JPH05340975A - Frequency meter - Google Patents

Abstract

PURPOSE:To make it possible to calculate frequency of digital signal by detecting the number of wares crossing with a reference level within a measuring range from discrete data, determining a total time based on the initial and final cross points of time, and then dividing the the number of waves by the total time. CONSTITUTION:Data is communicated between a CPU 1 and a digital signal processor 3 through a dual RAM 2, wherein the CPU 1 performs screen control and the processor 3 performs various analysises, operations, I/O setting for hard control, and the like based on input data. When frequency measuring operation is started, number of waves(n) crossing with a reference level within a measuring level, and the numbers m1, m2 of sampling data upto points of time t1, t2 when the waves cross initially and finally with the reference level are assumed. m1 and m2 are corrected according to a specific formula and a frequency (f) is operated according to a following formula based on m1, m2, m, and sampling frequency. f=nXfs/(m2-m1). Consequently, frequency measurement can be realized easily through a software.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a frequency measuring device suitable for measuring the frequency of a digital signal (discrete data).

[0002]

2. Description of the Related Art Conventionally, there is a frequency measuring device for measuring the frequency of an analog signal. In this case, when the average frequency to be measured is F, the analog signal San is at the reference level (generally the center level of the waveform) at the gate time T.
The number n of intersections with is counted to obtain F = n / T (see FIG. 12). In FIG. 12, * indicates an intersection.

[0003]

In the above frequency measurement, the number of times n does not include an error, but as is apparent from FIG. 12, the gate time T includes an error (A + B). In order for F to obtain N-digit precision, (A + B) / T <
It must be 1/10 ^N , and T> (A + B) × 10 ^N.

Here, since A and B each have a maximum of 1 cycle, the maximum of (A + B) is 2 cycles. Therefore, unless the gate time T is enormous, a certain degree of accuracy cannot be obtained.

Therefore, conventionally, the gate time T is increased, but the processing time becomes long, and there is the inconvenience that only the frequency of the continuous carrier can be measured because the frequency cannot be measured in a short section.

Further, the conventional frequency measuring device measures the frequency of the analog signal San, and it is impossible to measure the frequency of the digital signal. That is,
This is because the digital signal is discrete data obtained by sampling the analog signal San at a predetermined sampling frequency, and therefore sampling data * that intersects with the reference level is not always obtained (FIG. 13).
reference). In FIG. 13, the solid line shows the measurement frequency signal and the broken line shows the sampling frequency signal.

Therefore, an object of the present invention is to provide a frequency measuring device capable of measuring the frequency of a digital signal. Another object of the present invention is to provide a frequency measuring device that can perform high-precision frequency measurement with a short processing time.

[0008]

SUMMARY OF THE INVENTION According to the present invention, a wave number detecting means for detecting a wave number that crosses a reference level in a measurement range from discrete data, and a reference level that first and last crosses a reference level in a measurement range from discrete data. It comprises a time detecting means for detecting the time point and making the difference the whole time, and a calculating means for dividing the wave number detected by the wave number detecting means by the whole time detected by the time detecting means to obtain a frequency. It is a thing.

For example, the time detecting means obtains the crossing points by linear approximation from the time points corresponding to the discrete data located immediately before and after the crossing with the reference level.

Further, for example, the measurement range and the reference level are set by using a cursor on the screen.

[0011]

In the present invention, since the discrete data (digital signal) is processed to measure the frequency, the frequency measuring operation can be easily realized by software.

[0012] Further, by obtaining the points of time of intersection by linear approximation from the points of time corresponding to the discrete data located immediately before and after the intersection with the reference level, the whole time can be detected with high accuracy in spite of handling the discrete data. You can
It becomes possible to perform frequency measurement with high accuracy.

Further, by setting the measurement range and the reference level on the screen using the cursor, the measurement range and the reference level can be set by moving the cursor while observing the measurement waveform on the screen. The frequency measurement of the target signal can be set appropriately and easily.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram showing the configuration of the embodiment. In the figure, 1 is a CPU and 3 is a DSP
(Digital signal processor), these CP
Data exchange between U1 and DSP3 is via a dual RA
Via M2. The CPU 1 mainly performs screen control and performs parameter setting (editor), data request to the DSP 3, output of calculation result of the DSP 3, and the like. DSP3
Performs various analyzes based on input data, data editing (calculation) for analysis, and I / O setting for hardware control.

A hardware section 5 is connected to the DSP 3 via the I / O port 4. The hardware unit 5 includes a key panel 5a on which various operation keys are arranged, a display 5 for displaying an input waveform, a calculation result of a measurement frequency described later, and the like.
b, a RAM 5c for storing input data and calculation result data of the measurement frequency, and an A / D converter 5d for converting an input analog signal supplied to the input terminal 5e into a digital signal.

