JPH05291829A - Optional waveform generator - Google Patents

Abstract

PURPOSE:To provide the optional waveform generator generating a desired object waveform by correcting a deviation between its own output waveform and the desired object waveform. CONSTITUTION:The optional waveform generator having a waveform memory 4 storing waveform data, an address generator 8 addressing the waveform memory and a D/A converter 5 converting the waveform data outputted from the waveform memory into an analog output waveform is provided with a reference waveform memory storing reference waveform data corresponding to the analog output waveform, a measurement means A measuring the analog output waveform and generating the measured waveform data, a distortion detection means B comparing the measured waveform data from the measurement means and the reference waveform data to detect amplitude distortion and differentiation coefficient distortion and a correction means correcting the waveform data stored in the waveform memory in response to the output of the distortion detection means to obtain a desired analog output waveform.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an arbitrary waveform generator that outputs a desired waveform.

[0002]

2. Description of the Related Art Conventionally, an arbitrary waveform generator using a waveform function value generator is known. FIG. 2 is a block diagram of an arbitrary waveform generator using a conventional waveform function value generator. In the figure, the waveform function value generator 1 calculates the peak value of a desired waveform at each time, programs the calculated value as a time-series signal value, and writes it in the program memory 3. The program memory 3 is composed of a RAM or the like and is controlled by the CPU 2. The waveform time-series signal value data written in the program memory 3 is rewritten in the waveform memory 4 configured by a high-speed memory or the like for speeding up the processing. Rewriting of data from the program memory 3 to the waveform memory 4 is also controlled by the CPU 2. The data written in the waveform memory 4 is read according to the address value of the address generator 8 driven by the clock signal from the clock distributor 9 and input to the D / A converter 5. The digital signal value from the waveform memory 4 is D /
After being converted into an analog sample value signal by the A converter 5, it is added to a filter 6 such as a low pass filter or a band pass filter to form a smooth analog output signal.

The analog output signal is power-amplified by the amplifier 7 and then output from the output terminal OUT.

[0004]

However, the conventional arbitrary waveform generator described above has the following problems. (1) With respect to the calculated value of the waveform obtained from the waveform function value generator, nonlinearity and high-frequency characteristic distortion of the D / A converter, amplitude characteristic and phase of a filter such as a low-pass filter or a band-pass filter The analog output waveform is distorted due to the characteristics, various distortions of the amplifier, and the like, which causes a deviation from the target output waveform. (2) Further, the correction means or the D / A converter, the filter such as the low-pass filter or the band-pass filter, or the characteristic of the amplifier is changed to correct the deviation to correct the deviation, and the target waveform is the analog output waveform. Is difficult to approach.

The present invention eliminates the above-mentioned problems and corrects the deviation between the output waveform of the arbitrary waveform generator and the desired desired analog waveform to form the desired analog waveform. An object is to provide a waveform generator.

[0006]

In order to achieve the above object, the present invention provides a waveform memory for storing waveform data,
In an arbitrary waveform generator having an address generator for addressing the waveform memory and a D / A converter for converting the waveform data output from the waveform memory into an analog output waveform, reference waveform data corresponding to the analog output waveform A reference waveform memory for storing, a measuring means for measuring the analog output waveform to generate measured waveform data, an amplitude distortion and a differential coefficient by comparing the measured waveform data from the measuring means with the reference waveform data. A distortion detecting means for detecting distortion and a correcting means for correcting the waveform data stored in the waveform memory according to the output of the distortion detecting means are provided.

[0007]

According to the present invention, the analog output waveform is measured to generate the measured waveform data, the measured waveform data is compared with the reference waveform data, and by the comparison, the amplitude distortion and the differential coefficient distortion are caused by the distortion detecting means. Is detected and the waveform data stored in the waveform memory is corrected in accordance with the output of the distortion detecting means to obtain the waveform data stored in the waveform memory so that the analog output waveform of the target reference waveform is obtained. It can be modified in data units.

Further, by using fuzzy inference as the correction means for correcting the waveform data, it is possible to correct the waveform data for various waveforms with a small memory capacity.

[0009]

Embodiments of the present invention will be described in detail below with reference to the drawings. FIG. 1 is a block diagram of a first embodiment of an arbitrary waveform generator of the present invention. In the figure, a waveform function value generator 1, a CPU 2, a program memory 3, a waveform memory 4, a D / A converter 5, a filter 6 such as a low pass filter or a band pass filter, an amplifier 7, an address generator 8, and a clock distributor 9 are shown. 2 is similar to the conventional arbitrary waveform generator shown in FIG. In addition,
The waveform memory 4 of the present invention comprises a basic waveform memory and a modified waveform memory, and the reference waveform data from the program memory 3 is input to the basic waveform memory and the modified waveform memory.

