JPS5942493B2 - noise reduction system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a noise reduction system that detects the level of a signal and compresses and expands the dynamic range of the signal based on the signal level. The present invention relates to a noise reduction system that can increase compression effects with small transmission distortion by effectively switching.

As this type of noise reduction system, the one shown in FIG. 1 is known.

This noise reduction system includes a means (encoder) 11 for compressing the dynamic range of a signal, and a means (decoder) 12 for expanding the compressed signal to the original dynamic range, and a tape recorder or the like inserted between them This is to reduce noise in the medium 13 or transmission system.

Encoder 11 and decoder 12 are each VCA1
4.15 and a level sensor 6.17.

VCAI4.15 is a kind of multiplier, and the input signal ei
When a DC control signal E is given, an output signal e is obtained.

is ± eo-E 1×ei is given by

However, the exponent of E takes a negative sign during encoding and a positive sign during decoding.

Furthermore, the level sensor 6.17 is a circuit that detects the signal level, and when a signal e is given, the signal e corresponding to that level is detected.
The output is obtained.

By combining these blocks, when encoding eol-E-1×ei1 So, about the level EOl-EOl-1×Ei1 Than Eol = Ei 1 ”’ becomes.

On the other hand, when decoding, eo2:Ei2×ei2 To summarize the level of Eo2−Ei2′ becomes.

In other words, when encoding, the level change is 1 on a log scale.
/2, and expands to 2 times when decoding.

Figure 2 shows this input/output characteristic diagram.

During recording, the input signal of +20 ctH is +1 due to straight line A.
The peak margin is improved by suppressing the signal to 0 dB.

Also, an input signal of -60 dB becomes -30 ciB, and the dynamic range of the input signal as a whole is compressed in half and recorded.

During reproduction, the signal is expanded along the straight line B, the signal recorded at +10 aB is returned to +20 dB, and the signal recorded at -3 octH is lowered again to -6 octH.

At this time, the noise is also reduced by 30 dB.

That is, for a -60 dB input signal, 30 ct
The S/N improvement degree of B is obtained.

In this system, the S/N improvement degree is generally expressed as m-x (input signal level [dB]) [dB].

However, the OdB of the input signal level is the cross point in FIG.

Incidentally, in such a system, a peak detection circuit or an average value detection including RMS detection or the like has conventionally been used as a level detector.

FIG. 3 is an example of a level compression circuit using the simplest peak detection circuit.

4a to 4d show waveforms at various parts of the circuit diagram of FIG. 3.

The signal a applied to the input terminal 31 is applied to the peak detection circuit 33 together with the VCA 32.

This signal is subjected to double-wave detection by the absolute value detection circuit 34.

When the input signal is a sine wave as shown in FIG. 4a, the absolute value detection circuit 34 outputs a waveform as shown in FIG. The waveform shown in FIG. 4C is obtained.

The output signal d generated at the output terminal 38 is controlled by this waveform.
has the waveform shown in FIG. 4d, and each negative wave becomes a distorted wave.

FIG. 5 shows an example using the average value detection circuit 53.

FIG. 6a=d is a waveform diagram of each part of FIG. 5.

As in the case of peak detection, the signal a applied to the input terminal 51 is applied to the absolute value detection circuit 54.

The output signal of the absolute value detection circuit 54 is averaged by a resistor 55 and a capacitor 56, which are time constant elements, to obtain the waveform shown in FIG. 6C.

If the time constant is increased, the ripple (not shown) in the waveform shown in FIG. 6C can be almost ignored.

The output signal d of the VCA 52 controlled by this waveform has the waveform shown in FIG. 6d, and the amplitude of the wave front is larger than that of the steady portion.

That is, when peak detection is used, an amplitude equal to that of a steady state can be obtained from the wave front, but waveform distortion occurs for each wave, which is not preferable.

Furthermore, when average value detection is used, waveform distortion can be ignored by increasing the time constant, but the amplitude of the wave front becomes larger than that of the stationary portion.

However, when FM is used in the transmission system, carrier wave deviation increases as the amplitude increases, and redundant waves in the band increase.

Furthermore, when a signal is recorded as displacement information on a record or the like, the recording density decreases as the amplitude increases, which is also undesirable.

The present invention has been made in view of this point, and provides a noise reduction system that can effectively compress the level of even sharply rising signals and that is free from waveform distortion in the stationary portion.

FIG. 7 shows an embodiment of the present invention.

The input signal a applied to the input terminal 70 is double-wave rectified by the absolute value detection circuit 71.

This signal consists of a resistor 72 and a capacitor 7, which are time constant elements.
3 and output to an output terminal 75 through a batshear 74.

For example, if a sine wave with amplitude A as shown in FIG. 8a is input to input a, the output of the absolute value detection circuit 71 will be as shown in FIG. As shown in , it becomes 2A/π.

On the other hand, the output of the absolute value detection circuit 71 is resistor 76.77.
The signal is supplied to a buffer 78 through a 2/π coefficient multiplier consisting of a 2/π coefficient multiplier, and an amplitude waveform with an amplitude 2/π times the absolute value of the input signal is obtained as an output.

The difference between the output of the buffer 74 and the output of the buffer 78 is expressed as a ratio ΔV/Δt of the signal level difference ΔV to the delay time Δt caused by the capacitor 73, which is the rate of change in the signal level.

This rate of change is amplified by an amplifier 79, and the conduction state of the transistor 81 is controlled by the output of the nonlinear circuit 80.

The functions of the amplifier 79 and the nonlinear circuit 80 are as follows.

