JP2573262B2 - Quasi-peak value evaluation device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a quasi-peak value evaluation device for obtaining a quasi-peak value of an input signal, and particularly to pulse-like noise having a low repetition frequency in interference wave measurement. The present invention relates to a quasi-peak value evaluation device that improves the quasi-peak value measurement.

[Conventional technology]

The quasi-peak detector attempts to detect the interference wave by correcting the response of the human ear, that is, the audibility, in order to evaluate the degree of interference with the broadcast wave due to interference waves such as noise. . Therefore, conventionally, in order to realize this detection, a detection circuit having the same detection characteristics as a human hearing characteristic has been used, or a circuit which performs envelope detection and thereafter exhibits the hearing characteristics has been used. The basic characteristics of the quasi-peak detector having such audible characteristics are described in the International Commissioner for Wireless Interference (CI)
It is recommended by CISPR Publ.16 issued by SPR) (hereinafter referred to as CISPR standard).

Heretofore, such a quasi-peak detector has been configured by an analog circuit, and has been used by being mounted on an interference wave measuring device that exclusively measures an interference wave. In recent years, as regulations on noise emitted from electronic devices have become more stringent, this quasi-peak detector has come to be attached to a spectrum analyzer which is a general-purpose measuring instrument. Such a quasi-peak detector is described in “Nikkei Electronics” No. 417 published by Nikkei McGraw-Hill on March 23, 1987 (a part of the CISPR standard is described). Here, the configuration of the prior art shown in FIG. 4 will be described.

FIG. 4 is a block diagram after being converted to an intermediate frequency by heterodyne or the like in a spectrum analyzer, for example.

In FIG. 4, reference numeral 2c denotes an intermediate frequency signal, and 2d denotes a band of a required intermediate frequency signal 2c.
A quasi-peak detector 30 for detecting the quasi-peak value 10 having a detecting element and a predetermined charging time constant ( _Tc ) and discharging constant ( _Td ). 31 is a display means for displaying a quasi-peak value.

In these circuits, for example, 30
When measuring noise in the frequency range of up to 1000 MHz, the intermediate frequency signal 2c is band-limited by an intermediate frequency filter 2d having BW = 120 KHz, and the charging time constant T _c = 1 ms and the power transmission time constant T _d = 550
The detection is performed by the quasi-peak detector 30 having ms, and the result is displayed on the display unit 31. Specific examples of the quasi-peak detector 30 considered to satisfy such basic characteristics include:
There was one disclosed in JP-A-54-73559. The circuit diagram is shown in FIG.

In FIG. 5, the intermediate frequency signal is output by the diode 32 and the smoothing circuit 33.
Envelope detection of 2f to extract the amplitude fluctuation of noise, then
The resistors 36 and 39 and the capacitor 37 are driven by the differential amplifier 34 and the diode 35 having high impedance with respect to the impedance of the diode 32, and quasi-peak detection is performed. In other words, when the diode 35 is turned on by the positive voltage of the signal applied to its anode, the output impedance of the cathode can be regarded as a constant voltage source by feedback, so that the resistor 36
And are charged at the charging time constant T _c by capacitor 37, the diode 35 is the impedance becomes very high as seen on the cathode side becomes OFF by a negative voltage of such signals to the anode, the voltage charged in the capacitor 37 of its own capacity And a discharge time constant _Td determined by the resistor 39.