In this example, frequency measurements are made from digital signals, and thus discrete data.

First, the theory of frequency measurement using discrete data will be described. As described above, in the case of a digital signal, data that intersects with the reference level is not always obtained. Therefore, as shown in FIG. 2, when the previous data is below the reference level and the current data is above the reference level, it is considered that the intersection has occurred. In FIG. 2, * indicates sampling data.

Here, assuming that the time point of the first crossing in the measurement range is t1, the time point of the last crossing is t2, and the number of waves crossing is n, the frequency f to be obtained is obtained as shown in equation 1.

[0019]

[Equation 1]

Although t1 and t2 are times, the concept of time is difficult to handle in digital processing. Therefore, assuming that the sampling frequency is fs, the number of data sampled by time t1 is m1, and the number of data sampled by time t2 is m2, t1, t
2 becomes as shown in equation 2.

[0021]

[Equation 2]

Therefore, by substituting the equation 2 into the equation 1, the frequency f becomes as shown in the equation 3.

[0023]

[Equation 3]

Next, the number of waves crossed in this example is obtained as follows. In FIG. 3, the signal for one period is from a point a going up from below the reference level to a point c going up from below the reference level, or from a point b going up from below the reference level to It is from the upper level of the reference level to the lower point d. Therefore, the wave number crossing the reference level is obtained as shown in Equation 4.

[0025]

[Equation 4]

In the case of the example of FIG. 3, since there are three points a, c and e which go up from the lower level to the reference level, the wave number n = 3-1 = 2 is obtained from the equation (1) of the equation (4). .. Alternatively, since the points going from the top to the bottom of the reference level are the three points b, d, and f, the wave number n = 3-1 = 2 is obtained from the equation (2) of Equation 4.

The proper use of equations (1) and (2) in equation 4 is determined by the first point. In the case of the example of FIG. 3, since the first point a is the point going from the bottom to the top of the reference level, the equation (1) is used.

By the way, in Equation 3, n and fs have no error, but m is clear as is clear from the method of obtaining t1 and t2.
1 and m2 include an error. Therefore, all the errors of the frequency f obtained by the equation 3 are due to m1 and m2.
From FIG. 4, the error of m1 and m2 is one data maximum, and m
The error frequency of 2-m1 is 1 / (m2-m1).

Therefore, for example, the condition necessary to obtain a precision of 6 digits is as shown in Formula 5 when M = m2-m1. Therefore, to obtain a precision of 6 digits, at least 1
It is necessary to process, 000,000 pieces of data.

[0030]

[Equation 5]

If m1 and m2 can be measured accurately, the error of the frequency f shown in the equation 3 will be small. Therefore, in this example, m1 and m2 are obtained as follows. Since m1 and m2 are obtained in the same manner, the number at the time point t when the reference level N is crossed will be described as m.

As shown in FIG. 5, time t (n-1) (m (n-1)
When the level L of the discrete data p (n-1) of / fs) is below the reference level N and the level M of the discrete data pn of the next time point tn (mn / fs) is above the reference level N, The time point t at which the reference level N is crossed is m / fs.

Since the time points t (n-1) and tn are always adjacent on the discrete time axis, mn = m (n-1) +1.

Therefore, from the proportional relationship, the number m is expressed by the formula 6
The relationship is established.

[0035]

[Equation 6]

Then, in the equation 6, mn = m (n-
By substituting the relationship of 1) +1, the number m is obtained as shown in Expression 7. As is clear from Equation 7, the number m (n-1) of discrete data p (n-1) immediately before the intersection with the reference level N as the measurement data and the level L, and the discrete data pn immediately after the intersection with the reference level N By obtaining the level M of, the number m at the time point t when the reference level N is crossed can be accurately obtained.

[0037]

[Equation 7]

The first term m (n-1) in the equation (7) is an integer term, and the second term | L-N | / (| L-N | + | M-N
|) Becomes a decimal term. From this, the accuracy of the number m is determined only by the second term. That is, the L and M resolutions are directly proportional to the accuracy.

For example, assuming that the discrete data has 8 bits,
The accuracy of the number m is 1 / 256≈0.0039. Therefore, the condition necessary to obtain a precision of 6 digits as the precision of 1 / (m2-m1) is as shown in Formula 8 when M = m2-m1.

[0040]

[Equation 8]

Therefore, in this example, in order to obtain the precision of 6 digits, it is necessary to process 3907 data. However, since the linear approximation is used to obtain the number m as described above (see FIG. 5), the accuracy is somewhat degraded due to the error of the number m.

Next, the frequency measurement operation in this example will be described with reference to the flowchart of FIG. Although not shown,
This frequency measurement operation is performed by, for example, the key panel 5a (see FIG.
The operation is performed based on the parameters set by the operation (see). As the parameters, a sampling frequency fs for sampling the input analog signal, a reference level (threshold value), a time indicating how long the input signal waveform is displayed on the screen of the display, and a measurement range are set.

