JP7073826B2 - Long-period noise capture performance evaluation system and long-period noise capture performance evaluation method - Google Patents

Description

The present invention relates to a long-period noise capture performance evaluation system and a long-period noise capture performance evaluation method in an EMC (electromagnetic compatibility) measurement system.

FIG. 12 is an example of a radiation emission measuring device in EMC measurement, and is a turntable 1 arranged on the floor surface of an anechoic chamber, an antenna 3 supported by an antenna positioner 2 at a height set from the floor surface, and a turntable. It includes a mechanism control unit 4 that controls the rotation of 1 and the height of the antenna 3 via the antenna positioner 2, and a receiver 5. The EUT (Equipment Under Test) 7 to be tested, the turntable 1 on which the device is placed, the antenna 3 and the antenna positioner 2 are arranged in an anechoic chamber or an open site. The mechanism control unit 4 and the receiver 5 are arranged in a measurement room outside the anechoic chamber or in a measurement room laid out under the ground plane at an open test site so as not to affect the measurement.

In the EMC measurement, the electromagnetic wave (radiation emission) emitted by the EUT is measured. In the radiation emission measurement, the interference wave that may exceed the permissible value among the radiation interference waves from the EUT is selected by the first prescan. .. Prescan uses receivers, antenna positioners, turntables, etc. to identify the approximate frequency and maximum radiation direction of interfering waves that may exceed tolerances.

Conventionally, a spectrum analyzer has been mainly used as the receiver 5 in FIG. 12 in the prescan. The spectrum analyzer performs frequency sweep and measures the frequency characteristic of the noise radiated from the EUT 7. Therefore, if the maximum radiation position is specified by moving the antenna position via the antenna positioner 2 while taking a sufficient residence time (observation time) for each frequency, the measurement time becomes very long, which is not realistic. Therefore, it was not possible to set a sufficient residence time, it was difficult to detect waves other than short-period and continuous interfering waves, and the capture of long-period noise and pulsed noise depended on the experience level of the measurer. ..

In recent years, as a receiver for EMC measurement, the introduction of a measuring instrument using a time domain scan has been progressing. In a time domain receiver or a real-time spectrum analyzer, data in a wide frequency domain is basically obtained by FFT (Fast Fourier Transform) a signal sampled in the time domain. Therefore, in the spectrum analyzer, the same result as the measurement result obtained by taking the residence time for each measurement frequency can be obtained from the sampling data in one time domain. Therefore, it has become realistic to perform prescan with a long residence time, and it has become possible to capture long-period noise, which has been difficult to capture until now.

The com generator is an oscillator used to generate a reference signal for evaluating frequency characteristics in the start-up inspection of test equipment for radiation emission measurement. In the radiation emission measurement, the frequency band of the measurement standard is determined to be, for example, 30 MHz-1 GHz, 1 GHz-6 GHz or the like. One of the features of the com generator is that it has a wide frequency band so that these measurement bands can be evaluated collectively. Further, in the band, it has a comb-shaped spectrum having a certain step such as 1 MHz and 10 MHz. In the evaluation and inspection of the test equipment for radiation emission measurement, the signal emitted by the com generator is received and the amplitude value of each spectrum in the frequency band is measured to perform evaluation and abnormality detection.

Looking at the signal transmitted from the conventional comb generator in the time domain, as shown in FIG. 13, pulses having _a pulse width τ are continuously transmitted in a fixed period t1 without interruption. Period t ₁ is the reciprocal of the spectral spacing in the frequency domain. For example, the pulse transmission cycles t ₁ of a commonly used 1 MHz and 10 MHz step com generator are 1 μs and 100 ns, respectively. Further, the pulse width τ and the spectral bandwidth of the frequency domain are also in an inversely proportional relationship, and the narrower the pulse width τ is, the wider the bandwidth is. Specifically, each harmonic frequency n times (n: natural number) of the fundamental frequency 1 / t ₁ has a spectral component, and the amplitude of each harmonic frequency is proportional to the sampling function Sa (nπτ / t ₁ ). If the pulse transmission cycle t ₁ is constant, the narrower the pulse width τ, the smaller the coefficient of the variable n of the sampling function Sa (nπτ / t ₁ ). Therefore, the smaller the pulse width τ, the longer the change in the sampling function with respect to n, so that it has a high frequency component.