[Problems to be solved by the invention]

Such a quasi-peak detector 30, for example, when measuring pulse-like noise with a low repetition frequency and a long time interval, has a high peak value of the noise, but is filled with the quasi-peak detector 30. Since the level of the discharged and output level occupies less time as the repetition time interval is longer, there is a drawback that a large dynamic range from a high level to a low level is required. For this reason, the quasi-peak detector 30 and the circuit preceding it are easily saturated, and the measurement range cannot be widened. Also, because the measurement range is narrow,
An operation of finely adjusting the input noise level to the measurement range with an input attenuator or the like was required, and the measurement procedure was complicated. That is, when these are concretely described, for example, in the case of a measurement frequency range of 30 to 1000 MHz, the quasi-peak detector 30 becomes low when the repetition frequency of the input signal becomes low as shown in FIG.
The detection efficiency is greatly reduced, and conversely, an extra dynamic range (hereinafter referred to as an overload coefficient) is required by the reduced amount. When detecting independent pulses, a maximum overload coefficient of 44 dB is required. FIG. 7 shows the relationship among the dynamic range, the overload coefficient, and the measurement range in an easy-to-understand manner using the measurement of an isolated pulse as an example.

As can be seen from the figure, when measuring a pulse having a low repetition frequency, an overload coefficient of 44 dB and a measurement range of 10 dB require a minimum (when S / N = 12) dynamic range of 66 dB. Therefore, even in the conventional example shown in FIG. 5, the dynamic range is almost 56 dB, and there is almost no measurement range for an isolated pulse or the like, and the level of the pulse coincides with the very narrow measurement range (a value close to 0 dB). It could not be measured unless they matched.

The internal resistance (resistance at ON and OFF) of the detection diode 35 of the quasi-peak detector 30 shown in FIG. 5 changes according to the level of the input signal. Although various efforts have been made in the past, such as applying feedback, it is difficult to maintain this internal resistance at a constant value over the range of (56 dB + measurement range). This change in the internal resistance is a disadvantage that lowers the detection accuracy.

Further, when such a conventional quasi-peak detector 30 is applied to a spectrum analyzer, the spectrum
The level diagram of the circuit before the intermediate frequency of the analyzer, for example, the mixer, the input amplifier in front of it, also had to be changed so as to cover the dynamic range including the overload coefficient as described above (see “Nikkei Electronics, see Figure 17 on page 367).

In the case where a random signal is subjected to quasi-peak detection, an overload coefficient of the detection circuit may be smaller than that of an isolated pulse signal.

The present invention has been made in order to solve the above-mentioned conventional problems. Since it was a pulse-like signal having a low repetition frequency that required a maximum overload coefficient in the past, the present invention It is an object of the present invention to provide a quasi-peak value evaluation device in which the overload coefficient at the time of the quasi-peak value measurement is greatly reduced, the measurement range is widened, and a pulse-like signal having a low repetition frequency can be easily measured. .

In addition, another invention of the present invention can evaluate the quasi-peak value of a random signal or a high-frequency signal in addition to the above-mentioned object, and can load dynamic range and the like on a measuring instrument having a different purpose such as a spectrum analyzer. It is an object of the present invention to provide a quasi-peak value evaluation device which can be additionally used without applying the above.

[Means for solving the problem]

A quasi-peak evaluation apparatus according to the present invention includes a selection unit that extracts and outputs an evaluation signal to be evaluated from a pulse-like input signal, and a peak that detects a maximum level of the evaluation signal and outputs the value. Detecting means, frequency detecting means for detecting the repetition frequency of the input signal and outputting as frequency information,
Coefficient output means for outputting a response coefficient corresponding to the detection response of the quasi-peak detection based on the frequency information, and weighting the maximum level value with the response coefficient to output the quasi-peak value of the evaluation signal It has an arithmetic unit.

A quasi-peak evaluation apparatus according to another aspect of the present invention includes a selection unit that extracts and outputs an evaluation signal to be evaluated from an input signal, and detects a maximum level of the evaluation signal and outputs the value. Peak detecting means for outputting, frequency detecting means for detecting the repetition frequency of the input signal and outputting it as frequency information, and coefficient output means for outputting a response coefficient corresponding to a detection response of quasi-peak detection based on the frequency information; , A calculator for weighting the maximum level value with a response coefficient to output a quasi-peak value of the evaluation signal, and a detection element, a time constant element for determining a charge and discharge time constant, and detecting the evaluation signal. A quasi-peak detector that outputs a quasi-peak value, and a quasi-peak value that is output by a calculator based on the frequency information output by the frequency detection unit and a quasi-peak that is output by the quasi-peak detector. And compare them with any of them Those having a comparator switching means to quasi-peak.

