JPS6071967A - Digital spectrum analyzer - Google Patents

Abstract

PURPOSE:To shorten measurement time as the whole and to make high-precision measurement possible by setting a test signal for each measuring frequency range divided in plural areas and adjusting the input level of an AD converter. CONSTITUTION:First, a noise signal is given to an object 12 to be measured to measure all measuring bands F0-Fn, and a general transfer function characteristic is obtained. This characteristic is divided to parts, where the level change for frequency is sharp, and parts where it is slow, and the noise signal is set as the test signal to an FFT analyzer 14 in case of the area from the band F0 to the band F1 and the area from the band F2 to the band Fn, and a single sine wave signal is set as the test signal to the analyzer 14 in case of the area from the band F1 to the band F2. Next, the set test signal is supplied from a test signal generator 11 to the object 12 to be measured, and the input level of the AD converter in the FFT analyzer 14 is so set that an optimum conversion is attained in the AD converter, and FFT analysis is performed. The transfer function of the object 12 to be measured is measured in accordance with the input test signal of the object 12 to be measured and the Fourier conversion output of the output test signal.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention is a digital camera that supplies a test signal to a measurement target, performs fast Fourier transform on the input test signal and output test signal of the measurement target, and measures the transfer function of the measurement target from the converted output. Regarding spectrum analyzers.

<Background of the Invention> As shown in FIG.
The output test signal from the measurement object 12 is inputted to the fast Fourier transform analysis device (hereinafter referred to as FFT analysis device) 14 from the terminal 13, and the test signal on the input side of the measurement object 12 is inputted to the terminal 15, 9 FFT. The input test signals are input to the analysis device 14, and the input test signals are converted into digital signals in the analysis device 14, and then subjected to fast Fourier transformation to determine the transfer function of the measurement object 12.

In the FFT analyzer 14, as shown in FIG.
The output of 7 is sampled at a constant cycle by an AD converter 18 and converted into a digital signal, and the digital signal is stored in a buffer memory 19. When a certain number of samples, for example 1,024 samples, are stored in the buffer memory 19, fast Fourier transform is performed in the F'FT transformer 21.

At that time, in order to effectively perform the conversion in the AD converter 18, in other words, the sampled data is
When converting into a bit digital signal, the input range is set so as to use the entire 10-pit range as much as possible to increase analysis sensitivity. That is, buffer memory 1
If the sample value is too small, the gain of the amplifier 16 is increased, and if the sample value is too large, the gain of the amplifier 16 is decreased to optimize the conversion by the AD converter 18.

In the FFT analyzer 21, the sensitivity of the input signal is set as shown in step Sl in FIG. That is, as described above, an operation is performed to set the input level of the AD converter 18 so that the conversion is performed optimally. Thereafter, in step S2, fast Fourier transform is performed on, for example, 1,024 sample values taken into the buffer memory 19 in the setting state, and in step S3, the input test signal from the terminal 15 and the output test signal from the terminal 13 are processed. The power spectra Qaa, Qbb and the mutual spectrum Gab are calculated. That is, the sum Gaa of the square of the real part and the square of the imaginary part is calculated for each frequency component subjected to FFT analysis of the test signal input from the terminal 15,
Similarly, for the output test signal input through the terminal 13, the sum Gbb of the square of the real part and the square of the imaginary part of each frequency component subjected to FFT analysis is calculated to obtain each power spectrum, and further, these inputs are A mutual spectrum Gab is calculated by multiplying the corresponding frequency components of the fast Fourier transform of the test signal and the fast Fourier transform of the output test signal with the three-way joint yoke.

Each test signal is repeatedly acquired in the same way, the necessary number of power spectra and mutual spectra are obtained in the same way, and these are averaged in step S4, and the average power spectra <Gaa> and <Gbb>, the average mutual spectrum ( Enter Gab:). A transfer function is determined from these average values in step S5. That is, the average mutual spectrum 2
Divide the average power spectrum <Gab> by the average power spectrum <Gaa>. This is done for each frequency. The results analyzed by the FFT analyzer 21 are displayed on the display 22.

By the way, when the conventional test signal generator 11 generates a single sine wave, performs fast Fourier transform on each sine wave, and then performs fast Fourier transform while changing its frequency, this is repeated in sequence. In some cases, random noise is generated and applied to the measurement target as a test signal.

