JP2000183766A - Noise eliminator and noise elimination method - Google Patents

Abstract

(57) [Problem] To provide a noise eliminator capable of faithfully reproducing an original signal with a relatively small amount of calculation. SOLUTION: A demodulation circuit 3 for demodulating a signal from an antenna 1, a noise detection circuit 4 for detecting a noise of the signal demodulated by the demodulation circuit 3, and an A for converting the signal demodulated by the demodulation circuit 3 into a digital signal. A / D converter 5 and a noise elimination circuit 6 for receiving an output signal of the A / D converter 5. The noise elimination circuit 6 receives a signal from the noise detection circuit 4, immediately before and immediately after the noise detection period, and performs noise detection. At the sample time in the period, an operation for calculating the original signal before the noise is mixed is performed, and the calculated original signal data is replaced with a signal containing noise in the noise detection period.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a noise elimination device, and more particularly to a radio noise elimination device for controlling pulse noise generated in a vehicle.

[0002]

2. Description of the Related Art Various control systems (eg, a power window, a wiper, and a door mirror) in an automobile are mostly electronic, generate pulse noise during operation, and have an adverse effect on a car radio (noise source). ).

[0003] Electronic control of automobiles is progressing year by year, and the influence of noise generated therefrom on radio is becoming more serious. It is necessary to enhance the function of the control means for the detected pulse noise.

[0004]

In the prior art,
There were the following problems. (1) Sample-and-hold system When a pulsed external noise is detected from a demodulated radio audio signal, a sample-and-hold circuit operates to hold the previous value of the signal and control the noise. In this method, the distortion of the signal waveform due to the use of the sample and hold circuit is large and remains as noise. In the case of an analog sample-and-hold circuit, the previous value holding (holding) time is determined by an external CR, so that the time is not always appropriate.

(2) Digital filter system In the same manner as the above-mentioned system (1), when pulse noise is detected, the filter operates to control the noise.
In this method, since the filter itself performs the same operation as the moving average, the noise component spreads before and after the noise generation period. And ON / OF of filter processing
In order to fill the joint of the signal by F, processing such as fade-in / fade-out is required, and the amount of calculation becomes large.

An object of the present invention is to provide a noise elimination device and a noise elimination method capable of faithfully reproducing an original signal with a relatively small amount of calculation.

[0007]

According to a first aspect of the present invention, there is provided a noise removing apparatus for demodulating a signal from an antenna, a noise detecting circuit for detecting noise of a signal demodulated by the demodulating circuit, and a demodulating circuit. A noise removal circuit that receives the signal demodulated by the noise detection circuit, receives the signal from the noise detection circuit, and immediately before and after the noise detection period and at a sample time within the noise detection period, before the noise is mixed. An operation for calculating an original signal is performed, and the calculated original signal data is replaced with a signal containing noise in a noise detection period.

According to a second aspect of the present invention, there is provided a noise removing device for demodulating a signal from an antenna, a noise detecting circuit for detecting a noise of the signal demodulated by the demodulating circuit, and a signal for demodulating the signal demodulated by the demodulating circuit. An A / D converter that converts the signal into a digital signal; and a noise removal circuit that receives an output signal of the A / D converter. The noise removal circuit receives a signal from the noise detection circuit, Immediately afterwards and at a sample time within the noise detection period, an operation for calculating the original signal before the noise is mixed is performed, and the calculated original signal data is replaced with a signal containing noise in the noise detection period.

In the noise elimination method according to the third invention, noise of the received signal is detected in time series, and the value of the original signal before the noise is superimposed on the data before and after the noise detection period is predicted. is there.

In the noise elimination method according to a fourth aspect of the present invention, a step of detecting pulse noise of a received signal in time series, a step of calculating a tangent of a sample value immediately before the detection period, and a step of calculating a tangent of the sample value immediately after the detection period Calculating a straight line passing through the data value of the sample value immediately before the detection period and the data value of the sample value immediately after the detection period, calculating an average value of the tangent line and the straight line, and calculating the original signal value. And a step of replacing the calculated original signal value with noise-containing data of the received signal.

In the noise elimination method according to a fifth aspect of the present invention, a tangent line at the sample value immediately before the detection period is defined as a point between two points of the data value at the immediately preceding sample value and the data value at the immediately preceding sample value. In addition to the calculation based on the slope, the tangent at the sample value immediately after the detection period is calculated based on the slope between two points of the data value at the sample value immediately after the detection period and the data value at the sample value immediately after the sample value.

[0012]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiment 1 FIG. 1 shows an overall configuration diagram according to the first embodiment of the present invention. In the figure, 1 is an antenna, 2 is an FM front end circuit, 3
Is an FM-IM (intermediate frequency) demodulation circuit, 4 is a noise detection circuit, 5 is an A / D converter, 6 is a noise removal circuit, 7 is D / D
A converter 8 is an output terminal.

