JP3399274B2 - Tone signal detector - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a tone signal detector for detecting a tone signal transmitted on a communication line or the like.

[0002]

FIG. 22 shows, for example, Japanese Patent Laid-Open No. 63-2364.
FIG. 1 is a configuration diagram showing a conventional tone signal detector disclosed in Japanese Patent Laid-Open No. 60-61, in which 1 is a DFT for performing a discrete Fourier transform (hereinafter referred to as DFT) on an input signal to extract electric power W1 having a frequency component f1. An arithmetic unit 2 performs a DFT operation on the input signal to extract an electric power W2 having a frequency component f2.
The calculator 3 integrates the input signal to obtain the power Wb of the input signal.
The integrator 4 for calculating the power ratio W2 / W1 of the power W2 with respect to the power W1 and the power ratio Wb / W2 of the power Wb with respect to the power W2 are compared with predetermined threshold values a and b, respectively. It is a comparator that determines the presence or absence of a signal.

Next, the operation will be described. A signal transmitted using a telephone line may include a voice signal and the like in addition to a modem signal transmitted and received by a facsimile machine. Therefore, when a facsimile machine or the like transmits / receives a modem signal, it is necessary to detect a training signal appearing at the head of the modem signal and then execute processing such as demodulation in the subsequent stage. FIG. 23 is an explanatory diagram showing an example of the power spectrum of a signal transmitted using a telephone line. While the average power spectrum of voice is represented by the solid line in the figure, V.V. The power spectrum of the training signal of the 29 modem is represented by the dotted line in the figure. As can be seen from the figure, this training signal is 0.5 kHz.
z is substantially equivalent to a superposition of three tone signals having power components of respective frequencies of 1.7 kHz and 2.9 kHz.

As is clear from FIG. 23, in the vicinity of the frequency of 1 kHz, the power component of the voice signal is relatively large, whereas the power component of the tone signal does not exist. On the other hand, when the frequency is near 2.9 kHz, the power component of the tone signal is relatively large,
The feature is that the power component of the audio signal is small.

Therefore, the conventional tone signal detector detects a tone signal as follows by utilizing such characteristics.

First, the DFT calculator 1 performs a DFT operation on an input signal to extract electric power W1 having a frequency of 1 kHz from the electric power components in the frequency domain.

Further, the DFT calculator 2 executes the DFT calculation on the input signal and extracts the electric power W2 having a frequency of 2.9 kHz from the electric power components in the frequency domain.

The integrator 3 integrates the input signal,
The power Wb of the input signal is calculated. When the electric powers W1, W2 and Wb are obtained in this way, the comparator 4 obtains the electric power ratio W2 / W1 of the electric power W2 to the electric power W1 and the electric power ratio Wb / W2 of the electric power Wb to the electric power W2.

When the power ratio W2 / W1 is larger than the threshold value a and the power ratio Wb / W2 is smaller than the threshold value b, the comparator 4 determines that the input signal has the above-mentioned characteristics. Since it can be determined, it is determined that the tone signal has arrived, and a signal indicating that is output.

[0010]

In conventional telephone communication, for example, ringback tone (ringing tone), dial tone (dialing tone), busy tone (busy tone) and the like are used. Ringback tones are generally tone signals with amplitude modulation, and dial tones and busy tones are tone signals without amplitude modulation. As described above, for example, tone signals used in telephone communication include signals with amplitude modulation and signals without amplitude modulation. Since the conventional tone signal detector has the configuration described above, it is possible to detect that a tone signal has been input for any of the ringback tone, dial tone, and busy tone signals. It could not be detected whether it was a ringback tone (amplitude modulated signal), dial tone, or busy tone (amplitude modulated signal). Therefore, it has been impossible to detect only a specific tone of the tone signals used, for example, in a prior art device for telephone communication.

The present invention has been made to solve the above problems, and an object thereof is to obtain a tone signal detector for detecting a specific tone (for example, ringback tone) with a simple structure.

[0012]

According to a first aspect of the present invention, there is provided a tone signal detector comprising a signal strength calculating means for calculating the strength of an input signal and at least one delay for delaying the output of the signal strength calculating means. A buffer, a first feature quantity extracting means for extracting a feature quantity of an input signal based on the output of the signal strength calculating means and the output of the delay buffer, and an output of the signal strength calculating means and an output of the delay buffer. Based on the output of the second characteristic amount extracting means for extracting the other characteristic amount of the input signal, the output of the first characteristic amount extracting means, and the output of the second characteristic amount extracting means, the amplitude modulation is performed. It also comprises a determination means for determining the presence or absence of a tone signal.

A tone signal detector according to a second aspect of the present invention is a tone signal detector which selects and passes only a predetermined frequency component of an input signal, and the predetermined frequency component based on the output of the frequency selecting device. Of the input signal based on the output of the signal strength calculating means and the output of the delay buffer, the signal strength calculating means for calculating the strength of the signal, the at least one delay buffer for delaying the output of the signal strength calculating means, A first feature amount extracting means for extracting a quantity, a second feature amount extracting means for extracting another feature amount of the input signal based on the output of the signal strength calculating means and the output of the delay buffer; It is provided with a judging means for judging the presence or absence of the amplitude-modulated tone signal based on the output of the first characteristic quantity extracting means and the output of the second characteristic quantity extracting means.

According to a third aspect of the present invention, in the tone signal detector, the frequency selecting means includes a bandpass filter.

According to a fourth aspect of the present invention, in the tone signal detector, the frequency selecting means includes a DFT (discrete Fourier transform) calculating means.

According to a fifth aspect of the present invention, in the tone signal detector, the signal strength calculating means includes a square value calculating circuit.

According to a sixth aspect of the present invention, in the tone signal detector, the signal strength calculating means is provided with an absolute value calculating circuit.

According to a seventh aspect of the present invention, in the tone signal detector, the signal strength calculating means is provided with a detection circuit.

According to an eighth aspect of the present invention, in the tone signal detector, the signal strength calculation means integrates the output of the square value calculation circuit, the absolute value calculation circuit or the detection circuit. is there.

According to a ninth aspect of the present invention, in the tone signal detector, the signal strength calculating means smooths the output of the square value calculating circuit, the absolute value calculating circuit or the detection circuit. Is.

According to a tenth aspect of the present invention, the tone signal detector has a moving average circuit in the signal strength calculating means.