The former case of using a single sine wave has high measurement accuracy but requires a long measurement time. On the other hand, when random noise is generated, it can be measured in one go, but the measurement accuracy is poor. However, for example, in cases where the level change with respect to the frequency axis is relatively slow in the measurement of a transfer function, the measurement accuracy is not so bad even with measurement using random noise, and a relatively high-precision measurement can be obtained.

Depending on the object to be measured, resonance points or anti-resonance points may occur, and the accuracy of measurement using random noise deteriorates in such a portion where there is a sudden change with respect to the frequency axis. That is, in the transfer function as shown in FIG. 4, there is a resonance point 23a and an anti-resonance point 23b in the middle part on the frequency axis, and the characteristic curve changes rapidly. In such places, measurement using random noise has a particularly poor accuracy, and has the disadvantage that the sharp part of the resonance point becomes dull. However, in this region 23e, which changes rapidly, it changes relatively slowly.
It was found that even measurements using random noise can be performed with relatively high accuracy in this part.

<Summary of the Invention> From this point of view, the purpose of the present invention is to achieve high accuracy as a whole and a relatively short period of time in cases where the change in characteristics includes a rapid part and a part that does not change much with respect to the frequency axis. According to the present invention, the test signal generator is configured to generate multiple types of test signals, and the measurement frequency range is divided into multiple regions. The configuration is such that the test signal to be generated can be set for each divided area, and the test is performed for each frequency area using the set test signal suitable for that area. A to optimize the conversion characteristics of the AD converter for each
Adjust the input level of the D converter.

For this purpose, the test signal generator 11 is configured as shown in FIG. 5, for example. That is, the bus 25 includes a CPU 26 and a ROM 27 that stores its operating program. Readable and writable RA
M 28 is further connected to an input/output section 29 which is connected to the F"FT analyzer 14 shown in FIG. Data indicating the signal mode, such as a signal, multiple sine wave signal, single sine wave signal, 9-frequency swept pressure sinusoidal signal, the maximum and minimum values of the frequency range in which it occurs, as well as the amplitude level and, if necessary, the frequency change width. Data indicating Δf, etc. are sent, and together with these data, a clock that is synchronized with the sampling clock of the AD converter 18 in the FFT analyzer 14 and generally has a higher frequency than this is input.This clock is input from the input/output circuit 29. The signal is input to a frequency divider 31, which is connected to the bus 25, and a necessary frequency division ratio is set by the CPU 26 in accordance with the input data.

On the other hand, the waveform memory 32 stores digital values obtained by sampling each point of a multiple sine wave waveform, and a waveform of a frequency sweep signal, such as a swept sine wave signal whose frequency changes from Fo to Fl. The signal waveforms are stored, and which waveform of the signal to read is determined by the CP via the bus 25.
According to the signal mode given from the FFT analyzer 14, the readout area of the waveform memory 32 is selected and determined. Waveform memory 32 is frequency divider 3
1 output, generally read out with a clock of the same frequency synchronized with the AD converter 18. Therefore, the output of the frequency divider 31 is supplied to the address counter in the waveform memory 32 through the switch 33, and this address counter is incremented. It is also possible to read out the noise signal by switching the switch 33 and supplying random pulses from the noise generator 40 to the waveform memory 32.

The signal read out from the waveform memory 32 is converted into an analog signal, and is supplied to a level adjuster 36 through a multiplier 34 and a switch 35. The level adjuster 36 is set by the CPU 26 via the bus 25 in accordance with the level given by the FFT analyzer 14. A test signal having the set level is output from the terminal 37, and is applied to the measurement object 12 in FIG. In the case of generating a single sine wave signal, a sine wave memory 38
is read out using the output clock of the frequency divider 31. The frequency of the sine wave signal generated from the sine wave memory 38 is set by the CPU 26 via the bus 25. This read sine wave signal is supplied to a level setter 36 through a switch 39 and further through a switch 35. Further, this sine wave signal can be multiplied by a signal read out from the waveform memory 32 in a multiplier 34. This test signal generator uses FFT
Depending on the data provided by the analyzer 14, a test signal of the specified mode can be generated within the specified frequency range.