The reception signal tuned by the front-end circuit 2 functioning as a tuner is converted to an FM-IM demodulation circuit 3
The noise detection circuit 4 performs noise detection on the signal demodulated by the above. The noise detection circuit 4 performs detection of a noise detection period, that is, detection of a noise detection start time and a noise detection end time.

On the other hand, the signal demodulated by the FM-IM demodulation circuit 3 is converted into a digital signal by an A / D converter 5, and the A / D converted sampling data is applied to a noise removal circuit 6. The pulse noise removal processing is performed.

The noise elimination circuit 6 receives the signal from the noise detection circuit 4 and immediately before and immediately after the noise detection period and at a sample time within the noise detection period.
An operation for calculating an original signal before noise mixing is performed, and the calculated original signal data is replaced with a signal containing noise in a noise detection period to remove noise.

The noise removal processing by the noise removal circuit 6 will be described in detail. FIG. 2 shows a conventional example of a signal before noise removal and a pulse noise removal suppression image according to the present invention. FIG. 2A shows a signal waveform before noise removal. 2 (b-1) and 2 (b-
2) shows a signal waveform at the time of noise removal processing in a conventional example. FIG. 2 (b-1) shows noise removal by a sample hold method, and FIG. 2 (b-2) shows noise removal by a filtering method. . FIG. 2C shows a signal waveform at the time of noise removal according to the present invention. FIG. 2 (b-1)
In the conventional noise removal by the sample-and-hold method shown in FIG.
In the noise removal by the conventional filtering method shown in (b-2), the noise component is expanded by averaging exceeding the noise detection period. On the other hand, in the noise removal according to the present invention, the calculated original signal is replaced with a signal containing noise within the noise detection period, and the distortion is extremely small.

The embodiment according to the present invention solves the problems of the prior art because the replacement with the predicted value of the original signal obtained by the operation is performed only during the period in which the noise is detected. (See FIG. 2). sample·
The merit of the conventional method using the hold is that signal distortion due to correction is small (no noise is generated). An advantage of the conventional method by filtering is that noise is not spread because the processing is performed only for the detection period. Here, "noise removal" in the present invention refers to an operation of replacing a sample value in a noise detection period with a data value calculated as an original signal as described above.

Hereinafter, an algorithm for calculating an original signal before noise mixing, which is performed by the noise removing circuit 6 in the embodiment according to the present invention, will be described. Here, FIG. 3 shows a signal image before the noise removal processing. FIG.
Indicates an original signal prediction range before noise mixing. FIG. 5 shows the calculated data values of the original signal.

FIG. 3 shows an image of sample data targeted by generation of pulse noise.

Next, a description will be given of a method of specifying the existence range of the original signal, a method of defining the original signal, and a process of replacing the original signal with a signal containing noise. (1) Method for specifying the range of existence of the original signal (see FIG. 4) For the detection period of the pulse noise, a tangent B to the signal at the immediately preceding sample value and the straight line C connected by the two preceding and following sample values The original signal always exists in the area sandwiched by the symbols.

(2) Method for defining the original signal (see FIG. 5) The average signal is obtained from the average value of the value of the straight line C at each sample time in the noise detection period and the value of one of the tangents A and B at the same sample time. The value of the original signal to be taken. However, the selection condition of the tangents A and B is to select a tangent closer to the straight line C at each sample time.

In FIG. 5, Y3L to Y5L are the values of the straight line (C) at each sample time during the noise detection period, and Y3H to Y3L.
Y5H is the value of one of the tangents (A) and (B) at each sample time within the noise detection period, and the original signal values Y3 to Y5 are
This is an original signal value calculated as an average value of the value of the straight line (C) and one of the values of the tangent lines (A) and (B). That is,
The original signal value Yn is calculated by the following equation. Yn = (YnH
+ YnL) / 2

Tangent (A), tangent (B) and straight line (C)
Is determined as follows. a. How to determine the tangent (A) In order to reduce the amount of calculation, the slope a between the two points of the sample values Y1 and Y2 is set as the tangent (A). Slope a = (Y2-Y1) / (X2-X1) = (Y2-Y1) /
Δt b. How to Obtain Tangent Line (B) In order to reduce the amount of computation, the slope b between the two points of the sample values Y6 and Y7 is defined as the tangent line (B). Slope b = (Y7−Y6) / (X7−X6) = (Y7−Y6) /
Δt c. How to determine the straight line (C) The two contact points X2, Y2 and X2, Y6 of the above tangent lines (A) and (B)
Is a straight line (C) having a slope c that passes through. Slope c = (Y6-Y2) / (X6-X2) = (Y6-Y2) /
4Δt Here, Δt is a sampling period.

(3) Replacement processing of original signal and signal containing noise The calculated original signal is replaced with each sample data in the noise detection period, and the pulse noise removal processing is completed. The above processing is performed each time the pulse noise is detected (see FIG. 6), and the noise removal function is realized.