According to an eleventh aspect of the present invention, in the tone signal detector, the first feature quantity extracting means includes a maximum value calculating circuit, and the second feature quantity extracting means includes:
It is provided with a minimum value calculation circuit.

According to a twelfth aspect of the present invention, in the tone signal detector, the first characteristic amount extracting means is provided with a maximum value calculating circuit, and the second characteristic amount extracting means comprises:
It is provided with an average value calculation circuit.

According to a thirteenth aspect of the present invention, in the tone signal detector, the first feature quantity extracting means includes an average value calculating circuit, and the second feature quantity extracting means includes:
It is provided with a minimum value calculation circuit.

According to a fourteenth aspect of the present invention, in the tone signal detector, the first feature quantity extracting means selects the Nth largest value of the signals inputted to the first feature quantity extracting means. Then, the second characteristic amount extracting means selects and outputs the M-th smallest value among the signals input to the second characteristic amount extracting means.

According to a fifteenth aspect of the present invention, in the tone signal detector, the judgment means calculates a difference between the output of the first feature quantity extraction means and the output of the second feature quantity extraction means, The presence or absence of an amplitude-modulated tone signal is determined based on the value of this difference.

According to the sixteenth aspect of the present invention, in the tone signal detector, the determining means calculates a ratio of the output of the first characteristic amount extracting means and the output of the second characteristic amount extracting means, The presence or absence of the amplitude-modulated tone signal is determined based on the value of this ratio.

In the tone signal detector according to the seventeenth aspect of the present invention, in the judging means, the difference or ratio between the output of the first characteristic amount extracting means and the output of the second characteristic amount extracting means is calculated. Calculate and compare this difference or ratio value with a predetermined threshold value, and if the condition that this difference or ratio value exceeds the threshold value continues for a predetermined time or more, it is determined that there is an amplitude-modulated tone signal. It was done like this.

The tone signal detector according to the eighteenth aspect of the present invention is a tone signal detector which selects and passes only a predetermined frequency component of an input signal, and the predetermined frequency component based on the output of the frequency selecting device. Signal strength calculating means for calculating the signal strength of the input signal, second signal strength calculating means for calculating the signal strength of the input signal, and at least one delay for delaying the output of the first signal strength calculating means. A buffer, a first characteristic amount extracting means for extracting a characteristic amount of an input signal based on an output of the first signal strength calculating means and an output of the delay buffer, and an output of the first signal strength calculating means. Based on the output of the delay buffer, a second feature amount extracting means for extracting another feature amount of the input signal, an output of the first feature amount extracting means, and an output of the second feature amount extracting means. Basis And a determination means for determining the presence or absence of an amplitude-modulated tone signal, wherein the determination means is the difference between the output of the first characteristic amount extraction means and the output of the second characteristic amount extraction means, or the ratio. Is calculated and the value of the difference or ratio is compared with a predetermined threshold value determined based on the output of the second signal strength calculation means, and the presence or absence of the amplitude-modulated tone signal is determined based on the comparison result. It is something that is done.

According to a nineteenth aspect of the present invention, in the tone signal detector, the predetermined frequency component is selected based on the output of the frequency selecting means for selecting and passing only the predetermined frequency component of the input signal and the output of the frequency selecting means. Signal strength calculation means for calculating the signal strength of the input signal, second signal strength calculation means for calculating the signal strength of the input signal, and output of the first signal strength calculation means for the second signal A normalizing means for normalizing with the output of the intensity calculating means, at least one delay buffer for delaying the output of the normalizing means, and a feature amount of the input signal based on the output of the normalizing means and the output of the delay buffer. A first feature quantity extracting means for extracting the other feature quantity of the input signal based on the output of the normalizing means and the output of the delay buffer, and the first feature quantity extracting means for extracting the other feature quantity of the input signal. The output of the symptoms extracting unit, the second
The determination means for determining the presence or absence of the amplitude-modulated tone signal is provided based on the output of the feature amount extraction means.

[0031]

DETAILED DESCRIPTION OF THE INVENTION

Embodiment 1. FIG. 1 is a block diagram showing an embodiment of a tone signal detector according to the present invention.
00 is frequency selecting means for selecting and passing only a predetermined frequency component of the input signal, 101 is frequency selecting means 10
Signal strength calculation means for calculating the signal strength S of the input signal based on the output of 0, 102-1, 102-2, 102-
3 is a delay buffer for delaying the signal strength S output from the signal strength calculation means 101, and 103 is the signal strength calculation means 10.
1 output signal strength S, delay buffers 102-1 and 1
The first input signal based on the outputs of 02-2 and 102-3
Feature amount extraction means for extracting the feature amount C1 of
Is a second characteristic amount extracting means for extracting the second characteristic amount C2 of the input signal based on the signal intensity S output from the signal intensity calculating means 101 and the outputs from the delay buffers 102-1, 102-2, 102-3, Reference numeral 105 denotes a determination unit that determines the presence or absence of an amplitude modulation tone based on the first characteristic amount C1 and the second characteristic amount C2.

Next, the operation will be described. The frequency selecting means 100 selects and passes only a predetermined frequency component of the input signal. More specifically, for example, a bandpass filter can be used as the frequency selection unit 100. The characteristics of this band pass filter must be designed so that only the frequency components of the tone signal to be detected are passed and signals in other bands are blocked. The purpose of processing the input signal with this frequency selection means 100 is:
When a noise signal other than the tone signal of the frequency to be detected is mixed, the noise component is removed and the detection process is performed so that the noise can be detected without being affected by the noise.

The signal strength calculation means 101 calculates the signal strength S of the input signal based on the output of the frequency selection means 100. FIG. 2 is a configuration diagram showing an example of a detailed configuration of the signal strength calculation means 101. In the figure, 110 is a multiplier and 111 is an integrator. The multiplier 110 squares the output value of the frequency selection means 100 to calculate the power of the input signal and outputs the squared value. The integrator 111 adds the squared value output from the multiplier 110 by the number of n samples and outputs the added value in order to obtain information on the change in the power of the input signal due to the amplitude modulation. The value of n is preferably selected within a range that is sufficiently larger than the period of the carrier of the tone signal to be detected and sufficiently smaller than the period of amplitude modulation. With such a configuration, as the signal strength S, a value obtained by integrating the squared value of the output of the frequency selection unit 100 within the interval of the above n samples is obtained. The signal strength calculation means 101 uses the above n.
Each sample is used as one processing unit, and the signal strength S is calculated every time n samples are input, and the result is output. Hereinafter, the processing unit of the n samples will be referred to as a processing frame.

