JP2005233642A - Signal discrimination method and signal discriminating device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To discriminate a sine wave having arbitrary frequency having a simple configuration. <P>SOLUTION: A signal x(t) to be measured is sampled by an A/D converter 21, and the ratio of the sum of former and latter sample values to the middle sample value, or the ratio of the difference between the product of the former and latter sample values to the square of the middle sample value are calculated by a calculation part 22 relative to three sample values X(k), X(k-1), X(k-2) which are continuous in time series and acquired in the sampling, and the identity as to the acquired ratio or to the difference is determined by a determination part 27, to thereby discriminate whether the signal x(t) to be measured is a sine wave. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a technique for identifying a sine wave signal with a simple configuration.

  For example, in various devices (vibration detection devices and the like) that process output signals such as sensors, it may be necessary to distinguish between sine waves having periodicity and continuity and signals of other waveforms.

  As a general method for identifying whether or not an input signal is a sine wave having a constant frequency f, a signal to be identified is input to a narrow-band bandpass filter whose frequency is the center frequency of the pass and output to the filter. If there is, there is a method of determining the input signal as a sine wave signal of frequency f.

  However, as described above, in the method of identifying a sine wave using a narrow-band bandpass filter, when there are a large number of identification target frequencies, it is necessary to provide a filter for each frequency, which complicates the configuration. . Moreover, even if the waveform of the input signal is a rectangular wave, a triangular wave, or the like, if the frequency of the fundamental wave component is equal to f, a certain level of sine wave is output from the filter, and the distinction with respect to the frequency is High, but the waveform itself is low.

  In order to obtain this waveform discriminability, the band pass filter and a band elimination filter that greatly attenuates only a desired frequency component and passes other frequency components are used in combination, and the signal under measurement is input to two filters. When there is a signal output only to the band pass filter, a configuration for determining a sine wave is also conceivable, but in this case, two filters are required for each frequency, and this is a configuration in which this is prepared for the frequency. The configuration becomes complicated.

  An object of the present invention is to solve the above-described problems and provide a technique for identifying a sine wave with a simple configuration.

In order to achieve the object, the signal identification method of claim 1 of the present invention comprises:
Obtaining an instantaneous amplitude value of the signal under measurement every predetermined time as a sample value;
Calculating the ratio of the sum of the previous and subsequent sample values and the intermediate sample value for each set of three consecutive sample values in time series for the acquired sample value sequence;
Whether or not the signal under measurement is a sine wave is identified by determining whether or not the ratio of each set is equal.

The signal identification method according to claim 2 of the present invention includes:
Obtaining an instantaneous amplitude value of the signal under measurement every predetermined time as a sample value;
Calculating a difference between a product of preceding and succeeding sample values and a square of an intermediate sample value for each set of three consecutive sample values in time series for the acquired sample value sequence;
Whether or not the signal under measurement is a sine wave is identified by determining whether or not the difference of each set is equal.

The signal identification method of claim 3 of the present invention is:
Obtaining an instantaneous amplitude value of the signal under measurement every predetermined time as a sample value;
Four sample values X (k), X (k-1) continuous in time series with respect to the acquired sample value sequence
, X (k-2), X (k-3) for each set,
X (k) = X (k-1) [X (k-1) + X (k-3)] / X (k-2) -X (k-2)
Determining whether or not
Whether or not the signal under measurement is a sine wave is discriminated by determining whether or not the above-described equation holds.

A signal identification device according to claim 4 of the present invention is
An A / D converter (21) for sampling a signal under measurement at a predetermined period and outputting a digital sample value sequence;
An arithmetic unit that calculates a ratio between the sum of previous and subsequent sample values and an intermediate sample value for each set of three consecutive sample values in the sample value sequence output from the A / D converter. (22)
A determination unit (27) for determining whether or not the ratio for each set calculated by the calculation unit is equal;
Whether the signal under measurement is a sine wave is identified based on a determination result of the determination unit.

A signal identification device according to claim 5 of the present invention is
An A / D converter (21) for sampling a signal under measurement at a predetermined period and outputting a digital sample value sequence;
In the sample value sequence output from the A / D converter, the difference between the product of the preceding and succeeding sample values and the square of the intermediate sample value is calculated for each set of three consecutive sample values in time series. A calculation unit (22);
A determination unit (27) for determining whether or not the difference for each set calculated by the calculation unit is equal;
Whether the signal under measurement is a sine wave is identified based on a determination result of the determination unit.