The arbitrary waveform generator of the present invention feeds back the output obtained from the amplifier 7 to sequentially change the data of the time-series signal value of the waveform stored in the modified waveform memory of the waveform memory 4 to obtain the analog output waveform. For measuring the analog output waveform of the output terminal OUT to generate measured waveform data, and the measured waveform data and the reference waveform from the measuring means A. Distortion detecting means B for comparing the data to detect amplitude distortion and differential coefficient distortion, and correction means C for correcting the waveform data stored in the corrected waveform memory of the waveform memory according to the output of the distortion detecting means B. And have.

The measuring means A, the strain detecting means B and the correcting means C will be described below in order. First, the measuring means A will be described. The measuring means A is composed of a sampler 10 and an A / D converter 11, and converts the analog output from the amplifier 7 into a digital signal. Sampler 10 at this time
The A / D converter 11 is timed by the clock signal from the clock distributor 9.

The measured waveform data obtained by the measurement by the measuring means A is input to the distortion detecting means B. Next, the strain detecting means B will be described. Since the digital signal from the A / D converter 11 is obtained by digitizing the output from the amplifier 7, there is a conventional example between the value of the digital signal and the reference waveform data of the time-series signal value of the target waveform. There is a gap as explained in. The distortion detecting means B is composed of a scaler 13, a comparator 14, a distortion output section 15 and a reference waveform data generator 19, and the above-mentioned waveform deviation is caused by the reference waveform data from the reference waveform data generator 19 in the comparator 14. And the amplitude distortion and the differential coefficient distortion are output from the distortion output unit 15. The digital signal from the A / D converter 11 is scaled by the scaler 13 before being input to the comparator 14 as a pre-process for comparing the data value with the reference waveform data. On the other hand, the reference waveform data is output from the reference waveform data generator 19 based on the data stored in the reference waveform memory of the waveform memory 4. Further, the reference waveform memory for storing the reference waveform data may be the program memory 3 or another memory instead of the waveform memory 4. The reference waveform data generator 19 is driven by the clock signal from the clock generator 9 and the control signal from the CPU 2.

The comparison in the comparator 14 compares, for example, the deviation of the amplitude of the waveform and the slope of the waveform. FIG. 3 is a signal comparison diagram. In the figure, the straight line is the reference waveform data and the broken line is the measurement data. In comparing waveforms, the deviation and slope of the amplitude between the reference waveform data and the measured data at a certain reference value are obtained, and the value of the amplitude deviation or the ratio of the amplitude deviation and the reference waveform data value is used as the amplitude distortion, and The differential coefficient or the ratio of the differential coefficient and the differential coefficient of the reference waveform data is defined as differential coefficient distortion. The comparison value compared in the comparator 14 is output in the distortion output section 15 as amplitude distortion and differential coefficient distortion. The amplitude distortion and the differential coefficient distortion from the distortion detecting unit B are input to the correcting unit C to correct the waveform data stored in the waveform memory 14.

Next, the correction means C will be described. The correction means C includes a decision table 16, a correction data output unit 17, and a high speed addition processor 18. The decision table 16 is a table in which it is preset how to correct the waveform data stored in the waveform memory 14 based on the data of the amplitude distortion and the differential coefficient distortion, and depending on the desired waveform shape or the like. It is set.
The modified data output unit 17 forms modified waveform data according to the output of the decision table 16 and inputs the modified waveform data to the waveform memory 4 via the high speed addition processor 18. This is because the high-speed addition processor 18 performs speed matching with the waveform memory 4. The input modified waveform data modifies the waveform data in the modified waveform memory of the waveform memory 4. The corrected waveform data is read again according to the address value of the address generator 8, converted into an analog sample value signal by the D / A converter 5, and then added to the filter 6 such as a low pass filter or a band pass filter. , An analog output signal based on the modified data is obtained. The waveform data of this analog output signal is corrected again by the measuring means A, the distortion detecting means B, and the correcting means C.

By repeating this, the analog output approaches a desired waveform. In order to stop the repeating cycle, for example, the decision table 1
In 6, the correction data can be set not to be output when the value from the distortion output unit 15 becomes equal to or less than the set value. Next, a second embodiment of the present invention will be described. The second embodiment of the present invention is a drive with an equivalent time.

FIG. 4 is a signal diagram for explaining driving in real time and driving in equivalent time. FIG. 4A shows a signal in the case of driving in real time.
Are sampled at equal intervals t. In addition, FIG.
Is a signal in the case of driving by an equivalent time, and 5 cycles 5
Sampled at equal intervals t + T during T. In the case of a periodic signal, it is 1 at a long period interval depending on the equivalent time.
Since the data for the period T can be obtained, the subsequent signal processing speed can be reduced.