[Action]

According to the present invention, an evaluation signal to be evaluated is selectively extracted from a pulse-like input signal, a maximum level value of the evaluation signal is obtained, and a quasi-peak detection according to a repetition frequency of the input signal is performed. Calculate the response coefficient corresponding to the detection response of
Output the quasi-peak value.

In another aspect of the present invention, an evaluation signal to be evaluated is extracted from an input signal, and a quasi-peak value of the evaluation signal is obtained by the above-described calculation. on the other hand,
The quasi-peak value is obtained by detecting the evaluation signal with a quasi-peak value detector comprising a detecting element and a time constant circuit for determining a charging and discharging time constant, and the quasi-peak value obtained by calculation based on the repetition frequency of the input signal. The peak value is compared with the quasi-peak value obtained by detection, and one of the quasi-peak values is output.

〔Example〕

FIG. 1 shows the configuration of an embodiment of the present invention. This configuration is designed, for example, to measure a noise component in a frequency band of 30 to 1000 MHz included in noise that fluctuates in a pulse shape with a low repetition frequency in accordance with the CISPR standard.

In FIG. 1, reference numeral 1 denotes an input signal, which indicates noise in this example. 2 is a selecting means for extracting the noise component as an evaluation signal to be evaluated from now on;
Local oscillator 2b, intermediate frequency filter 2d, and envelope detector 2
e. Reference numeral 3 denotes an evaluation signal, and 4 denotes a peak detecting means. An A / D conversion circuit 4a for converting the evaluation signal 3 into a digital signal and a digital signal sequentially transmitted from the A / D conversion circuit 4a are sequentially compared. Meanwhile, it is constituted by a peak search circuit 4b for detecting the value of the maximum level. Reference numeral 5 denotes a frequency detecting means, which comprises a waveform shaping circuit 5a and a counter 5b for counting the repetition frequency of the evaluation signal 3. Reference numeral 6 denotes a coefficient output means, which calculates the detection efficiency of the quasi-peak value based on the response conditions of the quasi-peak value in the CISPR standard in advance, has the value as a ROM table, and outputs the value as a ROM table. The detection efficiency according to the frequency information to be output is output as a response coefficient. Reference numeral 7 denotes a digital arithmetic unit which multiplies the response coefficient output from the coefficient output means 6 as a maximum level value output from the peak detection means 4 and weights the result, and outputs a quasi-peak value of the evaluation signal 3. Here, the peak detecting means 4, the coefficient output means 6, and the arithmetic unit 7 are scheduled by a CPU (not shown) or the like so as to perform consistent operations.

With such a configuration, a thin pulse-like noise having a repetition of 10 Hz is received as the input signal 1, and 80% of the noise is received from the noise.
The operation will be described by taking as an example a case where a quasi-peak value of a noise component near MHz is evaluated.

First, noise near 80 MHz is mixed with a signal from the local oscillator 2b by the mixer 2a and heterodyne-received.
It is converted into an intermediate frequency signal 2c having an intermediate frequency of MHz. The intermediate frequency signal 2c is band-limited by the intermediate frequency filter 2d and output. The BW of this intermediate frequency filter 2d is CI
It is recommended by the SPR standard. In this example, the frequency is set to 120 KHz.
Therefore, the noise equivalently between 10MHz ± 60KHz (noise component is distributed every 10Hz in this BW) is 0.1 second (1 / 10Hz)
The intermediate frequency filter 2d is input to the intermediate frequency filter 2d at intervals, and receives the time response determined by the BW 120KHz of the intermediate frequency filter 2d to become an intermediate frequency signal 2f having an envelope of a distorted form, which is output to the next stage envelope detector 2e You. The envelope detector 2e performs envelope detection on the input intermediate frequency signal 2f, converts 10 MHz ± 60 KHz into direct current, and outputs the amplitude fluctuation of the envelope as the evaluation signal 3. Here, although the repetition frequency of the evaluation signal 3 is determined by the input signal 1, its waveform is determined by BW120KHz of the intermediate frequency filter 2d, and its level is a noise component.
It is determined by the amplitude of 80MHz ± 60KHz. That is, the difference in the waveform such as the pulse width of the input signal 1 directly affects the amplitude of the evaluation signal 3 and does not result in the difference in the shape of the waveform.