The delay buffer 102-1 delays the signal strength S output from the signal strength calculation means 101, the delay buffer 102-2 further delays the output of the delay buffer 102-1 and the delay buffer 102-3 is a delay buffer. 10
The output of 2-2 is further delayed. This delay buffer 1
The delay time in 02-1, 102-2, 102-3 needs to be equal to the time T corresponding to the processing frame.

The first feature quantity extraction means 103 is the signal strength S output from the signal strength calculation means 101, and the delay buffer 10
The first characteristic amount C1 of the input signal is extracted based on the outputs of 2-1, 102-2, and 102-3. More specifically, for example, a maximum value calculation circuit can be used as the first feature amount extraction means 103. When the maximum value calculation circuit is used, the signal intensity S is the first feature amount C1. , The maximum value of the outputs of the delay buffers 102-1, 102-2 and 102-3 is obtained.

The second characteristic amount extraction means 104 is the signal strength S output from the signal strength calculation means 101, and the delay buffer 10
The second feature amount C2 of the input signal is extracted based on the outputs of 2-1, 102-2, and 102-3. More specifically, for example, a minimum value calculating circuit can be used as the second characteristic amount extracting means 104. When the minimum value calculating circuit is used, the signal strength S is the second characteristic amount C2. , The minimum value of the outputs of the delay buffers 102-1, 102-2 and 102-3 is obtained.

The judging means 105 is the first feature quantity extracting means 1
The presence or absence of an amplitude modulation tone is determined based on the first feature amount C1 output from the output port 03 and the second feature amount C2 output from the second feature amount extracting means. FIG. 3 is a configuration diagram showing an example of a detailed configuration of the determination means 105. In the figure, 120 is a subtractor, 122 is a comparator, 123-1, 123-2, 1
Reference numeral 23-3 is a delay buffer, and 124 is an AND circuit.
The subtractor 120 uses the first characteristic amount C1 to the second characteristic amount C1.
Subtract two. The comparator 122 compares the output of the subtractor 120 with a predetermined threshold L. Delay buffers 123-1 and 12
3-2 and 123-3 delay the output of the comparator 122. These delay buffers 123-1, 123-2, 123
The delay time at -3 is equal to the time T corresponding to the processing frame. The logical product circuit 124 includes the comparator 12
2, delay buffers 123-1, 123-2, 123-3
When the output of the subtractor 120 exceeds the threshold value L for four consecutive frames, it is determined that there is amplitude modulation and "1" is output, and if not,
It judges that there is no amplitude modulation and outputs "0".

4 and 5 are explanatory views for explaining the operation of the tone signal detector according to this embodiment.
5 shows the output of each section when a tone signal not amplitude-modulated is given as an input signal, and FIG. 5 shows the output of each section when an amplitude-modulated tone signal is given as an input signal.

As shown in FIG. 4, when a tone signal that is not amplitude-modulated is applied to this tone signal detector,
The frequency selection means 100 removes the noise component mixed in this tone signal and passes only the tone signal of the frequency to be detected. The signal strength calculation means 101 calculates the signal strength S for the output of the frequency selection means 100. However, since the input tone signal is not amplitude-modulated, the signal strength S is the tone signal input. Maintain a certain size throughout the life. Delay buffers 102-1 and 1
The outputs of 02-2 and 102-3 are waveforms obtained by delaying the signal strength S by time T, 2 × T, and 3 × T, respectively.
The first feature amount extraction means 103 calculates the signal strength S and the maximum value of the outputs of the delay buffers 102-1, 102-2, 102-3, and the second feature amount extraction means 104 calculates the signal strength S, Delay buffers 102-1, 102-2, 102-3
Calculate the minimum value of the output of. Therefore, the determination means 10
5 is the output of the subtractor 120, that is, the first feature amount C1.
As shown in the figure, the difference value between the second feature amount C2 and the second feature amount C2 takes a large value for 3 × T after the tone signal is input. The state in which the output of 120 exceeds the threshold value L does not continue for four frames, and the detection result is "0" (no amplitude modulation) for this tone signal input which is not amplitude modulated.

Further, as shown in FIG. 5, when an amplitude-modulated tone signal is applied to this tone signal detector,
The frequency selection means 100 removes the noise component mixed in this tone signal and passes only the tone signal of the frequency to be detected. The signal strength calculation means 101 calculates the signal strength S with respect to the output of the frequency selection means 100. However, since the input tone signal is amplitude-modulated, the signal strength S is the input signal accompanying the amplitude modulation. Increases or decreases with changes in the power of. Delay buffer 102-
The outputs of 1, 102-2, 102-3 are the signal strength S,
The waveforms are delayed by time T, 2 × T, and 3 × T, respectively. The first feature amount extraction means 103 calculates the signal strength S and the maximum value of the outputs of the delay buffers 102-1, 102-2, 102-3, and the second feature amount extraction means 104 calculates the signal strength S, Delay buffers 102-1, 102-2, 10
Calculate the minimum of 2-3 outputs. Therefore, the output of the subtractor 120 in the determination unit 105, that is, the difference value between the first characteristic amount C1 and the second characteristic amount C2 is, as shown in the figure, compared with the case where amplitude modulation is not applied, Since it takes a large value, the state in which the output of the subtractor 120 exceeds the threshold value L continues, and the detection result becomes "1" (with amplitude modulation) for this tone-modulated tone signal input.

According to the above-mentioned structure, after the input signal is processed by the frequency selecting means, the signal strength S is calculated by the signal strength calculating means, and the signal strength S and the value obtained by delaying the signal strength S The presence or absence of amplitude modulation of the tone signal can be determined by extracting the first feature amount C1 and the second feature amount C2 and making a determination based on the first feature amount C1 and the second feature amount C2. There is an effect that can be done.

Further, when a bandpass filter is used as the frequency selecting means, when a noise signal is mixed in other than the tone signal of the frequency to be detected, the noise component is removed and the detection processing is performed, whereby noise is removed. There is an effect that can be detected without being affected by.

Further, when the signal strength calculating means calculates the squared value, there is an effect that the power of the input signal can be accurately calculated.

In addition, when the signal strength calculation means integrates the squared value, there is an effect that it is possible to extract information relating to the change in the power of the input signal due to the amplitude modulation.

When the maximum value is calculated by the first characteristic amount extraction means and the minimum value is calculated by the second characteristic amount extraction means, the characteristic amount that can be used for determining the presence or absence of amplitude modulation of the input signal is determined. There is an effect that it can be extracted.