The waveform data memory 12 is input to the scaler 13 in the same manner as in the first embodiment to obtain corrected data. The correction of the data in the waveform memory 4 in this case is intermittent compared to the real-time driving according to the cycle in the equivalent time. FIG. 5 is a block diagram of a second embodiment of the arbitrary waveform generator of the present invention. The first embodiment drives in real time, whereas this embodiment drives in equivalent time for a periodic waveform. By driving with the equivalent time, the signal processing for correcting the waveform data is facilitated and it is possible to deal with high frequency signals. This second embodiment is shown in FIG.
The waveform data memory 12 is installed between the / D converter 11 and the scaler 13, and the clock signal for driving the sampler 10 is gradually reduced.

In the first and second embodiments, the correction means C
The decision table 16 is used for the formation of the correction data of 1 to set the correction conditions corresponding to the desired waveform. Therefore, the decision table 16 is prepared corresponding to the desired waveform type and the waveform value of the waveform. FIG. 6 is a relationship diagram between a desired waveform and a decision table. In the figure, for example, for a sine wave, a decision table 1
6-1 is prepared, and the decision table 1 for a rectangular wave
6-2 is prepared, and the decision table 1 for the triangular wave
Prepare 6-3. Then, a set value for the amount of distortion is set in this decision table. Therefore, the amount of data set in the decision table increases according to the fineness of the setting.

In the third embodiment of the present invention, the decision table is executed by one fuzzy inference means so that the number of decision tables corresponding to the kind of waveform and the waveform value of the waveform in the above embodiment can be determined. The increase and the increase in the amount of setting data stored in the decision table are suppressed. FIG. 7 is a block diagram of a third embodiment of the arbitrary waveform generator of the present invention. In the third embodiment,
The decision table 16 in the correction means C in the first and second embodiments is replaced with fuzzy inference means D including a control rule section 20 and a defuzzifier 21.
The signal from the distortion output section 15 is input to the fuzzy inference means D, and the output of the fuzzy inference means D is the corrected data output section 1
7 is output. Fuzzy inference generally defines input / output relationships with ambiguity by control rules and converts the output into concrete values by defuzzification. Since this control rule has generality, it can be performed by switching the control rule of the control rule unit 20 according to the waveform without preparing a decision table corresponding to a desired waveform value. Since the setting data is not stored unlike the table, the required storage capacity can be reduced. This relationship is based on the decision tables 16-1 and 16- prepared according to the types of waveforms in FIG.
2 can be seen by replacing the fuzzy inference means D with the control rule unit 20 and the defuzzifier 21.

FIG. 8 shows an example of the control rule. The control rule table in the figure has a condition of amplitude distortion in the vertical direction and a condition of differential coefficient distortion in the horizontal direction at the upper end, and concludes the values in the table. In this table, NB is N
negative Big, NS is negative S
mall, ZO is Zero, PS is PositiveS
mall and PB represent Positive Big. Therefore, for example, the condition of the amplitude distortion of the table is N
If B is large in the negative direction and the condition of differential coefficient distortion is also large in the negative direction in NB, the conclusion is PB, and the conclusion that the correction amount is large in the positive direction is set.

The defuzzifier 21 forms a concrete correction value based on the conclusion of the control rule unit 20 and can be performed by various methods generally known as a membership function. The correction value from the defuzzifier 21 corrects the correction waveform memory of the waveform memory 4 via the correction data output unit 17 and the high speed addition processor 18. The subsequent operation is the same as in the first and second embodiments.
Further, the same means as described above may be used to stop the repetition of the correction, and a stopping rule may be included in the control rule or the defuzzifier 21.

FIG. 9 is a block diagram of a fourth embodiment of the arbitrary waveform generator of the present invention, which is an embodiment in which the control rule of the control rule unit 20 is changed according to the target waveform. In the figure, a waveform function generator 1 and a control rule unit 2
The control rule memory 22 is connected to 0. A control rule table as shown in FIG. 8 is stored in the control rule memory 22 according to the type of waveform, and the control rule corresponding to the type of waveform of the waveform function generator 1 is output to the control rule unit 20. The number of control rules stored in the control rule memory 22 corresponds to the type of waveform, and some waveforms can be commonly used. Therefore, compared with the number required for the above decision table. Can be reduced. Also, the amount of data required for the control rule itself can be small as shown in the control rule table of FIG. 8, and it is not necessary to have a large capacity for a large amount of data as in the decision table.

Next, the fuzzy inference means D of the third embodiment
The formation of the control rule of the control rule unit 20 when using is described. FIG. 10 is a block diagram of a fifth embodiment of the arbitrary waveform generator of the present invention. In the figure, the output end O
A switch SW is provided in the UT, one of which is a sampler 10 and the other is an oscilloscope 30 and a spectrum analyzer 31.
The calculation result of the distortion value calculation 32 is input to the distortion output unit 15 via the distortion value calculation 32. When the switch SW is connected to the sampler 10, it is the same as in the second embodiment and is processed online. On the other hand, when the switch SW is connected to the oscilloscope 30 and the spectrum analyzer 31 side, the processing is performed offline and the simulation is performed.