FIG. 9 shows the main screen 10 of the display 5b.
And the time and sampling frequency fs arranged in association therewith, as well as the display of the measured frequency f. The input signal waveform is displayed on the main screen 10 for the set time. Then, the reference level and the measurement range are arbitrarily set by changing the position and the length of the cursor 11 on the main screen 10.

The frequency measurement operation shown in FIG. 6 is executed in real time based on the parameters set as described above, and the measured frequencies f corresponding to the set parameters are sequentially displayed.

In the flowchart of FIG. 6, first, the wave number n crossing the reference level in the measurement range, the number m1 of sampling data up to the time t1 when the reference level is first crossed in the measurement range, and the last in the measurement range. Then, the number m2 of sampling data up to the time point t2 at which the intersection is crossed is obtained (step 21).

Next, the number to be corrected is set to m1 (step 22), the number 7 is executed to obtain a highly accurate number m (step 23), and the corrected number m is set to m1 (step 2).
4). Further, the number to be corrected is m2 (step 2
5), the equation 7 is executed to obtain a highly accurate number m (step 26), and the corrected number m is set to m2 (step 2).
7).

Then, using the wave number n, the numbers m1 and m2 obtained in the above steps, and the sampling frequency fs, the frequency f is calculated by the equation (3) (step 2).
8). Hereinafter, the operations of steps 21 to 28 will be repeated.

Next, using the flowchart of FIG.
The operation of obtaining the wave number n and the numbers m1 and m2 in step 21 of FIG. 6 will be described.

First, the wave number n is reset to "0" (step 30). Then, one piece of sampling data (discrete data) is read from the RAM 5c (shown in FIG. 1) (step 31), and the data counter is incremented by "1" (step 32).

Next, it is judged whether the read sampling data is lower than the reference level N (step 3).
3). If it is lower than the reference level N, one sampling data is read from the RAM 5c (step 3).
4) Determine whether it is within the measurement range (step 3)
5). If it is within the measurement range, the data counter is incremented by "1" (step 36).

Next, the sampling data is the reference level N.
It is determined whether or not it is higher (step 37). Reference level N
If not, turn on the flag (step 3
8) and returns to step 34.

If the sampling data is above the reference level N in step 37, it is judged whether or not the flag is on (step 39). If the flag is not on, the process returns to step 34. When the flag is on, the flag is turned off (step 40), and then the wave number n is incremented by "1" (step 4).
1).

Next, it is judged whether or not it is the first intersection (step 42). If it is the first crossing point, the count value of the data counter is set to the number m1 (step 43), and the process returns to step 34. On the other hand, when it is not the first crossing point, the count value of the data counter is set to the number m2 (step 44), and the process returns to step 34.

If it is not within the measurement range in step 35, the wave number n is decremented by "1" (step 45), and the operation is ended.

When the sampling data is not lower than the reference level N in step 33, one sampling data is read from the RAM 5c (step 4).
6) It is determined whether or not it is within the measurement range (step 4)
7). If it is within the measurement range, the data counter is incremented by "1" (step 48).

Next, the sampling data is the reference level N.
It is determined whether or not it is below (step 49). If it is not lower than the reference level N, the flag is turned on (step 50) and the process returns to step 46.

If the sampling data is below the reference level N in step 49, it is determined whether or not the flag is on (step 51). If the flag is not on, the process returns to step 46. When the flag is on, the flag is turned off (step 52) and then the wave number n is incremented by "1" (step 5).
3).

Next, it is judged whether or not it is the first intersection (step 54). If it is the first crossing point, the count value of the data counter is set to the number m1 (step 55), and the process returns to step 46. On the other hand, when it is not the first crossing point, the count value of the data counter is set to the number m2 (step 56), and the process returns to step 46.

If it is not within the measurement range in step 47, the wave number n is decremented by "1" (step 45), and the operation is terminated.

Next, using the flowchart of FIG.
The operation of improving the accuracy of the number (time) in steps 23 and 27 of FIG. 6 will be described.

First, the reference level from the level L of the data p (n-1) (data corresponding to the number m1-1 in step 23 and the number m2-1 in step 27) immediately before the time point t when the reference level N is crossed. N is subtracted to obtain the difference L-N of the previous point (step 61).

Next, it is judged whether or not the difference L-N of the preceding points is 0 or more (step 62). When it is 0 or more,
Go to step 64. On the other hand, when it is not 0 or more, that is, when it is a negative value, it is converted to a positive value (step 6).
3), go to step 64.

In step 64, the data pn immediately after the time point t at which the reference level N is crossed (the number m in step 23).
1. In step 27, the reference level N is subtracted from the level M of the data corresponding to the number m2) to obtain the difference MN of the rear points.

