JPS63202133A - Impulsive noise eliminating device - Google Patents

Abstract

PURPOSE:To prevent switching malfunction by controlling the detection sensitivity of an impulsive noise at a level of an audio signal so as to prevent the inversion. CONSTITUTION:An audio signal filter 10 is a filter where the K is changed by a frequency fS of an audio signal and an output signal of a gate circuit 6 is inputted thereto. A high pass filter 1 is a filter to exclude an audio signal component vs and to obtain only a noise component vn and receives an input signal of the gate circuit 6. As the audio frequency increases, the value K is increased to decrease the detection sensitivity of noise. Thus, the inversion is prevented. Moreover, when the audio frequency is fixed, proper noise elimination is applied depending on the noise level and the audio signal level.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a pulse noise removal device used in car radios and the like.

[Summary of the Invention] The present invention includes first and second variable attenuators that vary the signal levels of an audio signal component and a noise signal component in an input signal, and the first and second variable attenuators vary the signal levels of an audio signal component and a noise signal component in an input signal. By controlling the second variable attenuator and controlling the gate circuit for switching the input signal according to the output of the second variable attenuator,
The inversion phenomenon is prevented regardless of changes in the level and frequency of the audio signal.

[Prior Art] In car radios and the like, a noise removal device is used to remove pulsed noise generated by a car's spark plug or the like and mixed in.

This noise removal device uses a noise detection circuit to detect the presence of pulsed noise, and uses a gate circuit to switch and remove the pulsed noise contained in the audio signal.

FIG. 5 is a block diagram of a pulse noise removal device commonly used in car radios and the like.

In FIG. 5, a high-pass filter (HPF) 1 removes the audio signal S from the input signal v1 from the terminal 2 and obtains only the pulse noise N.

Pulse noise obtained from the noise filter 1 is amplified by an amplifier 3 and then led to an automatic gain control (AGC) circuit 4 and a monostable multicircuit 5.

When the pulse noise N is detected, the monostable multi-circuit 5 generates a predetermined switching pulse to shut off the gate circuit 6.

Generally, a pre-hold type, a linear signal compensation type, etc. is used as the gate circuit 6, but this gate circuit 6 is normally in a normal state, and the input signal v1 is passed directly to the terminal 7 as the output signal v0. Output from.

When a switching pulse is applied from the monostable multi-circuit 5 to the gate circuit 6, the gate circuit 6 is cut off, operates so as not to output the input signal v1, and removes the pulse noise N.

When the AGO circuit 4 continuously detects noise.

The gate circuit 6 is continuously cut off and operates so that the output signal does not stop continuously.

FIG. 6 shows an example of a signal switching waveform when noise is removed by a pre-hold type gate circuit.

In the figure, T'sw is a switching time for removing pulse noise, and is set to the time width of the pulse noise.

T is the repetition period of pulsed noise, and can actually take various values depending on the source of the noise. Note that this switching is performed by a switching pulse from the monostable multi-circuit 5 shown in FIG.

As mentioned above, FIG. 7 shows the audio signal frequency fs and signal distortion D when the audio signal is switched.
s, the distortion rate increases as the signal frequency fs increases, and the same tendency exists when the gate circuit 6 is of the linear interpolation type.

Figure 8 shows the reversal phenomenon when pulse noise is removed by switching, and the switching time T'
At sw, the audio signal S takes the form Pl-P, -P, and in this portion, a defective portion of only one area As occurs with respect to the original signal. On the other hand, when noise removal is not performed, the pulse noise component An shown in the shaded area in FIG. 1 is added to the signal S. Both of these components As and An cause distortion to the original signal. For component As, component A
.

is sufficiently large (pulse noise is large), by switching and performing an Il sound removal operation.

Signal distortion can be reduced. General noise removal devices aim at this effect.

[Problems to be Solved by the Invention] Conversely, if the component An is smaller than the component As (pulse noise is small), switching will increase distortion, and a so-called inversion phenomenon will occur. .

Therefore, in such a case, the noise removal operation must not be performed.

However, the loss area As of the signal due to switching varies depending on the signal frequency; it becomes smaller as the signal frequency is lower, and becomes larger as the signal frequency is higher. It is desirable to allow this to occur and to stop it at high frequencies where the reversal phenomenon occurs.

However, the current situation is that conventional noise removal devices do not take such measures.

SUMMARY OF THE INVENTION An object of the present invention is to provide a pulse noise removing device that can prevent the occurrence of a one-inversion phenomenon.