Further, when the determination means makes a determination based on the difference between the first characteristic amount and the second characteristic amount, and when the input signal is an unmodulated tone signal, this difference value is small and the input When the signal is an amplitude-modulated tone signal, it is possible to determine the presence or absence of amplitude modulation by using the property that the difference value becomes large.

Further, the judging means judges that there is an amplitude-modulated tone signal when the condition that the difference between the first characteristic amount and the second characteristic amount exceeds the threshold value continues for a predetermined time or longer. In this case, the presence or absence of amplitude modulation can be determined more accurately.

Embodiment 2. If the S / N ratio of the input signal is sufficiently high, the frequency selecting means 100 is not always necessary, and the input signal is directly calculated as the signal strength calculating means 101.
Even if the configuration is such that the signal strength is calculated, it is possible to determine the presence or absence of amplitude modulation of the tone signal, as in the first embodiment.

Embodiment 3. Further, in the first embodiment, the case where the bandpass filter is used as the frequency selecting means has been described, but not limited to the bandpass filter, for example, a discrete Fourier transform (hereinafter referred to as DFT) calculating means is used. Good.

FIG. 6 is a block diagram showing another embodiment of the tone signal detector according to the present invention, in which 100 is frequency selecting means, 101 is signal strength calculating means, 131,
132 is a multiplier and 133 is an adder. The operations other than the frequency selecting means 100 and the signal strength calculating means 101 are the same as those in the first embodiment, and therefore the frequency selecting means 10
0, only the operation of the signal strength calculation means 101 will be described.

The frequency selecting means 100 selects and passes only a predetermined frequency component of the input signal. More specifically, the frequency selecting means 100 is, for example, a DF.
T computing means can be used. This DFT operation means is a sequence of input signal samples x (0), x (1), ...
The DFT operation defined by the following equation is performed on x (n-1).

[0052]

[Equation 1]

In the equation (1), fi is the frequency of the tone signal to be detected, and fs is the sampling frequency.

When the frequency selecting means 100 performs the DFT operation as described above, it outputs the operation result to the signal strength calculating means 101. Since the operation result is usually a complex number, the real number part of the operation result. The output is divided into R and the imaginary part Q.

The frequency selecting means 100 performs a DFT operation every time n samples are input and outputs the operation result. The method of selecting the value of n is the same as in the first embodiment, and it is desirable to select it within a range that is sufficiently larger than the carrier cycle of the tone signal to be detected and sufficiently smaller than the amplitude modulation cycle.

The signal strength calculation means 101 calculates the signal strength of the input signal based on the output of the frequency selection means 100. The multiplier 131 calculates the squared value of the real part R of the output of the frequency selection means 100, and the multiplier 132 calculates the squared value of the imaginary part Q of the output of the frequency selection means 100. The adder 133 adds the output of the multiplier 131 and the output of the multiplier 132. With this configuration, the sum of the squared values of the real number part R and the imaginary number part Q of the output of the frequency selection means 100 is obtained as the output of the signal strength calculation means 101.

With the above-described configuration, as in the first embodiment, when a noise signal other than the tone signal of the frequency to be detected is mixed, the noise component is removed and the detection process is performed. This has the effect of enabling detection without being affected by noise.

Fourth Embodiment In the first embodiment, the signal strength calculating means calculates the square value, but the calculation is not limited to the square value calculation, and the absolute value calculation may be performed, for example.

FIG. 7 is a block diagram showing an example of the detailed configuration of the signal strength calculating means 101. In the figure, 140 is an absolute value circuit and 141 is an integrator. Since the operation other than the signal strength calculation means 101 is the same as that of the first embodiment, only the operation of the signal strength calculation means 101 will be described.

The absolute value circuit 140 comprises the frequency selecting means 10
Calculate the absolute value of the 0 output. FIG. 8 shows this absolute value circuit 1.
It is an explanatory view showing the operation of 40. As shown in the figure, the output value of the absolute value circuit 140 is always zero or a positive value.

The integrator 141 adds the absolute values output from the absolute value circuit 140 by the number of n samples and outputs the added value.

According to this structure, the calculation process of the power of the input signal performed by the signal strength calculation means 101 can be performed more easily than in the first embodiment.

Embodiment 5. Further, in the above-described first embodiment, the case where the signal strength calculation means calculates the squared value is described, but not limited to the calculation of the squared value, for example, detection may be performed.

FIG. 9 is a block diagram showing an example of the detailed configuration of the signal strength calculation means 101. In the figure, 142 is a detection circuit and 143 is an integrator. Signal strength calculation means 1
Since the operations other than 01 are the same as those in the first embodiment, only the operation of the signal strength calculation means 101 will be described.

The detection circuit 142 includes the frequency selecting means 100.
The detection processing is performed on the output of. Further, FIG. 10 is an explanatory diagram showing the operation of the detection circuit 142. As shown in the figure, when the input to the detection circuit 142 is a positive value,
The detection circuit 142 outputs the input value as it is, and outputs a zero level signal when the input to the detection circuit 142 is a negative value, and the output value of the detection circuit 142 is always zero or a positive value. Take

The integrator 143 adds the output value of the detection circuit 142 by the number of n samples and outputs the added value.

According to this structure, the calculation processing of the power of the input signal performed by the signal strength calculation means 101 can be performed more easily than in the first embodiment.

Sixth Embodiment In the first, fourth, or fifth embodiment described above, the signal strength calculation means has been described as one that integrates the squared value, the absolute value, or the detection output. Good.

FIG. 11 is a block diagram showing an example of the detailed structure of the signal strength calculation means 101. In FIG.
Is a multiplier and 144 is a smoothing circuit. Since the operation other than the signal strength calculation means 101 is the same as that of the first embodiment, only the operation of the signal strength calculation means 101 will be described.

The multiplier 110 calculates the squared value of the output of the frequency selecting means 100. The smoothing circuit 144 performs smoothing by processing the squared value output from the multiplier 110 with a low pass filter.

According to this structure, the signal strength calculation means smooths the squared value to extract information about the change in the power of the input signal due to the amplitude modulation, as in the first embodiment. There is an effect that can be.

