital low-pass filters with different time constants, and each digital low-pass filter supplied to each of the digital low-pass filters. Filter unit means are provided so that the level of the detection signal can be varied independently. Then, a value obtained by sampling the output level of the variable gain means at a predetermined timing is supplied to the filter unit means, and the output level is controlled with desired level characteristics through digital signal processing. .

[0007]

[Operation] Since the entire level control circuit is composed of a digital signal system feedback circuit,
The number of parts is reduced, and level control that is stable against changes over time and environmental changes can be provided at low cost.

[0008]

[Embodiment] Fig. 1 shows a block diagram of an auto gain controller of the present invention, in which 10 is a variable gain circuit, and an AD gain converter 10A that adjusts the level of an analog signal and converts it into a digital signal. , a DSP gain conversion section 10B configured with a multiplier.

[0009] The AD gain converter 10A is particularly configured with an input terminal T.
This is provided to roughly adjust the input signal level of the audio source supplied from the AD gain converter 10A.
The signal is supplied to the P gain converter 10B, and is input to the DSO gain converter 10B, so that a fine level adjustment of, for example, 0.1 to 0.2 dB is performed using a multiplication coefficient.

Reference numeral 11 denotes a maximum value detection unit that detects the maximum value of the level-adjusted output signal at a predetermined timing, and in the case of a stereo signal, it detects the larger value of either the left signal L or the right signal R. is being detected. The detected maximum value is then supplied to the next logarithmic conversion section 12, where it is subjected to level compression processing so as to correspond to human auditory sense.

Reference numeral 13 denotes a reverse backlash data addition circuit for preventing the influence of the backlash function from occurring in the loop even if a backlash function is added, as will be described later. The output data of 13 is supplied to a gain setting section 14.

As will be described in detail later, the gain setting section 14 includes a first filter unit that responds at low speed (10 to 100 seconds) to changes in the level of the output signal sent to the digital signal processing section 17. 14A, a second filter unit 14B that responds at a medium speed (1 to 10 seconds), and a third filter unit 14C that responds rapidly (0.1 seconds or less), and the output of each of these filter units is The signals are integrated in an adder 14D to form control data for setting the gain of the AGC circuit.

Incidentally, the first filter unit 14A
forms a gain control signal that corresponds to the level difference of the input audio source, and the third filter unit 14C forms a gain control signal that suppresses a signal that exceeds the dynamic range due to the occurrence of a rapidly rising high-level peak value. Form a control signal. Further, the second filter unit 14B responds to signals of medium level or higher and forms a gain control signal in a normal AGC circuit.

15 compares the gain control command value Yn-1 outputted at the previous timing with the control data Xn of the gain setting means 14 obtained at the current timing, so that fine vibrations due to disturbances etc. are suppressed. This is a backlash circuit that forms a gain control command value Yn, and gain control data for setting the gain of the variable gain circuit 10 described above based on this gain control command value Yn is supplied via an interface 16. ing.

As shown in the flowchart of FIG. 6, the operation of the backlash circuit 15 is based on the control data Xn obtained at the current timing and the gain control command value Yn of the variable gain circuit 10 output at the previous timing.
-1, and depending on whether or not this difference output is larger than a certain constant value Z, the gain command value Yn to be output next is determined.
The value of is determined.

That is, when |Xn - Yn-1 | is smaller than Z, the next gain control command value Yn is not particularly output, and remains the previous gain control command value Yn-1. Also, |Xn −Yn−1
When | is greater than Z, if Z is positive,
The gain command value Yn-1 that is output next is “Xn +
However, when Z is negative, the next gain command value Yn is output as "Xn - Z".

Therefore, in the auto gain controller of the present invention, minute vibrations in the output level due to disturbance noise, etc. are suppressed within the Z level range (within the dead band level) that is the backlash value, and the cause of click noise is suppressed. can be removed.

By the way, since the backlash control described above is a nonlinear closed-loop control system, the output value Xn of the gain setting section 14 is changed to, for example,
It fluctuates as shown in . Therefore, in the present invention, in order to suppress this unnecessary movement, the above-mentioned reverse backlash circuit 13 is inserted into the closed loop, for example, |Xn −
The (Xn - Yn) value when Yn | > Z is fed back to the reverse backlash circuit, and pseudo I'm letting you do it.

FIG. 2 shows the first, second, and third filter units 14 (A,
As shown in FIG. 3, C1 is the timing (T=2) corresponding to one track of the audio signal recorded on a rotating drum,
Positive peak value P of output level S captured every 0ms)
1, P2, P3..., and negative peak value P-1
, P-2, P-3, . . . are supplied to input terminals 21
A and 21B are slice circuits that slice the output signal according to a negative threshold level E (-) and a positive threshold level E (+), and 22A and 22B are rectifiers for obtaining positive and negative outputs. circuit, 23A, 2
3B is an amplifier that varies the rectified output level, 24
indicates an adder.

The subtracter 25, coefficient units 26A and 26B, adder 27, and one-sample delay circuit 28 form a digital low-pass filter that integrates positive and negative signals along their polarities. , switches S1 and S are switched by a switching signal Si supplied from a circuit not shown.
2 is switched so that, for example, when the calculation result of the digital low-pass filter is negative, contact a is selected, and when is positive, contact b is selected.