US Pat. No. 5,594,458 Patent Document 1 describes a technical description of a general comb generator itself and a calibration method of an EMC measuring facility using the comb generator. Japanese Unexamined Patent Publication No. 2005-49200 Patent Document 2 is a short pulse generation circuit, and shows a technique of combining a transistor and a delay element to shape a waveform so as to shorten the pulse width.

When an EMC measurement system incorporating a time domain receiver capable of capturing long-period noise as described above is constructed, it is verified whether long-period noise can be captured as designed over the entire frequency band specified by the measurement standard. There is a need to do. In particular, in the case of long-period noise, since the frequency of occurrence itself is low, if the noise cannot be captured due to an abnormality in the EMC measurement system, it is more difficult to notice the occurrence of the abnormality than in the short-period noise. Therefore, it is desirable to confirm whether long-period noise can be captured as expected by actual measurement.

However, when verifying the capture performance of long-period noise, a conventional comb generator cannot be used as a reference signal. As described above, the conventional com generator transmits pulses continuously in cycles of ns and μs orders, so noise with a period of ms order or more, which is generally regarded as long-period noise, It cannot be used for evaluation as an alternative.

In view of the above problems, an object of the present invention is to provide a long-period noise capture performance evaluation system and a long-period noise capture performance evaluation method capable of evaluating the capture performance of long-period noise by actual measurement. There is.

The first aspect of the present invention is a long-period noise capture performance evaluation system in an EMC measurement system including an antenna and a receiver to which a received signal of the antenna is input.
It is equipped with a com generator that generates a reference signal and a reception data determination device that inputs the output signal of the receiver.
When the observation time of the receiver is D (s), the comb generator has the spectral peak interval 1 / t ₁ (Hz) of desired equal intervals as the reference signal. A burst containing a plurality of pulses having a pulse period t ₁ (s), which is the reciprocal of t ₁ , is continuously transmitted in a burst period T (s) such that T> D.
The burst is a pulse having a pulse period of t ₁ (s) and a pulse width of 1 (ns) or less per burst.
k × D / t ₁ +1 (However, the coefficient k is a positive value less than 1)
Contains the following integers,
The EMC measurement system measures the reference signal a plurality of times, and in the reception data determination device, the EMC measurement system normally produces long-period noise from the ratio of the number of times the reference signal is received to the number of times the measurement is performed. It is characterized in that it is determined whether or not it has been captured.

The received data determination device compares the ratio of receiving the reference signal with the expected value in the range of (1 + k) × D / T or less at D / T or more, so that the EMC measurement system normally has a long period. It is good to judge whether or not the noise is captured.

A second aspect of the present invention is a method for evaluating long-period noise capture performance in an EMC measurement system.
When the observation time of the receiver included in the EMC measurement system is D (s),
A pulse with a pulse period t ₁ (s) that is the inverse of the spectral peak spacing 1 / t ₁ so as to have the desired equidistant spectral peak spacing 1 / t ₁ (Hz) from the comb generator that generates the reference signal. Bursts containing a plurality of bursts are continuously transmitted in a burst period T (s) such that T> D, and the burst is a pulse having a pulse period t ₁ (s) and a pulse width of 1 (ns) or less. Per burst
k × D / t ₁ +1 (However, the coefficient k is a positive value less than 1)
Include the following integers,
Whether or not the EMC measurement system normally captures long-period noise is determined from the ratio of the number of times the reference signal is received to the number of times the reference signal is measured by the EMC measurement system a plurality of times. It is characterized by making a judgment.

Whether or not the EMC measurement system normally captures long-period noise by comparing the rate of receiving the reference signal with the expected value in the range of (1 + k) × D / T or less at D / T or more. Should be determined.

The coefficient k is preferably a numerical value in which the coefficient k is larger than 0 and 0.1 or less.

It should be noted that any combination of the above components and a conversion of the expression of the present invention between methods, systems and the like are also effective as aspects of the present invention.