Embodiment 7. FIG. 12 is a configuration diagram showing an example of a detailed configuration of the signal strength calculation means 101. In the figure, 150 is an absolute value circuit, 151-1, 151-.
2, 151-3 are delay buffers, 152 is an adder, 15
3 is a multiplier. Since the operation other than the signal strength calculation means 101 is the same as that of the first embodiment, only the operation of the signal strength calculation means 101 will be described.

The absolute value circuit 150 includes the frequency selecting means 10
Calculate the absolute value of the 0 output.

Delay buffers 151-1, 151-2, 1
51-3 delays the output of the absolute value circuit 150. The delay time in each of the delay buffers 151-1, 151-2, 151-3 needs to be equal to the sampling frequency of the input signal.

The adder 152 outputs the output of the absolute value circuit 150 and the delay buffers 151-1, 151-2, 151-3.
Add the outputs of.

The multiplier 153 multiplies the output of the adder 152 by a predetermined coefficient P. In this embodiment, since the number of added samples in the adder 152 is 4, the coefficient P to be multiplied in the multiplier 153 is 0.25 which is the reciprocal of the number of added samples.

With this structure, the frequency selecting means 1
A moving average value obtained by averaging the absolute value of the output value of 00 for a plurality of past samples is obtained as the signal strength S output from the signal strength calculation means 101, and as in the first embodiment,
There is an effect that it is possible to extract information regarding a change in the power of the input signal due to the amplitude modulation.

Eighth Embodiment Further, in the first embodiment, the first feature amount extraction means 103 calculates the signal strength S and the maximum value of the outputs of the delay buffers 102-1, 102-2, 102-3, and the second feature amount. The extraction means 104 includes a signal strength S and delay buffers 102-1, 102-2, 102.
Although the minimum value of the output of −3 is calculated, the combination is not limited to the combination of the maximum value and the minimum value, and may be, for example, the combination of the maximum value and the average value.

In the present embodiment, the second feature quantity extraction means 104 has a signal strength S and a delay buffer 102-
The average value of the outputs of 1, 102-2 and 102-3 is calculated. The operation other than the second characteristic amount extraction means 104 is the same as that in the first embodiment.

13 and 14 are explanatory views for explaining the operation of the tone signal detector according to this embodiment.
13 shows the output of each section when a tone signal that is not amplitude-modulated is given as an input signal, and FIG. 14 shows the output of each section when an amplitude-modulated tone signal is given as an input signal. .

As shown in FIG. 13, when a tone signal that is not amplitude-modulated is given to this tone signal detector, the difference value between the first characteristic amount C1 and the second characteristic amount C2 is
As shown in the figure, time 3
Although it takes a large value only during × T, the value after that becomes 0, so that the state where the output of the subtractor 120 exceeds the threshold value L does not continue for 4 frames, and this tone signal that is not amplitude-modulated is used. The detection result is “0” for the input
(No amplitude modulation).

Further, as shown in FIG. 14, when the tone signal whose amplitude is modulated is applied to this tone signal detector, the difference value between the first characteristic amount C1 and the second characteristic amount C2 is
As shown in the figure, as compared with the case where the amplitude modulation is not applied, since it takes a large value, the state where the output of the subtractor 120 exceeds the threshold value L continues, and this amplitude-modulated tone signal input is input. On the other hand, the detection result is "1" (with amplitude modulation).

With the above-mentioned structure, as in the case of the first embodiment, there is an effect that it is possible to extract the feature quantity that can be used for determining whether or not the amplitude modulation of the input signal is present.

Ninth Embodiment Further, in the first embodiment, the first feature amount extraction means 103 calculates the signal strength S and the maximum value of the outputs of the delay buffers 102-1, 102-2, 102-3, and the second feature amount. The extraction means 104 includes a signal strength S and delay buffers 102-1, 102-2, 102.
Although the minimum value of the output of −3 is calculated, the combination is not limited to the combination of the maximum value and the minimum value, and may be, for example, the combination of the average value and the minimum value.

In this embodiment, the first feature quantity extraction means 103 has a signal strength S and a delay buffer 102-
The average value of the outputs of 1, 102-2 and 102-3 is calculated. The operations other than the first feature amount extraction means 103 are the same as those in the first embodiment.

15 and 16 are explanatory views for explaining the operation of the tone signal detector according to this embodiment.
FIG. 15 shows the output of each section when a tone signal that is not amplitude-modulated is given as an input signal, and FIG. 16 shows the output of each section when a tone signal that is amplitude-modulated is given as an input signal. .

As shown in FIG. 15, when a tone signal not amplitude-modulated is given to this tone signal detector, the difference value between the first characteristic amount C1 and the second characteristic amount C2 is
As shown in the figure, time 3
Although it takes a large value only during × T, the value after that becomes 0, so that the state where the output of the subtractor 120 exceeds the threshold value L does not continue for 4 frames, and this tone signal that is not amplitude-modulated is used. The detection result is “0” for the input
(No amplitude modulation).

Further, as shown in FIG. 16, when the tone signal whose amplitude is modulated is applied to this tone signal detector, the difference value between the first characteristic amount C1 and the second characteristic amount C2 is
As shown in the figure, as compared with the case where the amplitude modulation is not applied, since it takes a large value, the state where the output of the subtractor 120 exceeds the threshold value L continues, and this amplitude-modulated tone signal input is input. On the other hand, the detection result is "1" (with amplitude modulation).

With the above-mentioned structure, as in the case of the first embodiment, there is an effect that it is possible to extract the feature quantity that can be used for determining the presence or absence of amplitude modulation of the input signal.

[Embodiment 10] FIG. 17 is a block diagram showing another embodiment of the tone signal detector according to the present invention. In the figure, 100 is frequency selecting means for selecting and passing only a predetermined frequency component of the input signal, and 101 is a frequency. Signal strength calculation means 102-1 and 102-1 for calculating the signal strength of the input signal based on the output of the selection means 100.
2-2, 102-3, and 102-4 are signal strength calculation means 1
A delay buffer for delaying the output of 01, 161 is an output of the signal strength calculation means 101, delay buffers 102-1 and 1
A data rearrangement circuit that rearranges the outputs of 02-2, 102-3, and 102-4 in descending order of value, and 105 is a determination unit that determines the presence or absence of an amplitude modulation tone based on the output of the data rearrangement circuit 161. .

The operations of the frequency selecting means 100 and the signal strength calculating means 101 are the same as those in the first embodiment, and therefore their explanations are omitted.