This digital low-pass filter is shown in FIG.
Input data Xn for n samples as shown in a)
The output data Yn for is Yn = (Xn - Yn-1 ) τ + Yn-1 = Xn
・τ+Yn-1 (1-τ).

Therefore, the equivalent circuit is shown in FIG. 5(b).
In the case of (a) of Fig. 5, it is rewritten as shown in 1.
Since output data is obtained by multiple multiplications, calculation time can be shortened. τ indicating a time constant is τ<1, and the smaller τ is, the greater the influence of previous output data becomes, resulting in an integrating circuit with a large time constant.

In such a filter unit input, the output of the variable gain circuit 10 is detected by the maximum value detection section 11 at every predetermined timing period T as shown in FIG. It is compressed as follows. Then, this output level data is supplied via the reverse backlash circuit 13, and is supplied to the first filter unit 14A, the slice circuits 21A, 21B, the rectifier circuits 22A, 22B, and the amplifiers 23A, 23B. Processing is performed using input-output characteristics as shown in 4(a).

That is, the first filter unit 14A
In this case, the threshold levels E(-) and E(+) are both 0 level, and the angle θ indicates the gain of the amplifiers 23A and 23B.
1 is relatively small. Also, the second one has a similar configuration.
In the filter unit 14B, the threshold levels of the slice circuits 21A and 21B are set to E(-) and E(+), as shown in FIG. Only the output data is converted with a gain θ2.

Furthermore, in the third filter unit 14C having a similar configuration, the threshold level E(++) is set only for the + side output level data, as shown in FIG. 4(c). It is set such that only output level data greater than this E(++) is converted with a large gain of gain θ3.

Next, the integration time constant of the digital filter constituting the first filter unit 14A is 10 to 100.
It has a low Sec, and forms output data with speed changes that are performed manually by the user.

The integration time constant of the digital filter configured in the second filter unit 14B is 1 to 10 S.
Selected by ec. This time constant indicates the level change in the automatic recording level setting circuit (AGC circuit) of a general tape recorder, but as shown in FIG. Prevented from responding.

Furthermore, in the embodiment of the present invention, the third filter unit 14C has a fast integration time constant 0 that responds quickly.
．． By providing a digital filter of 1 S or less, momentary high-level signals are suppressed. In this case, the level to which the digital filter responds is, for example, as shown in FIG.
As shown in (c), the positive threshold level E(+
+) or above, and a strong suppressing effect is applied to steep signals exceeding this level.

In the gain setting section 14 of the above embodiment, for example, the third filter unit 14C mainly responds to a large input audio signal that is instantaneously input, and suppresses the output level. At the same time, the first filter unit 14A can respond to a small audio input level immediately after that, and can also respond to an instantaneous level drop. Further, the second filter unit 14B can respond to average level fluctuations and supply an output at an appropriate level to the digital signal processing section 17.

[0030] Also, the time constant τ(f) of each filter unit
, τ(r) and threshold level E(+) E
(-), θ1 θ2 θ2, etc. can be easily set individually, so the AGC can be adjusted as appropriate according to the user's preference.
The response characteristics of the circuit can be defined. Furthermore, by increasing the gain of the feedback loop, fine adjustment of the input level becomes possible, and an optimum recording level can be given to the magnetic tape.

[0031]

As explained above, the digital auto level controller of the present invention converts an input audio signal into a digital signal, supplies it to a variable gain circuit, and adjusts the output level detected at a predetermined timing. Since the gain control signal is formed by a filter unit having three types of response characteristics for the peak value, the entire circuit can be controlled by an AG that can be controlled according to a program by a microprocessor, for example.
It can be made into a C circuit.

In particular, since the feedback circuit system is divided into multiple band processing systems, there is an advantage that appropriate level control can be performed for all changes in the input signal. In addition, because it uses digital signal processing, it is excellent against aging and environmental changes.
The characteristics of the C circuit can be easily changed, and the cost and space of the circuit can be reduced.

[Brief explanation of drawings]

FIG. 1 is an overall diagram of a digital auto gain controller of the present invention.

FIG. 2 is a block diagram showing a specific example of a filter unit that detects an output level and sets a gain.

FIG. 3 is an explanatory waveform diagram of peak level detection points.

FIG. 4 is a gain characteristic diagram of output levels supplied to three digital filters.

FIG. 5 is an equivalent circuit diagram of a digital filter.

FIG. 6 is a flowchart for adding backlash.

FIG. 7 is a diagram showing fluctuations in control data occurring in a circuit within a nonlinear loop.

[Explanation of symbols]

10 variable gain circuit 11 Maximum value detection section 12 Logarithmic compression circuit 13 Reverse backlash addition circuit 14A First filter unit 14B Second filter unit 14C Third filter unit

Claims (1)

[Claims] 1. Variable gain means; gain setting means for setting the gain of the variable gain means; detecting the output of the variable gain means at every predetermined timing to obtain its peak value; - level detecting means for logarithmically compressing the peak value and supplying the compressed value to the gain setting means; A digital auto gain controller characterized by being equipped with a level setting means that can independently vary the signal level.