According to the long-period noise capture performance evaluation system and the long-period noise capture performance evaluation method according to the present invention, it is possible to evaluate the long-period noise capture performance by actual measurement.

The block diagram which shows the embodiment of the long-period noise acquisition performance evaluation system and method in the EMC measurement system which concerns on this invention. The waveform diagram of the reference signal (time domain display) transmitted by the com generator as the reference signal generator in the embodiment. A spectrum diagram of the reference signal (frequency domain display) transmitted by the com generator. An explanatory diagram in the case where the burst length of the reference signal is very short compared to the observation time in the embodiment. An explanatory diagram in the case where the burst length of the reference signal cannot be ignored as compared with the observation time in the embodiment. Power spectrum diagram of a single pulse with a pulse width of 0.8 ns. FIG. 3 is a time waveform diagram of a reference signal having a burst length of 100 ns that can be used in the embodiment. The power spectrum diagram of the reference signal of FIG. 7. FIG. 3 is a time waveform diagram of a reference signal having a burst length of 500 ns that can be used in the embodiment. The power spectrum diagram of the reference signal of FIG. The block diagram of the com generator in the embodiment. The block diagram which shows an example of a radiation emission measurement system. Waveform diagram of the reference signal (time domain display) of the conventional com generator.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. The same or equivalent components, members, processes, etc. shown in the drawings are designated by the same reference numerals, and duplicate description thereof will be omitted as appropriate. Further, the embodiment is not limited to the invention but is an example, and all the features and combinations thereof described in the embodiment are not necessarily essential to the invention.

FIG. 1 is an example of the EMC measurement system 10 according to the embodiment of the present invention, and is supported by a turntable 1 and an antenna positioner 2 arranged on the floor surface of an anechoic chamber at a height set from the floor surface. The antenna 3 and the receiver 5 that receives the received signal of the antenna 3 (the received signal is input) are provided. The EUT (Equipment Under Test) to be tested, the turntable 1 on which the device is placed, the antenna 3 and the antenna positioner 2 are arranged in an anechoic chamber or an open site. The receiver 5 is arranged in a measurement room outside the anechoic chamber or in a measurement room laid out under the ground plane at an open test site so as not to affect the measurement. Further, although not shown, a mechanism control unit for controlling the rotation of the turntable 1 and the height of the antenna 3 via the antenna positioner 2 is provided. The operation of the turntable 1, the antenna positioner 2, the receiver 5, and the like is controlled by the predetermined measurement control program 8. The receiver 5 is, for example, a time domain receiver, a real-time spectrum analyzer, a digital storage oscilloscope, or the like.

A long-period noise capture performance evaluation system 20 for verifying the capture performance of long-period noise of the EMC measurement system 10 is added to the EMC measurement system 10. The long-period noise capture performance evaluation system 20 receives the received data obtained by executing the measurement control program 8 and the comb generator 21 as a reference signal generator that radiates a reference signal to the antenna 3 (received data is input). It has a received data determination device 22 for determining received data.

As a reference signal, a comb generator 21 that transmits a periodic burst wave (burst pulse) shown in FIG. 2 in the time domain is used. Here, in the periodic burst wave of FIG. 2, a burst, which is a short-period pulse group having a pulse period t ₁ (s), is continuously (repeatedly) in a long period (burst period = reference signal generation period T). It is a generated waveform. The number of pulses per burst is set to an integer of (T / t ₁ ) -1 or less when the pulse period is t ₁ . By providing a period during which no pulse of at least one pulse cycle is transmitted between bursts, the signal in the time domain is divided for each burst. The pulse width τ of the pulse having the pulse period t ₁ is, for example, 1 ns or less, and when the observation time of the receiver 5 is D (s), the number of short-period pulses per burst is, for example, 0.1 × (D / t). ₁ ) Set to +1. In this case, the reference signal is a signal having a bandwidth of 1 GHz or more (the frequency spectrum does not cross zero up to 1 GHz) at a spectral peak interval of 1 / t ₁ (Hz) in the frequency domain, as shown in FIG. Therefore, for example, when the measurement standard is 30 MHz-1 GHz, the frequency characteristics of the entire band can be evaluated collectively. Further, by regarding the generation cycle of one burst shown in FIG. 2 {reference signal generation cycle T (s)} as one long-period noise cycle, noise is generated for each cycle T in the time domain of the reference signal. It can be treated as a signal that simulates long-period noise.