A signal identification device according to claim 6 of the present invention is
An A / D converter (21) for sampling a signal under measurement at a predetermined period and outputting a digital sample value sequence;
Of the sample value sequence output from the A / D converter, four sample values X (k) continuous in time series
, X (k-1), X (k-2), X (k-3) for each of the three sample values excluding the latest sample value,
X (k-1) [X (k-1) + X (k-3)] / X (k-2) -X (k-2)
An arithmetic unit (22 ′) for performing
A determination unit (27) for determining whether the calculation result by the calculation unit is equal to the latest sample value of the set;
Whether the signal under measurement is a sine wave is identified based on a determination result of the determination unit.

  As described above, the signal identification method and apparatus according to the present invention determines the identity of the ratio or difference obtained by the predetermined calculation for the three sample values consecutive in the time series obtained by sampling the signal under measurement. Since whether or not the signal under measurement is a sine wave is identified, a sine wave having an arbitrary frequency can be identified with a simple configuration.

  In addition, for the three sample values consecutive in time series obtained by sampling the signal under measurement, except for the latest sample value, the identity between the value obtained by the predetermined calculation and the latest sample value is determined. Thus, since it is identified whether or not the signal under measurement is a sine wave, a sine wave having an arbitrary frequency can be identified with a simple configuration.

  Hereinafter, embodiments of a signal identification device of the present invention will be described with reference to the drawings. First, the principle of a signal identification method for identifying a sine wave of the present invention will be described.

  A sine wave signal x (t) having a frequency f, an amplitude A, and an initial phase θ is expressed by the following equation.

  x (t) = A sin (2πft + θ) (1)

Further, a sample value sequence X (n) obtained by sampling the sine wave signal x (t) with a period T.
Is represented by the following equation (n = 0, 1, 2,...).

  X (n) = A sin (2πfnT + θ) (2)

  Here, sample values X (k−2), X (k−1), and X (k) of three consecutive points in time series where n = k−2, k−1, and k are as follows: expressed.

X (k−2) = A sin [2πf (k−2) T + θ]
X (k−1) = A sin [2πf (k−1) T + θ]
X (k) = A sin (2πfkT + θ)

  Of the above three consecutive sample values, the sum of the sample values before and after is expressed as follows.

X (k) + X (k-2)
= A sin (2πfkT + θ) + A sin [2πf (k−2) T + θ]
= 2A sin [2πf (k−1) T + θ] cos (2πfT)
= 2X (k-1) cos (2πfT) (3)

  Here, 2 cos (2πfT) is a constant C that does not depend on k, and the above equation (3) holds regardless of the value of k. Therefore, if it is a sine wave, the following formula will be materialized about arbitrary k.

  X (k) + X (k-2) = CX (k-1) (4)

The maximum value of the period T for satisfying the sampling theorem is the signal x (t)
In this case, the positive peak value and the negative peak value of the signal are alternately sampled, or 0 is continuously sampled to characterize the sine waveform. May not be sampled.

Therefore, the condition for determining the sine wave in the above equation (4) is that the sampling period T is smaller than 1/2 of the period 1 / f of the signal x (t), that is,
| C | = | 2 cos (2πfT) | <2 (5)
Is required.

  Therefore, if the above equation (4) is steadily established for three consecutive sample values obtained by sampling with the period T, the signal that is the basis of the sample value sequence is determined to be a sine wave of frequency f. be able to.

  However, the constant C is a value determined by the sampling period T and the signal frequency f. When the signal frequency f is unknown, the constant C is not fixed.

  However, the fact that the above equation (4) is constantly established means that the following equations (6a) and (6a ′) are established, that is, the sum of two sample values before and after the three sample values. The ratio to the intermediate sample value is the same for different sets of sample values.