The delay buffer 102-1 delays the output of the signal strength calculation means 101, the delay buffer 102-2 further delays the output of the delay buffer 102-1 and the delay buffer 102-3 of the delay buffer 102-2. The output is further delayed, and the delay buffer 102-4 further delays the output of the delay buffer 102-3. The delay buffers 102-1, 102-2, 102-3, 102-4
The delay time in is equal to the time T corresponding to the processing frame.

The data rearrangement circuit 161 outputs the output of the signal strength calculation means 101 and the delay buffers 102-1 and 102.
-2, 102-3, 102-4 are rearranged in descending order of value, and signal lines S161-1, S161-2, S1 are rearranged.
61-3, S161-4, and S161-5. Here, it is assumed that the data output to the signal line S161-1 has the maximum value and the data output to the signal line S161-5 has the minimum value.

Of the outputs of the data rearrangement circuit 161,
The data output to the signal line S161-2, that is, the value of the second largest signal strength, and the data output to the signal line S161-4, that is, the value of the second smallest signal strength are supplied to the determination means 105. It

The determination means 105 is the data rearrangement circuit 1
Of the outputs of 61, the signal line S161-2 and the signal line S1
Based on the data output to 61-4, the presence / absence of an amplitude modulation tone is determined.

The determination criterion of the presence / absence of the amplitude modulation tone in the determining means 105 is the same as that in the first embodiment, and the second largest signal strength output to the signal line S161-2 is the first characteristic amount C1. Value of the second feature amount C
As 2, the value of the second smallest signal strength output to the signal line S161-4 is used.

According to this structure, by using the data excluding the maximum value and the minimum value in the output of the signal strength calculation means 101, the influence of noise or the like can be reduced and the amplitude modulation can be performed more accurately. There is an effect that the presence or absence can be determined.

Eleventh Embodiment In the above embodiment, the determination means determines the presence or absence of amplitude modulation by using the second largest signal strength value and the second smallest signal strength value. However, the present invention is not limited to this. More generally, even if the determination is performed using the N-th largest signal strength value and the M-th smallest signal strength value, the same effect as in the above-described embodiment can be obtained.

Twelfth Embodiment In the first embodiment,
In the determination means 105, from the first feature amount C1 to the second
Although the feature amount C2 is subtracted and the presence or absence of amplitude modulation is determined based on the comparison result between the subtraction result and a predetermined threshold value L, the present invention is not limited to the subtraction, and for example, division may be performed.

FIG. 18 is a block diagram showing an example of the detailed structure of the judging means 105. In FIG. 18, 125 is a divider,
122 is a comparator, 123-1, 123-2, 123-3
Is a delay buffer, and 124 is an AND circuit. Since the operation other than the determining means 105 is the same as that of the first embodiment, only the operation of the determining means 105 will be described.

The divider 125 divides the first characteristic amount C1 by the second characteristic amount C2. The comparator 122 is the divider 12
The output of 5 is compared with a predetermined threshold L. Delay buffer 12
3-1, 123-2, 123-3 delay the output of the comparator 122. These delay buffers 123-1 and 123
The delay time at −2, 123-3 is equal to the time T corresponding to the processing frame. AND circuit 124
Is a comparator 122 and delay buffers 123-1 and 123-
If the output of the divider 125 exceeds the threshold value L for four consecutive frames, the logical product of the outputs of 2 and 123-3 is calculated, and it is determined that the amplitude modulation is present, and "1" is output, and otherwise. , Determines that there is no amplitude modulation and outputs "0".

According to this structure, the output of the divider 125 does not depend on the signal level of the input signal, so that the presence or absence of amplitude modulation can be favorably detected regardless of the signal level of the input signal.

Thirteenth Embodiment In the first embodiment,
In the comparator 122 in the judging means 105, the subtractor 12
Although the output value of 0 is compared with the predetermined threshold value L, the threshold value may be adaptively determined according to the level of the input signal.

FIG. 19 is a block diagram showing another embodiment of the tone signal detector according to the present invention.
Reference numeral 100 is a frequency selecting means for selecting and passing only a predetermined frequency component of the input signal, and 101-1 is a first signal strength for calculating a first signal strength S1 of the input signal based on the output of the frequency selecting means 100. Calculating means 101-2 is a second signal strength calculating means for calculating the second signal strength S2 of the input signal, and 102-1, 102-2 and 102-3 are first signal strength calculating means 101-1. A delay buffer for delaying the output first signal strength S1, 103 is the first signal strength S output from the first signal strength calculation means 101-1.
1, delay buffers 102-1, 102-2, 102-3
Based on the output of the first signal C1 of the input signal based on the output of the first feature amount extraction means 104, the first signal strength calculation means 101-1 outputs the first signal strength S1, the delay buffer 102- Second characteristic amount extraction means for extracting the second characteristic amount C2 of the input signal based on the outputs of 1, 102-2, and 102-3, and 105 is the first characteristic amount C1, the second characteristic amount C2, and the second characteristic amount C2. It is a judging means for judging the presence or absence of the amplitude modulation tone based on the signal strength S2 of 2.

Next, the operation will be described. The frequency selecting means 100 selects and passes only a predetermined frequency component of the input signal.

The first signal strength calculating means 101-1 calculates the first signal strength S1 of the input signal based on the output of the frequency selecting means 100. More specifically, as the first signal strength calculation means 101-1, for example, as in the first embodiment, the squared value of the output of the frequency selection means 100 is used as the first signal strength S1. , It is sufficient to obtain an integrated value within the interval of the n number of samples. Further, as in the case of the first embodiment, it is desirable to select the value of n within a range that is sufficiently larger than the carrier cycle of the tone signal to be detected and sufficiently smaller than the amplitude modulation cycle.

The second signal strength calculation means 101-2 calculates the second signal strength S2 of the input signal. More specifically, as the second signal strength calculation means 101-2, for example, the square value of the input signal to the tone signal detector is
It suffices that the integrated value be obtained within the interval of m number of samples. The value of m is preferably selected to be sufficiently larger than the amplitude modulation cycle of the tone signal to be detected. By selecting the value of m in this way, the second signal strength S2 averages the effect of increase or decrease in the power of the input signal due to the amplitude modulation even when a tone signal with amplitude modulation is given as the input signal. The signal has a relatively small temporal variation.

Delay buffers 102-1, 102-2, 1
02-3, the first feature amount extraction means 103, and the second feature amount extraction means 104 are the same as those in the first embodiment.

The judging means 105 judges the presence / absence of an amplitude modulation tone based on the first characteristic amount C1, the second characteristic amount C2, and the second signal intensity S2 output from the second signal intensity calculating means 101-2. To do.