As mentioned earlier, the input level of the AD converter 18 is
Control is performed so that the conversion operation of the D converter is optimized and the most effective conversion data is obtained. For this purpose, a case where the signal level input to the FFT analysis device 14 is too large is detected. For example, as shown in FIG. 6, within the variable gain amplifier 16, a differential input signal is applied to a common input terminal of an amplifier 43 through resistors 41 and 42, and the output of the amplifier 43 is applied to a comparator 44 as a reference voltage of a reference power supply 45. It is compared with the voltage, and if this common mode noise component is larger than a certain level, the output of the comparator 44 is configured to become high level. Further, the output of the variable gain amplifier 16 is branched and supplied to a comparator 46, and this amplified output is compared with the reference voltage of a reference power supply 47. If the amplified output is at a level higher than the maximum conversion level in the Ω converter 18, the comparison is made. The conversion output of the OAD converter 18, in which the output of the converter 46 is set to a high level, is branched to a digital comparator 48, and is compared with a threshold value in a register 49.
If the output of the D-converter 18 is overflowing data, a higher level than the comparator 48 will be generated.

The conversion output of the AD converter 18 is, for example, 12 bits,
The most significant 12 bits of the parallel bit output line 52 are AD.
Used for conversion output. On the other hand, comparators 44, 46, 4
Each output of 8 is input through an OR circuit 51 to the least significant bit of an output line 52 of the AD converter. Therefore, if the input signal level or noise is too large, the lowest bit in the data input to the rough memory 19 becomes logic 1, and the lowest If the bit is 1, it is determined that the input signal is too large, and the gain of the variable gain amplifier 16 is lowered by a certain value, for example, 10 db.

This range setting operation is performed, for example, as shown in FIG.
,024 pieces are taken in, this FFT analysis analyzer 1
The buffer memory 19 is stored in a predetermined number, for example 1, in step S+.
024 is taken out, and in step S2 it is checked whether or not the gain of the variable gain amplifier 16 has been both increased and decreased up to that point. If such an operation has not been performed, it is read out in step S3. Checks whether each least significant bit of the data is 1 or not (−
It is checked whether there is any overflow, and if there is overflow, the gain of the variable gain amplifier 16 is lowered by a certain amount, for example 10 db, in step S4. On the other hand, in step S3, it is checked whether the input is too small compared to the read data, and if the upper example bit of the data is always O, it is determined that the input level data is too small; In S6, the gain of the variable gain amplifier 16 is set to a certain amount,
For example, it increases by 10db. After repeating this process, that is, taking 1,024 new data each time and checking whether there is an overflow or whether it is too small, or after repeating it several times, or
If all of the input data is neither overflow nor underinput in the first operation, it is determined in step S7 that the range is appropriate, and the variable gain amplifier 1 is set in this range.
A gain of 6 is kept set. In some cases, the gain is too large in a certain input range, so if the gain is lowered, the gain becomes too small for the next captured data, making it impossible to proceed to step S7. In other words, when the gain is increased and decreased at the same time, this is detected in step S2, and the gain of the side determined to be under-input range at that time is changed in step S.
6, it is determined that the input is too low, and the gain is set to the increased state in step S8.

Next, the operation of the apparatus according to the present invention will be explained. As shown in FIG. 8, first, in step S1, the entire band F is measured.

to Fn, the noise signal or multiple sine wave signal is the measurement target 1
2 and measure. As a result, a rough transfer function characteristic as shown in Fig. 4 is obtained, and the level change with respect to frequency is divided into a steep part and a gentle part. In this figure, the area from Fo to Fl and the area from Fr to F2 are divided. For example, as shown in FIG. 9, the test signals to be generated for each region are divided into three regions, F2 to Fn, and Fo to Fl and F2 to Fn are noise signals, and Fl to F2 are
The region of is a single sine wave signal, the amplitude of each generated signal is Vl + F2 * F3, and for a sine wave signal, the range of change in the frequency of the sine wave Δf is, for example, F.0 Such a test The data 00 is set in the FFT analysis device 14. Also, at this time, r is 1
From now on, in step S3, the designated area rl
A test measurement is performed for the frequency range Fo to F+, and after the measurement, r is incremented by 1 in step S4, and in step S5, r is the number of regions R1, that is, in this example, it is divided into three regions, so R = 3. If it is not a dog, the process returns to step S3 to perform measurements on the next specified area, that is, the second specified area Ft to F2 in this example.