The repetition frequency of the evaluation signal 3 is determined by the frequency detecting means 5.
And outputs the value to the coefficient output means 6. The coefficient output means 6 obtains the detection efficiency y of the quasi-peak detection with respect to the repetition frequency in advance using the following approximate expression, and calculates the value.
It is stored as a ROM table.

Here, T _c : charging time constant T _d : discharging time constant n: repetition frequency B: BW of the intermediate frequency filter 2 d Therefore, the coefficient output means 6
When a frequency of 10 Hz is input, n = 10 Hz, C
A value (0.058 in this case) obtained from the ISPR standard as T _c = 1 ms, T _d = 550 ms, and B = 120 KHz is output to the calculator 7 as a response coefficient.

The computing unit 7 multiplies the value of the maximum level of the evaluation signal 3 detected by the peak detection unit 4 by the response coefficient output by the coefficient output unit 6 and outputs the value as a quasi-peak value 8. The reason why the quasi-peak value 8 is established by this calculation is that the waveform of the evaluation signal 3 is the intermediate frequency filter 2d as described above.
Because it depends on If the shape of the waveform of the evaluation signal 3 is directly affected by the waveform of the input signal 1, the response coefficient cannot be determined only by the above approximate expression.

Also, in the above embodiment, when the input signal 1 is a single isolated pulse, the counter 5b receives the first one pulse, and when the next pulse does not arrive even after an appropriate time, for example, 5 seconds, , The repetition frequency of this pulse is 0.2Hz
And outputs the result to the coefficient output means 6. Although there are some errors due to the time limit, there is no practical problem.

In the quasi-peak value evaluation device of the present invention, since the detection efficiency is obtained by calculation using the calculated detection efficiency, the range corresponding to the overload coefficient shown in FIGS. This is unnecessary, and there is a great effect that the amount can be transferred to the measurement range. At the same time, accuracy is improved because no diode is used.

In the above embodiment, the envelope detector 2e is not always required. However, in this case, the next-stage A / D conversion circuit 4a must respond to the intermediate frequency. Further, the frequency detecting means 5 can directly detect the repetition frequency from the input signal 1. When measuring the quasi-peak value 8 of only the fixed frequency component, the selecting means 2 can use a narrow band filter.

Further, the peak detecting means 4, the coefficient output means 6, and the computing unit 7 in the above embodiment can be constituted by analog circuits. For example, the arithmetic unit 7 can charge the evaluation signal 3 through a diode to charge a capacitor connected to the input of an operational amplifier having a high impedance input. The coefficient output means 6 only needs to generate a DC voltage corresponding to the detection efficiency y. The arithmetic unit 7 may use an analog multiplier.

 Next, another embodiment of the present invention is shown in FIG.

FIG. 2 shows an example in which random noise and noise having a high repetition frequency can be measured.

2, the same reference numerals as those in FIG. 1 denote the same parts, and 9 denotes a quasi-peak detector, for example, the differential amplifier 34, the diode 35, the resistors 36 and 39, and the capacitor 3 in FIG.
7 and the like, and a detection circuit 9a that outputs a quasi-peak value and an A / D converter 9b that converts the quasi-peak value to a digital signal and outputs a quasi-peak value 11. Numeral 12 denotes comparison switching means. The magnitudes of the quasi-peak value 8 output by the arithmetic unit 7 and the quasi-peak value 11 output by the quasi-peak detector 9 based on the frequency information from the frequency detection means 5 are shown. And outputs one of the quasi-peak values. Since the operation of the quasi-peak detector 9 is the same as that described in the section of the prior art, the operation of this embodiment will be described based on the operation of the comparison switching means 12.