Consider a case where the observation time is set to D and the reference signal transmitted from the com generator 21 is measured for the receiver 5 incorporated in the EMC measurement system 10 to be evaluated. The receiver 5 samples the time signal at the set sampling cycle during the observation time D. Then, by FFTing this time signal, the characteristics of the frequency domain of the required band are acquired.

First, consider a case where the burst length of the reference signal is set to be negligibly shorter than the burst cycle (= reference signal generation cycle T) and the observation time D of the receiver 5, as shown in FIG. When the receiver 5 receives a control command from the measurement control program 8 of the EMC measurement system 10 and starts observation at an arbitrary time, the expected value of the probability of receiving the reference signal transmitted from the com generator 21 is D. / T. This is because the comb generator 21 always transmits the reference signal in the period T, so that the probability that the comb generator 21 transmits the reference signal during the observation time D, in other words, the probability that the receiver 5 receives the reference signal. Is D / T. The EMC measurement system 10 incorporating the receiver 5 observes the reference signal a sufficient number of times, and the reception data determination device 22 determines whether or not the reference signal is received for each measurement and peak frequency, and receives the received data. When the probability is calculated, if the EMC measurement system to be evaluated is correctly constructed, the reception probability of the reference signal converges to D / T. On the contrary, when the reception probability deviates significantly from the D / T, it is possible to detect that there is some abnormality. When performing the above evaluation, the measurement control program 8 of the EMC measurement system 10 to be evaluated controls not only the receiver 5 but also other devices included in the system such as the turntable 1 and the antenna positioner 2, and receives data. By evaluating the received data acquired while performing the same processing as in the actual EMC measurement such as the calculation of the above by the received data determination device 22, it is possible to evaluate the capture performance of long-period noise in actual use.

Up to this point, the case where the burst length is negligible compared to the observation time D has been described, but in reality, the burst of the reference signal has a length. Therefore, as shown in FIG. 5, when the burst length B of the reference signal cannot be ignored as compared with the observation time D, the burst may be located at the end of the observation time D. In that case, the amplitude value of the received signal is lower than that in the state where it is completely within the observation time. When the burst length is B, the probability that the burst length B is 100% within the observation time D is (DB) / T, and the probability that a part of the burst length B is within the observation time D is 2B / T.

In the present embodiment, when the number of pulses per burst is k × (D / t ₁ ) + 1 (however, a positive number less than the coefficient k: 1 is preferably 0.1 or less), the burst length B in FIG. Is limited to the observation time D × k. The probability that part or all of the burst length B enters the observation time D is (1 + k) D / T. Moreover, the probability that only a part of the burst enters within the observation time is 2Dk / T. When a part of the burst length falls within the observation time D, the amplitude value of the received signal obtained by the receiver 5 fluctuates in proportion to the length of the time entering the observation time D. Therefore, the expected value is changed by adjusting the threshold level for determining whether or not the reference signal of the received data determination device 22 that receives the received signal is received. For example, if the threshold level is set close to the level in the state where there is no received signal (the state in which the reference signal is not received), the expected value becomes the maximum (1 + k) D / T. If the threshold level is set near the level when 100% of the burst length is within the observation time, the expected value is (1-k) D / T.