FIG. 20 is a block diagram showing an example of the detailed configuration of the judging means 105. In the figure, 120 is a subtractor, 1
Reference numeral 22 is a comparator, 123-1, 123-2 and 123-3 are delay buffers, 124 is an AND circuit, and 126 is a threshold setting circuit.

Subtractor 120, delay buffer 123-1,
123-2, 123-3, comparator 122, AND circuit 1
Since the operation of 24 is similar to that of the first embodiment, the operation of the threshold setting circuit 126 will be mainly described.

The threshold value setting circuit 126 sets the threshold value L used in the comparator 122 based on the second signal strength S2 output from the second signal strength calculation means 101-2. More specifically, the threshold setting circuit 126 multiplies, for example, the second signal strength S2 by a predetermined coefficient, and the value is calculated by the comparator 12
The threshold value L used in 2 is output.

According to this structure, the threshold value setting circuit sets the threshold value based on the output of the second signal strength calculating means, thereby setting an appropriate threshold value regardless of the signal level of the input signal. Thus, there is an effect that the presence or absence of amplitude modulation can be favorably detected.

Fourteenth Embodiment FIG. 21 is a block diagram showing another embodiment of the tone signal detector according to the present invention. In the figure, 100 is frequency selecting means for selecting and passing only a predetermined frequency component of the input signal, 101-
1 is a first signal strength calculating means for calculating the first signal strength S1 of the input signal based on the output of the frequency selecting means 100, and 101-2 is a second signal strength calculating means for calculating the second signal strength S2 of the input signal. The signal strength calculation means 106 is the first signal strength S1 output from the first signal strength calculation means 101-1.
And the second output from the second signal strength calculation means 101-2.
Normalizing means for obtaining the value of the normalized signal strength based on the signal strength S2 of
-3 is a delay buffer for delaying the output of the normalizing means 106, 103 is an output of the normalizing means 106, the delay buffer 1
02-1, 102-2, 102-3, a first feature quantity extracting means for extracting the first feature quantity C1 of the input signal based on the outputs of the input signals 02-1, 102-2, 102-3; 104 is an output of the normalizing means 106; Second characteristic amount extraction means for extracting the second characteristic amount C2 of the input signal based on the outputs of 102-2 and 102-3, and 105 is the first characteristic amount C1 and the second characteristic amount C2.
It is a judging means for judging the presence or absence of the amplitude modulation tone based on the above.

Next, the operation will be described. The operations of the frequency selecting means 100, the first signal strength calculating means 101-1 and the second signal strength calculating means 101-2 are the same as those in the thirteenth embodiment.

The normalizing means 106 includes the first signal strength S1 output from the first signal strength calculating means 101-1 and the second signal strength S1.
Based on the second signal strength S2 output from the signal strength calculation means 101-2, the normalized signal strength value is obtained. More specifically, the normalizing means 106 divides the first signal strength S1 by the second signal strength S2, for example, and outputs the divided value.

Delay buffers 102-1, 102-2, 1
02-3, the first feature amount extraction means 103, the second feature amount extraction means 104, and the determination means 105 are the same as those in the first embodiment.

According to this structure, the output of the normalizing means does not depend on the signal level of the input signal by dividing the first signal strength by the second signal strength. However, there is an effect that the presence or absence of amplitude modulation can be detected well.

[0119]

As described above, according to the present invention, it is possible to obtain a tone signal detector for detecting only a specific tone signal by simply detecting the presence or absence of amplitude modulation.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing a tone signal detector according to a first embodiment of the present invention.

FIG. 2 is a configuration diagram showing a signal strength calculation means of the tone signal detector according to the first embodiment of the present invention.

FIG. 3 is a configuration diagram showing a judging means of the tone signal detector according to the first embodiment of the present invention.

FIG. 4 is an explanatory diagram illustrating an operation of the tone signal detector according to the first embodiment of the present invention.

FIG. 5 is an explanatory diagram illustrating an operation of the tone signal detector according to the first embodiment of the present invention.

FIG. 6 is a configuration diagram showing frequency selecting means and signal strength calculating means of a tone signal detector according to a third embodiment of the present invention.

FIG. 7 is a configuration diagram showing a signal strength calculation means of a tone signal detector according to a fourth embodiment of the present invention.

FIG. 8 is an explanatory diagram illustrating an operation of the tone signal detector according to the fourth embodiment of the present invention.

FIG. 9 is a configuration diagram showing a signal strength calculating means of a tone signal detector according to a fifth embodiment of the present invention.

FIG. 10 is an explanatory diagram illustrating an operation of the tone signal detector according to the fifth embodiment of the present invention.

FIG. 11 is a configuration diagram showing a signal strength calculating means of a tone signal detector according to a sixth embodiment of the present invention.

FIG. 12 is a configuration diagram showing a signal strength calculating means of a tone signal detector according to a seventh embodiment of the present invention.

FIG. 13 is an explanatory diagram illustrating an operation of the tone signal detector according to the eighth embodiment of the present invention.

FIG. 14 is an explanatory diagram illustrating an operation of the tone signal detector according to the eighth embodiment of the present invention.

FIG. 15 is an explanatory diagram illustrating an operation of the tone signal detector according to the ninth embodiment of the present invention.

FIG. 16 is an explanatory diagram illustrating an operation of the tone signal detector according to the ninth embodiment of the present invention.

FIG. 17 is a configuration diagram showing a tone signal detector according to a tenth embodiment of the present invention.

FIG. 18 is a configuration diagram showing a judging means of a tone signal detector according to Embodiment 12 of the present invention.

FIG. 19 is a configuration diagram showing a tone signal detector according to a thirteenth embodiment of the present invention.

FIG. 20 is a configuration diagram showing a judging means of a tone signal detector according to Embodiment 13 of the present invention.

FIG. 21 is a configuration diagram showing a tone signal detector according to a fourteenth embodiment of the present invention.

FIG. 22 is a configuration diagram showing a conventional tone signal detector.

FIG. 23 is an explanatory diagram illustrating an operation of a conventional tone signal detector.