The comparison switching means 12 operates based on the frequency information received from the frequency detection means 5, and FIG. 3 is a flowchart at that time. In FIG. 3, (1) to (5) show each step of the operation. Note that the frequency detection means 5 outputs a zero as frequency information when the counter 5b does not operate when the noise to be measured is random noise or a mixed noise of random noise and pulse noise, and tries to measure. The noise is mainly composed of pulsed noise. When the counter 5b operates and counts the repetition frequency of the pulsed noise, the value is output as frequency information. Therefore, the comparison switching means 12 operates as follows, and outputs either the quasi-peak value 8 or 11 described above.

(A) First, it is determined whether or not the frequency information is zero (step (1)). If the frequency information is zero, that is, if it is mainly random noise as described above, a quasi-peak value 11 is output (step (5)). ).

(B) When the frequency information is a certain repetition frequency f,
That is, in the case of mainly panel-like noise, it is determined whether or not f> 20 Hz (step (2)).
11 is output (step (5)).

(C) If f20 Hz, the quasi-peak value 8 and the quasi-peak value 11 are compared (step (3)), and the larger quasi-peak value is selected and output (steps (4) and (5)). )).
In this case, the larger value is selected when the random noise is included in the pulse-like noise, and as the magnitude approaches the pulse-like noise, the quasi-peak value 8 becomes a value lower than the real value. This is because

In this case, the value of the quasi-peak value 11 is not a problem because the effective overload coefficient of the overload coefficient necessary for the operation of the quasi-peak value detector 9 decreases due to random noise.

In this embodiment, the case where the frequency output by the frequency detecting means 5 is 20 Hz is one of the criteria for determining which of the quasi-peak values 8 and 11 is to be selected. This is because, when the repetition frequency of the pulse-like noise is about 20 Hz, the overload coefficient of the detection circuit 9a that performs quasi-peak detection is reduced by half from the overload coefficient of the isolated pulse from FIG. In this example
20Hz is used as a decision boundary. As described above, this frequency may be selected so that the overload coefficient of the detection circuit 9a does not increase. Accordingly, the overload coefficient of the circuit before the selection means 2 is also reduced, and the measurement range can be extended accordingly.

Neither of the two embodiments described above limits the measurement target to noise.

〔The invention's effect〕

As described above, the present invention extracts and outputs an evaluation signal to be evaluated from a pulse-like input signal, and a selection unit that detects and outputs a maximum level of the evaluation signal and a peak detection unit that outputs the value. Frequency detection means for detecting the repetition frequency of the input signal and outputting it as frequency information; coefficient output means for outputting a response coefficient corresponding to the detection response of the quasi-peak detection based on the frequency information; and response to the maximum level value A calculator that performs quasi-peak value of the evaluation signal by performing weighting operation with coefficients is provided, so that the overload coefficient when measuring the quasi-peak value of a pulse-like signal at a low frequency is significantly reduced. Accordingly, there is an effect that the measurement range can be widened and the measurement itself becomes easy. Further, since the measurement range is widened and the diode need not be used at the same time, there is an effect that the accuracy is improved.