In the off-line simulation, the output signal from the output terminal OUT is input to the oscilloscope 30 for waveform observation, and the output signal from the output terminal OUT is input to the spectrum analyzer 31 to perform frequency analysis. Then, the amount of distortion such as amplitude distortion and differential coefficient distortion with the reference waveform is calculated by comparing with the reference waveform for the purpose of this waveform observation and frequency analysis, and the calculation result is input to the distortion output unit 15 and the calculation is performed. The control rule of the control rule unit 20 is modified while considering the result. The distortion output unit 15 inputs the data to the control rule unit 20. In the control rule unit 20, the data from the distortion output unit 15 is modified based on the modified control rule, and the data in the modified waveform memory of the waveform memory 4 is modified.

This off-line simulation is performed manually, not by the processing by the signal circuit. With this offline simulation,
The control rule of the control rule unit 20 is corrected and the data of the corrected waveform memory of the waveform memory 4 is corrected.
The output signal based on the correction control rule and the correction data is simulated off-line again,
Further, the control rule and the correction waveform data of the waveform memory 4 are corrected. By repeating this modification, the control rule in fuzzy reasoning is set.

When the control rule is set by this manual simulation, the SW is connected to the sampler 10 side to perform online processing. It should be noted that the present invention is not limited to the above-described embodiments, and various modifications can be made based on the spirit of the present invention, and they are not excluded from the scope of the present invention.

[0027]

As described above, according to the present invention,
The analog output waveform is measured to generate measured waveform data, the measured waveform data is compared with the reference waveform data, and the amplitude distortion and differential coefficient distortion are detected by the distortion detection means by the comparison, and the distortion detection means Since the waveform data stored in the waveform memory is corrected according to the output, (1) the waveform data stored in the waveform memory is corrected in data units so that the analog output waveform of the target reference waveform can be obtained. be able to. (2) By using fuzzy inference as the correction means for correcting the waveform data, it is possible to correct the waveform data for various waveforms with a small memory capacity. With such an effect, the deviation between the output waveform of the arbitrary waveform generator and the desired desired waveform can be corrected to generate the desired desired waveform.

[Brief description of drawings]

FIG. 1 is a block diagram of a first embodiment of an arbitrary waveform generator of the present invention.

FIG. 2 is a block diagram of a conventional arbitrary waveform generator.

FIG. 3 is a signal comparison diagram.

FIG. 4 is a signal diagram for explaining driving in real time and driving in equivalent time.

FIG. 5 is a block diagram of a second embodiment of the arbitrary waveform generator of the present invention.

FIG. 6 is a relationship diagram between a desired waveform and a decision table.

FIG. 7 is a block diagram of a third embodiment of the arbitrary waveform generator of the present invention.

FIG. 8 is a control rule table of the present invention.

FIG. 9 is a block diagram of a fourth embodiment of the arbitrary waveform generator of the present invention.

FIG. 10 is a block diagram of a fifth embodiment of the arbitrary waveform generator of the present invention.

[Explanation of symbols]

 1 Waveform function value generator 2 CPU 3 Program memory 4 Waveform memory 5 D / A converter 6 Filter 7 Amplifier 8 Address generator 9 Clock distributor 10 Sampler 11 A / D converter 12 Waveform data memory 13 Scaler 14 Comparator 15 Distortion output Section 16 decision table 17 corrected data output section 18 high-speed addition processor 19 reference waveform data generator 20 control rule section 21 defuzzifier 22 control rule memory 30 oscilloscope 31 spectrum analyzer 32 distortion value calculation SW switch A measuring means B distortion detecting means C correction means D fuzzy inference means

Claims (2)

[Claims] 1. A waveform memory for storing waveform data; (b) an address generator for addressing the waveform memory; and (c) converting the waveform data output from the waveform memory into an analog output waveform. An arbitrary waveform generator having a D / A converter, (d) a reference waveform memory for storing reference waveform data corresponding to the analog output waveform, and (e) measurement of the analog output waveform to generate measured waveform data. And (f) distortion detecting means for comparing the measured waveform data from the measuring means with the reference waveform data to detect amplitude distortion and differential coefficient distortion, and (g) output of the distortion detecting means. An arbitrary waveform generator, the correction means correcting the waveform data stored in the waveform memory according to the above. 2. The correction means includes fuzzy inference means for correcting the waveform data in the waveform memory according to the output of the distortion detection means based on a plurality of predetermined control rules. The arbitrary waveform generator according to claim 1.