[Explanation of symbols]

100 frequency selection means 101 signal strength calculation means 101-1 first signal strength calculation means 101-2 second signal strength calculation means 102-1, 102-2, 102-3, 102-4 delay buffer 103 first Feature amount extraction means 104 Second feature amount extraction means 105 Judgment means

─────────────────────────────────────────────────── --- Continuation of the front page (56) References JP-A-3-250961 (JP, A) JP-A-5-300551 (JP, A) JP-A-8-237696 (JP, A) JP-A-8- 265457 (JP, A) JP 63-236460 (JP, A) JP 63-236099 (JP, A) International Publication 95/20272 (WO, A1) (58) Fields investigated (Int. Cl. ⁷⁾ , DB name) H04M 11/00-11/10

Claims (19)

(57) [Claims] 1. A signal strength calculation means for calculating the strength of an input signal, at least one delay buffer for delaying the output of the signal strength calculation means, and an output of the signal strength calculation means and an output of the delay buffer. A first feature amount extracting means for extracting a feature amount of the input signal, and a second feature amount extracting means for extracting another feature amount of the input signal based on the output of the signal strength calculating means and the output of the delay buffer. Means and
A tone signal detector comprising: a determination unit that determines the presence or absence of an amplitude-modulated tone signal based on the output of the first characteristic amount extraction unit and the output of the second characteristic amount extraction unit. . 2. A frequency selecting means for selecting and passing only a predetermined frequency component of the input signal, and a signal strength calculating means for calculating the strength of the signal of the predetermined frequency component based on the output of the frequency selecting means. At least one delay buffer for delaying the output of the signal strength calculation means, and first feature quantity extraction means for extracting the feature quantity of the input signal based on the output of the signal strength calculation means and the output of the delay buffer. A second feature quantity extracting means for extracting another feature quantity of the input signal based on the output of the signal strength calculating means and the output of the delay buffer; an output of the first feature quantity extracting means; 2. A tone signal detector comprising: a determination unit that determines the presence or absence of an amplitude-modulated tone signal based on the output of the feature amount extraction unit 2. 3. The tone signal detector according to claim 2, wherein the frequency selecting means includes a bandpass filter. 4. The tone signal detector according to claim 2, wherein the frequency selecting means includes a DFT (discrete Fourier transform) calculating means. 5. The tone signal detector according to claim 1, wherein the signal strength calculation means includes a square value calculation circuit. 6. The tone signal detector according to claim 1 or 2, wherein the signal strength calculation means includes an absolute value calculation circuit. 7. The tone signal detector according to claim 1, wherein the signal strength calculation means includes a detection circuit. Wherein said signal intensity calculating means is characterized by integrating the output of the square value calculating circuit or the absolute value calculation circuit or the detection circuit, 5 to claim 7 Neu
Tone signal detector according to Zureka. 9. The signal strength calculation means smoothes the output of the square value calculation circuit, the absolute value calculation circuit, or the detection circuit .
The tone signal detector according to any of the above. 10. The tone signal detector according to claim 1, wherein the signal strength calculation means includes a moving average circuit. 11. The first feature quantity extraction means comprises a maximum value calculation circuit, and the second feature quantity extraction means comprises a minimum value calculation circuit. The tone signal detector according to item 2. 12. The first feature quantity extraction means comprises a maximum value calculation circuit, and the second feature quantity extraction means comprises an average value calculation circuit. The tone signal detector according to item 2. 13. The first feature quantity extraction means comprises an average value calculation circuit, and the second feature quantity extraction means comprises a minimum value calculation circuit. The tone signal detector according to item 2. 14. The first feature quantity extraction means is the first feature quantity extraction means.
Of the signals input to the feature amount extraction means, the N-th largest value is selected and output, and the second feature amount extraction means selects one of the signals input to the second feature amount extraction means. , The M-th smallest value is selected and outputted, and the tone signal detector according to claim 1 or 2. 15. The determining means calculates the difference between the output of the first feature quantity extracting means and the output of the second feature quantity extracting means, and the amplitude-modulated tone is calculated based on the difference value. The tone signal detector according to claim 1 or 2, wherein the presence or absence of a signal is determined. 16. The determination means calculates a ratio between the output of the first feature quantity extraction means and the output of the second feature quantity extraction means, and the amplitude-modulated tone is calculated based on the value of the ratio. The tone signal detector according to claim 1 or 2, wherein the presence or absence of a signal is determined. 17. The difference between the output of the first characteristic amount extracting means and the output of the second characteristic amount extracting means,
Alternatively, if a ratio is calculated and the value of the difference or ratio is compared with a predetermined threshold value, and the condition that the value of the difference or ratio exceeds the threshold value continues for a predetermined time or more, there is an amplitude-modulated tone signal. The tone signal detector according to claim 1 or 2, characterized in that 18. A frequency selecting means for selecting and passing only a predetermined frequency component of an input signal, and a first signal strength for calculating the strength of the signal of the predetermined frequency component based on the output of the frequency selecting means. The calculating means, the second signal strength calculating means for calculating the signal strength of the input signal, the at least one delay buffer for delaying the output of the first signal strength calculating means, and the first signal strength calculating means. A first feature quantity extracting means for extracting a feature quantity of the input signal based on the output and the output of the delay buffer; and an input signal other than the input signal based on the output of the first signal strength calculating means and the output of the delay buffer. Based on the output of the first feature amount extracting means and the output of the second feature amount extracting means, the presence / absence of the amplitude-modulated tone signal is determined. Judge A determining unit is provided, and the determining unit calculates a difference or a ratio between the output of the first characteristic amount extracting unit and the output of the second characteristic amount extracting unit, and calculates the value of the difference or the ratio as described above. A tone signal detector characterized in that it is compared with a predetermined threshold value determined based on the output of the second signal strength calculating means, and the presence or absence of an amplitude-modulated tone signal is determined based on the comparison result. 19. A frequency selecting means for selecting and passing only a predetermined frequency component of an input signal, and a first signal strength for calculating the strength of the signal of the predetermined frequency component based on the output of the frequency selecting means. Normalizing means for normalizing the output of the calculating means, the second signal strength calculating means for calculating the signal strength of the input signal, and the output of the first signal strength calculating means with the output of the second signal strength calculating means. And at least one delay buffer for delaying the output of the normalizing means, first feature amount extracting means for extracting the feature amount of the input signal based on the output of the normalizing means and the output of the delay buffer, Second feature quantity extracting means for extracting another feature quantity of the input signal based on the output of the normalizing means and the output of the delay buffer, the output of the first feature quantity extracting means, and the second feature quantity extracting means. Feature extraction hand Based on the output, the tone signal detector, characterized in that it comprises determination means for determining whether the amplitude modulated tone signal.