Further, as described above, another aspect of the present invention is a selection means for extracting and outputting an evaluation signal to be evaluated from an input signal, and a peak detection for detecting a maximum level of the evaluation signal and outputting the value. Means, a frequency detection means for detecting a repetition frequency of the input signal and outputting the same as frequency information, a coefficient output means for outputting a response coefficient corresponding to a detection response of quasi-peak detection based on the frequency information, A computing unit that performs a weighting operation on a value with a response coefficient to output a quasi-peak value of the evaluation signal, and a detecting element and a time constant element that determines a charge and discharge time constant, and detects the quasi-peak by detecting the evaluation signal. A quasi-peak detector that outputs a peak value is compared with a quasi-peak value output by a calculator and a quasi-peak value output by a quasi-peak detector based on frequency information output by a frequency detection unit. And
Since comparison switching means for quasi-peak value of any of them is provided, there is an effect that so-called unspecified signals other than pulse-like signals having a low repetition frequency can be measured more accurately than conventional quasi-peak value detection. . Further, since the overload coefficient is significantly reduced, there is an effect that a general-purpose measuring instrument such as a spectrum analyzer can be additionally used without imposing a large load.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of one embodiment of the present invention, FIG. 2 is a diagram showing a configuration of another embodiment of the present invention, and FIG. 3 is a flow of operation of another embodiment of the present invention. FIG. 4 is a diagram showing a configuration of a conventional example in which quasi-peak detection is performed after frequency conversion. FIG. 5 is a circuit diagram showing a specific example of a conventional quasi-peak detector. FIG. 7 and FIG. 7 are diagrams for explaining the characteristics of the conventional example. In the figure, 1 is an input signal, 2 is a selection means, 2a is a mixer, 2b is a local oscillator, 2c is an intermediate frequency signal, 2d is an intermediate frequency filter, 2e is an envelope detector, 3 is an evaluation signal, and 4 is an evaluation signal. Peak detection means, 4a and 9b are A / D converters, 4b is a peak search circuit, 5
Is a frequency detecting means, 5a is a waveform shaping circuit, 5b is a counter,
6 is a coefficient output means, 7 is a calculator, 8, 10, and 11 are quasi-peak values,
9, 30 is a quasi-peak detector, 9a is a detection circuit, 12 is comparison switching means, 31 is display means, 32 and 35 are diodes, 33 is a smoothing circuit, 34 is a differential amplifier, and 36 and 39 are resistors. , 37 are capacitors.

Claims (3)

(57) [Claims] A selecting means for extracting and outputting an evaluation signal to be evaluated from a pulse-like input signal; and a peak detecting means for detecting a maximum level of the evaluation signal and outputting the value. 4) and frequency detecting means (5) for detecting the repetition frequency of the input signal and outputting it as frequency information.
Coefficient output means (6) for outputting a response coefficient corresponding to a detection response of the quasi-peak detection based on the frequency information;
An arithmetic unit (7) for performing a weighting operation on the value of the maximum level with the response coefficient and outputting a quasi-peak value of the evaluation signal
A quasi-peak evaluation device comprising: 2. The method according to claim 1, wherein the selecting means (2) heterodyne-receives a frequency component to be evaluated from the pulse-like input signal, limits the band with an intermediate frequency filter, and outputs this signal as an evaluation signal. The quasi-peak value evaluation device according to claim 1, wherein 3. A selecting means (2) for extracting and outputting an evaluation signal to be evaluated from an input signal, and a peak detecting means (4) for detecting a maximum level of the evaluation signal and outputting the value. Frequency detecting means (5) for detecting a repetition frequency of the input signal and outputting it as frequency information; and coefficient output means (6) for outputting a response coefficient corresponding to a detection response of quasi-peak detection based on the frequency information. ), A computing unit (7) for performing a weighting operation on the value of the maximum level with the response coefficient and outputting a quasi-peak value of the evaluation signal, and a time constant element for determining a detection element, a charge and discharge time constant A quasi-peak detector (9) for detecting the evaluation signal and outputting a quasi-peak value, and a quasi-peak output from the arithmetic unit based on the frequency information output from the frequency detecting means. The peak value and the quasi-peak detector are output. Quasi-peak values and compares the comparison switching means for one of them and quasi-peak (12) quasi-peak evaluation apparatus characterized by having a to.