For example, a case where the coefficient k is selected to be 0.1 or less will be specifically described. That is, when the number of pulses per burst is 0.1 × (D / t ₁ ) +1 or less, the burst length B in FIG. 5 is limited to 10% or less of the observation time D. When the burst length is set to 0.1D at the maximum, the probability that a part or all of the burst length B enters the observation time D is 1.1 × D / T. In addition, the probability that only a part of the burst will be within the observation time is 0.2 × D / T or less. When a part of the burst length falls within the observation time D, the amplitude value of the received signal obtained by the receiver 5 fluctuates in proportion to the length of the time entering the observation time D. Therefore, the expected value is changed by adjusting the threshold level for determining whether or not the reference signal of the received data determination device 22 that receives the received signal is received. For example, if the threshold level is set close to the level in the state where there is no received signal (the state in which the reference signal is not received), the expected value becomes the maximum 1.1 × D / T. If the threshold level is set near the level when 100% of the burst length is within the observation time, the expected value is 0.9 × D / T. However, in either case, it is not preferable because it is easily affected by noise in determining whether or not the reference signal is received. Therefore, preferably, if the S / N ratio of the received signal is sufficient, a part of the received signal can be used for the observation time by shortening the burst length as much as possible compared to the reference signal generation cycle and the observation time of the receiver. It is desirable to reduce the rate of entry. If the ratio of a part of the received signal entering the observation time is sufficiently small, the expected value of the probability of receiving the reference signal may be regarded as D / T without adjusting the threshold level.

In view of the above points, the coefficient k is preferably set to 0.1 or less, and if it is larger than 0.1, the expected value due to the threshold level tends to fluctuate, which is not preferable.

<Example of reference signal transmission waveform>
An example of the transmission waveform of the reference signal generated by the comb generator 21 will be described with reference to FIGS. 6 to 10. As a desired reference signal, the transmission waveform of the reference signal when the bandwidth is 1 GHz, the spectrum peak interval is 100 MHz, and the reference signal generation cycle (burst cycle) is 1 ms will be described.

First, in order to output the desired reference signal, the setting of each parameter for the transmission waveform in the time domain will be described. First, considering the pulse width τ of a short-period pulse contained in a burst, the Fourier transform of a single pulse wave represented by a rectangular function in the time domain becomes a sinc function. If the pulse width τ is set to 1 ns or less, the frequency spectrum does not cross zero up to 1 GHz. FIG. 6 shows the power spectrum of a single pulse when the pulse width τ is 0.8 ns. The power spectrum value near 1 GHz is about 10 dB lower than that near DC.

Next, considering the pulse period t ₁ of the short-period pulse, the pulse period t ₁ is 10 ns because it is the reciprocal of the desired spectral peak interval of 100 MHz. Further, since the number of transmission pulses per burst is 0.1 × (burst cycle / pulse cycle) +1 or less, the number of pulses is set to 10,001 or less in this example. However, as described above, in the evaluation of pulse acquisition performance, it is desirable that the burst length is short because it is desired to reduce the ratio of a part of the burst length entering the observation time. Therefore, FIG. 7 shows a waveform in the time domain when the number of pulses is 11 and the burst length is 100 ns. FIG. 8 shows the power spectrum of the waveform of FIG. 7.

Further, FIG. 9 shows a time waveform when the number of pulses is set to 51 and the burst length is set to 500 ns as a comparison with FIG. 7. FIG. 10 shows the power spectrum of this waveform.

The spectral peak spacing of the power spectra of FIGS. 8 and 10 is both the desired 100 MHz. Further, the shape of the envelope of the spectrum peak is the same as the spectrum shape of the single pulse shown in FIG. 6, and the power spectrum value at 1 GHz is about 10 dB lower than the value at DC. Therefore, a desired reference signal can be obtained by transmitting the time waveform shown in FIG. 7 or 9 at a cycle of 1 ms. Further, when comparing the power spectrum levels at the respective spectral peaks of FIGS. 8 and 10, the number of pulses per burst is increased, and FIG. 10 is about 13 dB higher. When used for evaluation of pulse acquisition performance as described above, it is desirable that the burst length is short. However, for example, when the reference signal from the comb generator 21 is propagated over a long distance and used for evaluation, the S / N ratio may become small. In such a case, it is possible to improve the S / N ratio by increasing the number of pulses per burst, if necessary.

<Evaluation of noise capture performance in long-period noise capture performance evaluation system>
As shown in FIG. 1, a long-period noise capture performance evaluation system 20 is added to the EMC measurement system 10, and a receiver 5 having an observation time of 100 μs is assumed in the EMC measurement system 10 assuming a long-period noise with a period of 1 ms. The following is an example of evaluating the long-period noise capture performance by incorporating the above.