When measuring using a noise signal in the designated region r = l, as shown in FIG. 10, the test data required from the FFT analyzer 14, that is, the frequency region is F. to F1, data indicating that the generated test signal is a noise signal and that the amplitude is 1 is supplied to the test signal generator 11.

Based on this, the test signal generator 11 switches the switch 33 to the noise signal source 44 side in FIG. , the signal is supplied to a level setter 36 through a switch 35, and the level setter 36 sets the level to Vl and outputs a test signal. Switch 39 is turned off.

Upon receiving this test signal, the FFT analyzer 14 changes the level of the input side of the AD converter 18 so that the conversion characteristics of the AD converter 18 are the best, as described above with reference to FIGS. 6 and 7. Controls the gain of gain amplifier 16. Thereafter, in step S4, FFT analysis is performed at the set level. This measurement is performed, for example, several times and the average is taken as described above. When the measurement for one designated area is performed in this way, the next designated area, in this example, the area Fl to F2, is measured. Test data indicating that S+ is in the range F+ to F2, the test signal is a single sine wave, the amplitude is F2, and the frequency change is 5 is transferred to the test signal generator 11. In step S2, the lowest frequency F in this designated area is
+ is set to Fo, and in step S3 the test signal generator 11
generate more signals. In this case, the switch 39 is turned on in FIG. 5, and the switch 35 is switched to the switch 39 side. The sine wave memory 38 is set to generate the frequency Fi.

Therefore, at the beginning, a sine wave of frequency F+ is generated and output, and based on this sine wave signal, step S4
In this step, level setting is performed to obtain the optimum sensitivity. After that, in step S5 for the sine wave of frequency F1, F
FT analysis is performed, and in step S6, Δf is added to frequency F1 to obtain Fi, and Fi is the highest frequency F in the specified area.
It is checked in step S7 whether it is larger than 2, and if it is smaller, the process returns to step S3, and from this, a sine wave signal with a frequency of F1+Δf is generated. In this way, Δf
Generate a high frequency sine wave signal and perform FFT analysis by setting the sensitivity settings to be optimal each time. During the analysis, data is taken multiple times at the optimal setting level and the FFT conversion output is averaged. 0 When the frequency Fl of the generated single sine wave becomes larger than the highest frequency F2 of the specified area, the test using the single sine wave in the specified area is finished. 0 In this example, next, the measurement of the area of frequencies F2 to Fh is performed. In that case, the test is performed in the same manner as in FIG. To divide the measurement frequency range and generate test signals for each region, for example, in the example of FIG. 4, a frequency swept sine wave signal is generated for the frequency range Fo to F1 and the frequency range F2 to Fn. Good too. Alternatively, all of these regions may be configured to generate frequency swept sine wave signals, for example, as shown in FIGS. 12 and 13, in the frequency range Fo to Fl, an intermediate wave number of Δ:h = 40 Hz is swept. A signal is generated, and then a sine wave signal that sweeps a frequency within the range of a frequency Fo+Δf1 to Fo+2Δf1, which is an addition of Δft, is generated, and the frequency domain Fl
For F2 to F2, the sweep signal width Δfz is made small, for example, about 4 Hz, and measurement is performed using a slowly sweeping signal, and for the frequency range F2 to Fn, the sweep width Δfs =
Measurement may be performed with a large sweep width such as 100 Hz. Also, the measurement range is measured with a noise signal or multiple sine waves, and then a single sine wave or a slow frequency swept sine wave is used only in a range where the frequency domain changes rapidly, for example, in the range from Fl to F2 in the previous example. However, the measurement data of that part may be used in place of the previous data using multiple sine waves or noise signals.The frequency domain is divided and measured in this way, and the optimal sensitivity is determined each time. Since the analysis results are calculated in the frequency range Fo = Fll
In order to see the whole image, it is necessary to perform level correction. That is, the FFT analysis processing in step S4 in FIG. 10 and step S5 in FIG. 11 is performed as shown in FIG. In other words, FFT conversion is performed in step Sl, and the analysis result is subjected to size correction in step S2.