That is, when the output of buffer 74 is lower than the output of buffer 78, transistor 81 is rendered non-conductive.

Conversely, when the output of the buffer 74 is higher than the output of the buffer 78, the transistor 81 becomes active, and the greater the difference between the two, the lower the equivalent resistance becomes, controlling the time constant for charging the capacitor 73.

FIG. 8d shows the conduction state of the transistor 81.

In the shaded area of the diagram, the transistor 81 becomes conductive and is charged to a peak level of 2/π of the amplitude of the input sine wave. After that, the transistor becomes non-conductive and is repeatedly charged and discharged by the resistor 72, and the potential of the capacitor 73 is the average of the signal. The value 2A/π is maintained.

In other words, in this embodiment, the circuit is configured such that the larger the rise of the signal level, the closer the detection is to peak detection, and the smaller the change in the signal level, the closer the detection is to average value detection.

By doing this, the level of the wave front can be detected reliably, and since the average value is detected in the stationary part, an increase in distortion is hardly recognized.

By incorporating this into a level compression/expansion noise reduction system, narrowband transmission and high-density recording as described above become possible.

In the above embodiment, an example has been described in which the rise time constant changes continuously depending on the magnitude of the rate of change.

Continuously changing the temperature in this way increases the tolerance for adjustment accuracy and temperature drift, and is industrially and practically advantageous.

However, in cases where the transfer characteristics do not require such precision, the circuit shown in FIG. 9 is more advantageous in terms of cost.

This will be explained by assigning the same reference numerals to the parts corresponding to those in FIG. 7.

This circuit is a simplified version of the control circuit for the transistor 81, in which the potential of the output of the buffer 74 and the output of the buffer 78 are compared by a Schmitt circuit 91, and the transistor 81 is turned on or off depending on the state of the output. It is designed to control either of them.

The sharpness of the rise is determined based on the magnitude of the history of the Schmitt circuit 91.

In other words, when the rise is sharp, the buffers 78 and 74
The difference in output is large.

If this is greater than the magnitude of the Schmitt circuit's history, the Schmitt circuit inverts its output.

FIG. 10 shows a more reliable version of the method shown in FIG.

In the method shown in FIG. 9, when the level rises significantly as shown in 111 in FIG. 11, and then reaches an even higher level as shown in 112 and then settles down, 111 with a large level change is
Although the wave crest is detected correctly, the buffer 74
The output will be as shown in Figure 11, and if the level is compressed using the output of such a level detector, it will become as shown in Figure C.

In order to solve this problem, as shown in FIG. 10, a monostable circuit 101 may be provided at the next stage of the Schmitt circuit 91 in FIG. 9.

The monostable length may be set to a time that encompasses the envelope of a musical instrument that has a fast-rising musical tone and successively increases in amplitude.

These instruments include cymbals, maracas, guiro, and marcato violins.

The respective envelope waveform diagrams are shown in FIGS. 12a, b, and c.

As we will see, the maximum monostable length is 150 m5
It's good to have.

However, as described above, the effects of the present invention can be obtained even without providing a monostable circuit as shown in FIG.

FIG. 13 shows a more natural effect.

That is, when the time constant is switched in the circuit shown in FIG. 10, the waveform may be distorted.

Therefore, it is only necessary to make this switching operation gradual.

That is, the time constant circuit 1 is included in the control circuit of the transistor 81.
31 is provided.

In FIG. 14, a log conversion circuit is provided at the input part of the Schmitt circuit, which allows the rate of change in the rise to be determined in dB, which is more practical.

Furthermore, a LOG conversion circuit may be provided in the input section of the circuit shown in FIG. 9 or the like.

[Brief explanation of drawings]

Fig. 1 is a principle diagram of an example of a conventional noise reduction system, Fig. 2 is an operational characteristic diagram of the system shown in Fig. 1, Fig. 3 is an example of a conventional level detection circuit, and Fig. 4 is a diagram of an example of a conventional level detection circuit.
Figure 5 is a waveform diagram of each part of the circuit shown in the figure, Figure 5 is another example of a conventional level detection circuit, Figure 6 is a waveform diagram of each part of the circuit of Figure 5, and Figure 7 is a circuit of one embodiment of the present invention. 8 is a waveform diagram of each part of the circuit of FIG. 7, FIG. 9 is a circuit diagram of another embodiment of the present invention, and FIG. 10 is a circuit diagram of still another embodiment of the present invention.
FIG. 1 is a special example of a waveform diagram using the circuit of FIG. 9, FIG. 12 is a rise envelope diagram of several types of musical tones, FIG. 13 is a circuit diagram of another embodiment of the present invention, and FIG. 14 is a diagram of the rise envelope of several musical tones. FIG. 7 is a circuit diagram showing still another embodiment. 71... Absolute value detection circuit, 72... Resistor, 73... Capacitor f, 74, 78...
...Buffer, 79...Amplifier, 80...
...Nonlinear circuit, 81...Transistor.

Claims (1)

[Claims] 1 A means for compressing the dynamic range of an input signal, comprising a level detector for detecting the amplitude level of a human input signal and a voltage gain control amplifier whose gain is controlled by the output voltage of the level detector; In the noise reduction system, the level detector includes a means for rectifying the input signal and a means for rectifying the input signal. , means for smoothing the rectified signal obtained by the means with a predetermined time constant, means for detecting a rate of change in the amplitude level of the input signal, and a rate of change is large depending on the rate of change in the detected amplitude level. A noise reduction system comprising: means for controlling the time constant of the smoothing means to become smaller as the time constant increases.