[Means for Solving the Problems] In order to achieve such an object, the present invention provides noise detection means for extracting noise signal components contained in an input signal;
Audio signal detection means for extracting an audio signal component included in an input signal, gate means for switching the input signal, and first and second levels for varying the signal levels of the extracted noise signal component and audio signal component. and a first level variable means according to the output of the second level variable means.
The present invention is characterized in that it includes level control means for controlling the second level variable means, and gate control means for controlling switching of the gate means in accordance with the output of the first level variable means.

[Function] The present invention controls the detection sensitivity of pulse noise based on the level of the audio signal, so it is possible to prevent the inversion phenomenon, and also to prevent switching malfunctions (harmonics of the audio signal) that occur when the audio signal increases in particular. It is possible to prevent this from occurring (determining that pulse noise is mixed in and performing a noise removal operation).

[Embodiments of the Invention] First, the principle of the present invention will be explained with reference to FIGS. 2 and 3.

FIG. 2 explains the improvement of the inversion phenomenon with respect to the signal frequency according to the present invention. When a constant noise Na that causes signal distortion Da is input, the noise is removed from the audio signal frequency fsa at which the inversion phenomenon occurs. Control behavior. That is, the noise removal operation is performed at frequencies below fsa.

At frequencies above fsa, the noise removal operation is stopped. As a result, above the frequency fsa, the signal has a distortion Da, and below the frequency fsa, the signal distortion can be less than or equal to Da, and no inversion phenomenon occurs.

Furthermore, when the noise level increases from noise Na that causes signal distortion Da to noise Nb that causes signal distortion Db, the frequency at which the inversion phenomenon occurs changes from fsa to fsb, and the noise removal operation starts from this frequency fsb. control.

As a result, although the maximum signal distortion increases from Da to Db, it can be reliably suppressed to below the signal distortion Db.

Such a noise removal operation may be performed by controlling the detection level of the noise detection circuit depending on the signal frequency.

By the way, human hearing senses can distinguish between noises even if they are at the same level if the audio signal level is smaller than the noise level, but as the audio signal level increases, it becomes impossible to distinguish between the noises. . Namely.

Masking effect works.

Conventional pulse noise removal removes noise regardless of the audio signal level, so if there is an audio signal, there is no problem even if small pulse noise is not removed, but if the audio signal is small ,for example,
In a state where there is no signal, the small pulse noise is noticeable.

In addition to preventing the above-mentioned inversion phenomenon caused by changes in audio signal frequency, the present invention can also prevent inversion phenomena with respect to changes in the signal level by comparing pulse noise and audio signal level.

Figure 3 shows this operation. When the audio signal input to the gate circuit is vsl and the frequency is fst, the pulse noise detection level is set to VDz, and switching is performed for small pulse noise that causes an inversion phenomenon. Try not to let it happen. When the input level of the audio signal increases to VSx, the noise detection level is raised to Vpl to prevent switching even for pulse noise that is larger than when the input is vsi.

When the frequency of the audio signal is increased from fs to fsz, the detection level is increased from VDL to v01' even at the input level of vs, and is set so as not to detect small noises. As a result, it is possible to prevent an inversion phenomenon with respect to both changes in signal frequency and level.

FIG. 1 shows a configuration diagram of an embodiment of a pulse noise removing device according to the present invention, in which 10 is an audio signal filter, 11 and 12 are variable attenuators, 13 is an audio amplifier, and 14 is an audio AGC circuit. It is.

In FIG. 1, the input signal v1 is expressed by the following equation.

7□= v5+ vo...(1) However, vs: audio signal component vn: pulse noise component At this time, detect noise for the noise component vn that satisfies the following (2) formula, and perform a noise removal operation. .

This K indicates the ratio between the audio signal component vs(K) and the pulse noise component vn(K) when the inversion phenomenon begins to occur.

The audio signal filter 10 is a filter for changing K depending on the frequency fs of the audio signal, and receives the output signal of the gate circuit 6 as input. The high-pass filter 1 is for eliminating the audio signal component VS and obtaining only the noise component VQ, and receives the input signal of the gate circuit 6.

The outputs of filters 1 and 10 are passed through variable attenuators 11 and 12, respectively;
2. Audio amplifier 13 and audio AGC circuit 1
It is controlled by a signal through 4.

If the voltage gains of the variable attenuators 11 and 12 are Go, then when the audio signal component v s:':Q, GO=1
As VS increases, Go approaches 0.

The AGC circuit 4 is provided to prevent the gate circuit 6 from being continuously cut off (to prevent the gate circuit output VQ from stopping) when the pulse noise period becomes extremely small, and does not normally operate.