C = [X (k) + X (k-2)] / X (k-1)
= [X (k + 1) + X (k-1)] / X (k)
= [X (k + 2) + X (k)] / X (k + 1)
………
(6a)

1 / C = X (k-1) / [X (k) + X (k-2)]
= X (k) / [X (k + 1) + X (k-1)]
= X (k + 1) / [X (k + 2) + X (k)]
………
...... (6a ')

  Therefore, if the equations (6a) and (6a ′) are continuously established, it can be determined as a sine wave regardless of the signal frequency f.

  Although the above formulas (6a) and (6a ′) cannot be calculated with the denominator 0, in this case, the calculation result may be ignored and the determination based on the next calculation result may be performed.

  Further, if the following equations (6b) and (6b ′) obtained by modifying the above equations (6a) and (6a ′) are used, identification is possible even if the sample value is zero. In this case, for three sets of sample values, the difference between the square of the intermediate sample values and the product of the two sample values before and after is equal for each set.

X (k-1) ^2- X (k) X (k-2)
= X (k) ² -X (k + 1) X (k-1)
= X (k + 1) ² -X (k + 2) X (k)
…………
(6b)

X (k) X (k-2) -X (k-1) ²
= X (k + 1) X (k-1) -X (k) ²
= X (k + 2) X (k) -X (k + 1) ²
…………
(6b)

Also, four consecutive sample values X (k) to X (k-3) in the above formula (6a)
The following equation (6c) is obtained by solving the equation defined by:

X (k) = X (k-1) [X (k-1) + X (k-3)] / X (k-2)
-X (k-2)
...... (6c)

  This equation (6c) shows that the result of calculating the right-hand side using three sample values of the four consecutive sample values excluding the latest one is equal to the latest sample value. If the above equation (6c) is constantly established for successive sample values, the signal can be determined as a sine wave.

FIG. 1 shows a configuration of a signal identification device 20 based on the above principle.
In FIG. 1, an A / D converter 21 samples a signal under measurement x (t) in synchronization with a clock signal Ck having a period T, and measures the signal under measurement x (t).
Is converted to a digital sample value sequence X (n) indicating an instantaneous amplitude value at each time T and output to the arithmetic unit 22.

  The arithmetic unit 22 stores the sample value sequence X (n) in the first latch circuit 23 and the second latch circuit 24 while sequentially shifting in synchronization with the clock signal Ck, and calculates the stored value together with the input value. It outputs to the circuit 26.

The calculation circuit 26 calculates the latest three sample values X (k) in the sample value sequence X (n).
, X (k-1), X (k-2), and performs any one of the above formulas (6a), (6a '), (6b), (6b'), and the result Y ( k) is output to the determination unit 27.

  When the calculation circuit 26 uses the equations (6a) and (6a ′), the calculation cannot be performed when the intermediate sample value X (k−1) is 0. Therefore, the intermediate sample value X (k−1) ) Is also output to the determination unit 27.

The determination unit 27 calculates the operation result Y (k) when the intermediate sample value X (k-1) is not 0.
Are sequentially stored and compared with the previously stored calculation result Y (k-1), and when both are in an equal state, the input signal x (t) indicates that it is a sine wave, for example, 1 (high Level) identification signal H is output, and when it is not equal or when the equal state does not continue, the input signal x (t)
For example, an identification signal H of 0 (low level) indicating that is not a sine wave is output.

In addition, the determination unit 27 calculates the calculation result Y (k) when the intermediate sample value X (k−1) is 0.
Is ignored and the next calculation result Y (k + 1) is determined.

  Note that this determination mode may be performed according to the application. Under ideal conditions that do not consider noise and errors included in the calculation, if the state where the calculation result Y is equal is interrupted even once, it is determined that it is not a sine wave. However, in actual signals and circuits, there are noises and errors that are included in the calculation. Taking these errors into account, for example, when comparing the calculation results Y, an allowable error range is provided, and within that range Compare both.

  When an identification signal indicating a sine wave is output from the determination unit 27, the following expression is obtained from the calculation result Y on the right side (or left side) and the known sampling period T in the calculation of Expression (6a). By (7), the frequency f of the input sine wave signal can be obtained.

f = (1 / 2πT) cos ⁻¹ (Y / 2) (7)

  The signal identification device 20 ′ shown in FIG. 2 performs sine wave identification using the above equation (6c). A third latch circuit 25 is added to the calculation unit 22 ′, and the calculation circuit 26 ′ The right side of Expression (6c) is calculated for the stored values of the three latch circuits 23 to 25, and the calculation result Z and the latest output value X (k) of the A / D converter 21 are input to the determination unit 27. Then, it is determined whether or not both match continuously, and the sine wave is identified based on the determination result.