This correction is performed, for example, in the frequency range FO to F that was first captured.
Regarding the FFT conversion output in the frequency domain when the gain is made larger than the gain of the variable gain amplifier 16 at that time, using the level range setting made for l as a reference, for example, if the gain is increased by 10 db, each measurement If the magnitude of the frequency component is lowered by 10 db and conversely the gain is lowered by 10 db, correction is performed to increase the magnitude of each frequency component of the FFT transform by 10 db. Thereafter, in step S3, the lower spectrum Gaa, Qbb, and mutual spectrum Gab are calculated for each of the corrected frequency components.New data is also taken in in the same manner, that is, the process shown in FIG. 10 or 11. The multiple data acquired in this way are averaged for each frequency component in step S4, that is, the average power spectrum <
Gaa ) (Gbb ), average mutual spectrum〈
Gab> is calculated. Furthermore, a transfer function is calculated from this Tsubo-yen spectrum in step S5.

<Effects> As described above, according to the present invention, first, rough measurements are performed over the entire measurement frequency band, and then, by looking at the results, in the vicinity where the measurement results are changing rapidly, single-sine wave signals or High-precision measurements can be made by measuring with a slow frequency sweep signal, and the characteristics change relatively slowly. By using a high-speed sweep signal, high-precision measurements can be made in a short amount of time. Therefore, overall measurement time is short and highly accurate measurement can be performed.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing the measurement system of the digital spectrum analyzer, Fig. 2 is a block diagram specifically showing a part of the FFT analysis device, Fig. 3 is a flow chart showing an example of FFT analysis operation, and Fig. 4 is a diagram showing an example of the characteristics of the measured transfer function,
Figure 5 is a block diagram showing an example of a test signal generator, Figure 6 is a block diagram showing an example of a test signal generator.
Figure 7 is a block diagram showing an example of overflow detection.
The figure is a flowchart showing an example of operation for setting the optimum level range.
FIG. 8 is a flowchart showing an example of the operation of the spectrum analyzer according to the present invention, FIG. 9 is a diagram showing an example of setting the frequency range of the test signal 9, and FIGS. 10 and 11 each show an example of the operation for each designated region. Flow chart, FIG. 12 shows the assignment data of other test signals, FIG. 13 shows the assignment of frequency characteristics corresponding to FIG. 12, and FIG. 14 shows step S41 of FIG. 10. FF of step S5
It is a flowchart which shows the example of operation of T analysis. 11: Test signal generator, 12: Measurement target, 14: FFT
Analyzer, 16: Variable gain amplifier, 17: Low frequency waver,
18: AD converter, 19: buffer memory, 21:
FFT analyzer, 22: Display. Patent applicant: Takeda Riken Kogyo Co., Ltd. Agent: Takashi Kusano 7178 Figure? 9 Figure 110 Figure 11 Figure 7i712Figure 71713ToO meeting 8T1 14th Bile procedural amendment (spontaneous) October 17, 1980 2, Title of invention Digital spectrum analyzer 3, Amendment person case Relationship between patent applicant Takeda Riken Kogyo Co., Ltd. 4, agent Sumo Building 5, 4-2-21 Shinjuku, Shinjuku-ku, Tokyo, date of notice of reasons for refusal Initiator 6, subject of amendment Column for detailed description of the invention and drawings 7. Contents of amendment (1) Page 4 of the specification, lines 12-14 “Furthermore...
...and the product of the fast Fourier transform of the output test signal of the corresponding frequency component.'' (9) Page 5, line 6 of the same book “From test signal generator 11”
Correct it to "by the test signal generator 11." Circular page 7 key ← Correct line a "synchronized with the sampling clock" to "synchronized with the sampling clock". (4) In the same book, page 12, line 6, change ``Is it the upper bit of the data?'' after ``The upper bit of the absolute value of the data (5).'' 6) Correct it to "noise signal source 40". (7) Add the numeral "25" to Figure 5 as shown in red in the attached copy diagram. that's all

Claims (1)

[Claims] (1) A digital spectrum analyzer that supplies a test signal from a test signal generator to a measurement target, converts the signal from the measurement target into a digital signal with an AD converter, and performs fast Fourier transform on the digital signal for analysis. In the test signal generator, the test signal generator is configured to selectively generate multiple types of test signals, and divides the measurement frequency range into multiple areas and sets the type of test signal to be generated in each area. a test signal setting means; a means for generating a set test signal set by the setting means and corresponding to each frequency domain from the test signal generator; A digital spectrum analyzer comprising level setting means for setting an input level range of an AD converter so as to obtain optimum conversion.