Next, it is judged whether or not the difference MN between the rear points is 0 or more (step 65). When it is 0 or more,
Go to step 67. On the other hand, when it is not 0 or more, that is, when it is a negative value, it is converted to a positive value (step 6).
6) and then to step 67.

In step 67, the difference between the front points is divided by the sum of the difference between the front points and the difference between the rear points, and the result is used as a correction value.
Then, this correction value is added to the number of previous points (m1-1 in step 23, m2-1 in step 27) to obtain the number m after correction (step 68), and the operation of improving accuracy is completed.

As described above, in this example, the input analog signal is converted into a digital signal, the digital signal is processed, and the frequency f is measured.
The frequency measurement operation can be easily realized by software.

Although the digital signal is discrete data, the numbers m1 and m2 for calculating the frequency f are calculated by the operation of high precision (see the flowchart of FIG. 8 and the formula 7).
Is corrected to obtain with high accuracy, the frequency f
Can be measured with high accuracy.

Further, since the frequency f can be measured with high accuracy, the processing time can be shortened and the frequency can be measured with high accuracy in a short section. Therefore, it is also possible to measure the chroma frequency directly from the video signal. In the past, due to the error, it was not possible to measure in a short interval, and the frequency of the chroma had to be picked up directly from the circuit transmitting the chroma.

10 and 11 show examples of display on the main screen 10 of the display 5b (see FIG. 1) and other display portions during frequency measurement. Figure 10
Is an example of measurement of a single carrier frequency. In this case, the execution speed and accuracy change depending on the length (measurement range) of the cursor 11. On the other hand, FIG. 11 is an example of measuring the frequency of the chroma of the video signal. In this case, the frequency of the chroma is measured by setting the cursor 11 within the chroma of the video signal.

Further, in this example, the display 5b
Cursor 11 while observing the measurement waveform on the main screen 10 of
Can be used to set the measurement range and the reference level, and there is the advantage that they can be set appropriately and easily so that the frequency measurement of the desired signal is made.

[0071]

According to the present invention, since the discrete data (digital signal) is processed to measure the frequency, the frequency measuring operation can be easily realized by software.

Further, by obtaining the points of time of intersection by linear approximation from the points of time corresponding to the discrete data located immediately before and after the intersection with the reference level, the whole time can be detected with high accuracy in spite of handling the discrete data. You can
The frequency measurement can be performed with high accuracy. Further, by this, the processing time can be shortened, the frequency measurement in the short section can be performed with high accuracy, and the chroma frequency can be directly measured from the video signal.

Further, by setting the measurement range and the reference level on the screen using the cursor, the measurement range and the reference level can be set by moving the cursor while observing the measurement waveform on the screen, There is an advantage that the frequency measurement of the target signal can be set appropriately and easily.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an embodiment.

FIG. 2 is a diagram for explaining how to obtain an intersecting point.

FIG. 3 is a diagram for explaining how to obtain a wave number that intersects a reference level.

FIG. 4 is a diagram for explaining a number error.

FIG. 5 is a diagram for explaining how to improve the accuracy of the number (time).

FIG. 6 is a flowchart for explaining a frequency measurement operation.

FIG. 7 is a flowchart for explaining an operation of obtaining a wave number and a number.

FIG. 8 is a flowchart for explaining an operation of improving the accuracy of the number (time).

FIG. 9 is a diagram showing a main screen and other display units.

FIG. 10 is a diagram showing a display example of a main screen and other display units.

FIG. 11 is a diagram showing a display example of a main screen and other display units.

FIG. 12 is a diagram for explaining a conventional frequency measurement method.

FIG. 13 is a diagram showing a relationship between discrete data (digital signal) and a reference level.

[Explanation of symbols]

 1 CPU 2 Dual RAM 3 DSP (Digital Signal Processor) 4 I / O Port 5 Hardware Section 5a Key Panel 5b Display 5c RAM 5d A / D Converter 5e Input Terminal 10 Main Screen 11 Cursor

Claims (4)

[Claims] 1. A wave number detecting means for detecting a wave number that crosses a reference level within a measurement range from discrete data, and a time point at which the reference level crosses first and last within the measurement range from the discrete data. A frequency measuring device comprising: time detecting means for making the difference the total time; and calculating means for dividing the wave number detected by the wave number detecting means by the total time detected by the time detecting means to obtain a frequency. 2. The frequency measurement according to claim 1, wherein the time detecting means sets a time point at which the discrete data goes from the bottom of the reference level to the top or from the top of the reference level to a crossing point. apparatus. 3. The time detecting means obtains the crossing time point by linear approximation from the time points corresponding to the discrete data located immediately before and after the crossing of the reference level. Frequency measuring device. 4. The frequency measuring device according to claim 1, wherein the measurement range and the reference level are set by using a cursor on a screen.