Now, assuming that the loss of the noise component Vl caused by the filter 1 is 0, the output Vna of the amplifier 3 is expressed by the following equation (4).

vna=vn 0Go 1IGvn −(4)
However, Go: voltage gain of variable attenuator 11.12 Gvn: gain of amplifier 3 Next, when the audio signal component vs is input, the output vsa of the audio amplifier 13 is as follows: vsa=vs-Go-
Gz·Gva=va ”(5) However, G□: Voltage gain of audio signal filter 10 Gva: Gain of audio amplifier 13 va: Detection level of audio AGC circuit 14. Therefore, the following equation (6) is obtained. .

Therefore, the following equation (7) can be obtained from equations (4) and (6).

Incidentally, it is assumed that when the output of the noise amplifier 3 reaches VC, the monostable multi-circuit 5 operates to provide the gate circuit 6 with a predetermined switching pulse for noise removal.

a is expressed as the following equation (8).

Furthermore, when vn = vn (KL vs - v5 (K), it is assumed that the inversion phenomenon begins to occur, and vna = VC. Therefore, from equations (3) and (8), the following (
9) Equation is obtained.

In formula (9), G y a HG y nt K
If 1 is fixed and G1 is changed depending on the frequency, the value of K can be changed with the frequency.

As explained in Fig. 3, in order to prevent the inversion phenomenon, K should be increased as the audio frequency increases;
It is necessary to lower the noise detection sensitivity. Therefore, if a high-pass filter is used as the audio signal filter 10, this operation can be performed.

Further, when the audio frequency is fixed, an appropriate noise removal operation can be performed depending on the noise level and audio signal level.

Therefore, if a noise removal operation that satisfies equation (9) is performed using the apparatus shown in FIG. 1, the inversion phenomenon can be eliminated.

In the embodiment shown in FIG. 1, the audio signal filter 1
Although the input signal to 0 is taken from the output of the gate circuit, it goes without saying that the invention is not limited to this and may be taken from its input.

In addition, in an FM receiver, the noise removal device shown in Fig. 1 performs noise removal at the composite signal stage, but as shown in Fig. 4, the inversion phenomenon is removed using the stereo demodulated signal of the FM multiplex. It becomes possible to provide an audio signal for improvement. That is, in the embodiment shown in FIG. 4, the output of the FM tuner 15 is input to the gate circuit 6 and the filter 1, and the stereo signals LC:H and RCH, which are the outputs of the gate circuit 6, are synthesized by the synthesis circuit 16. , is input to the audio signal filter 1o.

[Effects of the Invention] As described above, according to the present invention, the following effects can be achieved.

(1) Reversal phenomena can be prevented with respect to changes in both the level and frequency of audio signals.

(2) Since the detection sensitivity of pulsed noise is controlled by the level of the audio signal, switching malfunctions that occur when the audio signal increases in particular (harmonics of the audio signal are determined to be mixed with pulsed noise and noise removal is performed) ) can be prevented.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of an embodiment of a pulse noise removing device according to the present invention, FIGS. 2 and 3 are explanatory diagrams for explaining the present invention in detail, and FIG. 4 is a block diagram of a pulse noise removing device according to the present invention. FIG. 3 is a configuration diagram of another embodiment of the removal device. FIG. 5 is a block diagram of a conventional pulse noise removing device, and FIGS. 6 to 8 are explanatory diagrams for explaining the operation of FIG. 5. 1... High-pass filter, 4... Monostable multi circuit. 6... Gate circuit, 10... Audio signal filter, 11.12... Variable attenuator. 14...Audio AGC circuit. Patent applicant Clarion Co., Ltd. Figure 1 fsa fsb audio signal Zhoutang 4! K fs Figure 6 Figure 8 Procedural amendment June 2, 1986

Claims (2)

[Claims] (1) Noise detection means for extracting a noise signal component in an input signal, gate means for switching the input signal, audio signal detection means for extracting an audio signal component contained in the input signal, and the extracted noise a first level variable means for varying the signal level of the signal component; a second level variable means for varying the signal level of the extracted audio signal component; and an audio signal varied by the second level variable means. level control means for controlling the first and second level variable means according to the component;
and gate control means for controlling switching of the gate means in accordance with the noise signal component varied by the level variable means. (2) The gate control means includes noise amplification means for amplifying the noise signal component varied by the first level variable means, and gain control means for controlling the gain of the amplification means according to the output of the amplification means. and pulse generating means for generating a pulse for controlling switching of said gate means in accordance with the output of said amplifying means.