The noise removal processing flow shown in FIG. 6 includes the following steps. First, in step S1, pulse noise is detected by the noise detection circuit 4, and a noise detection period is detected. In step S2, the noise removal circuit 6 calculates a tangent (A) at the sample value immediately before the noise detection period in the received signal waveform. Here, as described above, the tangent line (A) is, as shown in FIG. 5, a point between the data value Y2 at the sample value X2 immediately before the noise detection period and the data value Y1 at the immediately preceding sample value X1. Is calculated by the inclination a of In step S3, the tangent (B) at the sample value immediately after the noise detection period is calculated. Here, as described above, the tangent (B) indicates the data value Y in the sample value X6 immediately after the noise detection period, as shown in FIG.
6 and the data value Y at the next sample value X7
7 is calculated from the slope b between the two points. In step S4, the data values Y3L, Y4 of the straight lines passing through the two contact points X2, Y2 and X2, Y6 (FIG. 5) of the tangents (A), (B).
Calculate L, Y5L. In step S5, each sample X3
, X4, X5, the data values Y3H, Y4H, Y5H after tangent selection (determination) are calculated. In step S6, an original signal value is calculated by averaging the data values Y3H, Y4H, Y5H and the data values Y3L, Y4L, Y5L. In step S7, replacement of the noise mixed data with the original signal value is performed.

According to the embodiment of the present invention, elimination of pulse noise can be realized with an extremely small amount of data and a small amount of computation while solving the problems of the prior art.

[0027]

According to the first aspect of the present invention, it is possible to obtain a noise eliminator capable of faithfully reproducing an original signal with a relatively small amount of calculation.

According to the second aspect, it is possible to obtain a noise eliminator capable of reproducing the original signal more faithfully with a relatively small amount of calculation.

According to the third aspect of the invention, it is possible to obtain a noise elimination method capable of faithfully reproducing the original signal with a relatively small amount of calculation.

According to the fourth aspect, it is possible to obtain a noise elimination device and a noise elimination method capable of reproducing the original signal more faithfully with a relatively small amount of calculation.

According to the fifth aspect, it is possible to obtain a noise elimination device and a noise elimination method capable of reproducing the original signal more faithfully with a smaller amount of calculation.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a system configuration according to an embodiment of the present invention.

FIG. 2 is a diagram showing a pulse noise suppression image in the embodiment according to the present invention and the prior art.

FIG. 3 is a diagram showing a signal image before noise removal processing in the embodiment according to the present invention;

FIG. 4 is a diagram showing an original signal prediction range before noise is mixed in the embodiment according to the present invention;

FIG. 5 is a diagram showing calculated data values of an original signal in the embodiment according to the present invention.

FIG. 6 is a flowchart showing a processing flow in the embodiment according to the present invention;

[Explanation of symbols]

1 antenna, 2 FM front-end circuit, 3 FM
−IM (intermediate frequency) demodulation circuit, 4 noise detection circuit, 5
A / D converter, 6 noise removal circuit, 7 D / A converter, 8 output terminals.

Claims (5)

[Claims] A demodulation circuit for demodulating a signal from an antenna; a noise detection circuit for detecting a noise of the signal demodulated by the demodulation circuit; a noise removing circuit for receiving a signal demodulated by the demodulation circuit; The elimination circuit receives the signal from the noise detection circuit, and performs an operation for calculating an original signal before noise mixing at a sample time immediately before and immediately after the noise detection period and at a sample time within the noise detection period. A noise removing device, wherein data is replaced with a signal containing noise during a noise detection period. 2. A demodulation circuit for demodulating a signal from an antenna, a noise detection circuit for detecting noise of the signal demodulated by the demodulation circuit, and an A / D converter for converting the signal demodulated by the demodulation circuit into a digital signal. And a noise elimination circuit for receiving an output signal of the A / D converter, the noise elimination circuit receiving a signal from the noise detection circuit, and immediately before and immediately after a noise detection period and a sampling time in the noise detection period. A noise removing device for performing an operation for calculating an original signal before noise mixing, and replacing the calculated original signal data with a signal containing noise in a noise detection period. 3. Detecting noise of a received signal in time series,
A noise removal method for predicting the value of an original signal before noise is superimposed on data before and after a noise detection period. 4. A step of detecting pulse noise of a received signal in time series, a step of calculating a tangent with a sample value immediately before the detection period, a step of calculating a tangent with a sample value immediately after the detection period, and a detection period. Calculating a straight line passing through the data value at the immediately preceding sample value and the data value at the sample value immediately after the detection period; calculating an average value of the tangent line and the straight line to calculate an original signal value; A noise elimination method including a step of replacing a value with noise mixed data of a received signal. 5. A tangent line at a sample value immediately before a detection period,
Calculated by the slope between two points between the data value at the immediately preceding sample value and the data value at the immediately preceding sample value, and the tangent at the sample value immediately after the detection period is calculated as the data value at the immediately following sample value. 5. The noise elimination method according to claim 4, wherein the calculation is performed based on a slope between two points between the data value and a data value of a sample value immediately after that.