Even in this configuration, when the intermediate sample value X (k−2) stored in the second latch circuit 24 becomes 0, the determination becomes impossible. The above operation is performed when the stored value is not 0. When the stored value is 0, the operation result Z (k)
And the output value X (k) of the A / D converter 21 are ignored, and the determination is made based on the calculation result Z (k + 1) of the next timing and the output value X (k + 1) of the A / D converter 21. I do.

  In any of the signal identification devices 20 and 20 'having the above-described configuration, the calculation for the determination is only a simple four-calculation calculation, and if three or four consecutive sample values are obtained, the sine wave can be identified. It is possible, and a signal waveform of an arbitrary frequency can be identified almost in real time with a simple configuration.

The figure which shows the structure of embodiment of the signal identification apparatus of this invention. The figure which shows the structure of other embodiment of the signal identification apparatus of this invention.

Explanation of symbols

20, 20 '... signal identification device, 21 ... A / D converter, 22, 22' ... arithmetic unit,
23 to 25: Latch circuit, 26: Calculation circuit, 27: Determination unit

Claims (6)

Obtaining an instantaneous amplitude value of the signal under measurement every predetermined time as a sample value;
Calculating the ratio of the sum of the previous and subsequent sample values and the intermediate sample value for each set of three consecutive sample values in time series for the acquired sample value sequence;
A signal identification method for identifying whether or not the signal under measurement is a sine wave by determining whether or not the ratio of each set is equal. Obtaining an instantaneous amplitude value of the signal under measurement every predetermined time as a sample value;
Calculating a difference between a product of preceding and succeeding sample values and a square of an intermediate sample value for each set of three consecutive sample values in time series for the acquired sample value sequence;
A signal identification method for identifying whether or not the signal under measurement is a sine wave by determining whether or not the difference of each set is equal. Obtaining an instantaneous amplitude value of the signal under measurement every predetermined time as a sample value;
Four sample values X (k), X (k-1) continuous in time series with respect to the acquired sample value sequence
, X (k-2), X (k-3) for each set, X (k) = X (k-1) [X (k-1) + X (k-3)] / X (k -2) -X (k-2)
Determining whether or not
A signal identification method for discriminating whether or not the signal under measurement is a sine wave by determining whether or not the expression is continuously established. An A / D converter (21) for sampling a signal under measurement at a predetermined period and outputting a digital sample value sequence;
An arithmetic unit that calculates a ratio between the sum of previous and subsequent sample values and an intermediate sample value for each set of three consecutive sample values in the sample value sequence output from the A / D converter. (22)
A determination unit (27) for determining whether or not the ratio for each set calculated by the calculation unit is equal;
A signal identification device for identifying whether or not the signal under measurement is a sine wave based on a determination result of the determination unit. An A / D converter (21) for sampling a signal under measurement at a predetermined period and outputting a digital sample value sequence;
In the sample value sequence output from the A / D converter, the difference between the product of the preceding and succeeding sample values and the square of the intermediate sample value is calculated for each set of three consecutive sample values in time series. A calculation unit (22);
A determination unit (27) for determining whether or not the difference for each set calculated by the calculation unit is equal;
A signal identification device for identifying whether or not the signal under measurement is a sine wave based on a determination result of the determination unit. An A / D converter (21) for sampling a signal under measurement at a predetermined period and outputting a digital sample value sequence;
Of the sample value sequence output from the A / D converter, four sample values X (k) continuous in time series
, X (k-1), X (k-2), X (k-3) for each of the three sample values excluding the latest sample value,
X (k-1) [X (k-1) + X (k-3)] / X (k-2) -X (k-2)
An arithmetic unit (22 ′) for performing
A determination unit (27) for determining whether the calculation result by the calculation unit is equal to the latest sample value of the set;
A signal identification device for identifying whether or not the signal under measurement is a sine wave based on a determination result of the determination unit.

Images