As described with reference to FIGS. 4 and 5, if the burst length B is short, the probability that the end of the observation time D of the receiver 5 will be reached is low, so that the burst length B is preferably short. A case where the burst length B is set short to 100 ns, which is 0.1% of the observation time D (in the case of the reference signals of FIGS. 7 and 8), will be described.

When observing the reference signal transmitted from the comgenerator 21 so that the number of times is sufficient under this condition, the approximate expected value of receiving the reference signal is observation time / burst period = 0.1, which is about 10%. It can be received with the probability of. However, as described above, when the burst length B is finite, the burst may be applied to the end of the observation time and the reception level may fluctuate. In the case of this example, the probability that a part of the burst length is included in the observation time is 0.02% from 2 × burst length / burst period. Further, the probability that the burst length will be 100% within the observation time is 9.99% from (DB) / T, which is close to the expected value of 10% when the burst length is negligible. In this case, the probability of receiving an amplitude value other than the amplitude value when the burst enters 100% within the noise level and the observation time is 0.02%, which is sufficiently low, so that the threshold for determining whether or not the reference signal is received is determined. The effect of setting the level on the expected value of the capture rate is reduced. For example, in order to reduce the influence of noise, it is considered that a value of 10% can be applied to the expected value of the capture rate even when the amplitude value is set to half when the burst enters 100% within the observation time.

Further, as described above, when the S / N ratio of the received reference signal is low, the S / N ratio is improved by lengthening the burst length B and increasing the number of pulses per burst. However, when the burst length B is set long, the probability that a part of the burst length B falls within the observation time increases. As an example, a case where the burst length B is increased to 5 times the previous example and the burst length B is set to 500 ns, which is 0.5% of the observation time D (in the case of the reference signals of FIGS. 9 and 10) will be described. do. In this case, the probability that a part of the burst length B is included in the observation time D is 0.1%, which is 5 times higher than that of 2B / T. Further, the probability that the burst length B is 100% within the observation time D is reduced to 9.95% from (DB) / T. In this case, if the threshold level is set to half of the amplitude value when the burst enters 100% within the observation time as in the previous example, the signal of the reception level close to the threshold level increases, so compared to the previous example. , Susceptible to noise.

<Hardware configuration of the com generator that serves as the reference signal generator>
FIG. 11 shows a configuration example of the hardware of the comb generator 21 that outputs a periodic burst wave with a pulse width of 1 ns or less and an arbitrary period and a number of pulses. The comb generator 21 has a burst wave generation circuit unit 31 that periodically generates a desired burst wave with a pulse having a pulse width wider than 1 ns, and a pulse width τ of the burst wave output from the burst wave generation circuit unit 31 of 1 ns or less. It is provided with a waveform shaping circuit unit 32 for shaping. As the burst wave generation circuit unit 31, for example, when an FPGA (Field-Programmable Gate Array) operating at 200 MHz, which is a kind of logic circuit, is used, the clock period is 5 ns, so that an arbitrary logic waveform can be generated with a time resolution of 5 ns. It is possible to occur. The waveform shaping circuit unit 32 can be configured by a well-known technique such as Patent Document 2.

When a periodic burst wave containing a plurality of pulses having a pulse period t ₁ shown in FIG. 2 is transmitted as a reference signal from the comb generator 21 in the frequency domain, for example, the spectral peak interval is as shown in FIG. 3 in the frequency domain. It becomes a signal having a bandwidth of ₁ GHz or more at 1 / t1. Further, the pulse having the pulse period t ₁ is set so as to include an integer of (T / t ₁ ) -1 or less per burst. By providing a period during which no pulse of at least one pulse cycle is transmitted between bursts, it is possible to divide the signal in the time domain for each burst, and one burst is noise generated during one cycle of long-period noise. By considering it as, long-period noise (for example, noise with a period of ms order) can also be simulated (the burst period is the sum of the burst length and the time until the next burst arrives).

<Evaluation method of EMC measurement system>
The long-period noise capture performance evaluation method of the present invention can be used for evaluation of a radiation emission measurement system as shown in FIG. Here, instead of the EUT7 which is the device under test, a comb generator 21 as a reference signal generator is installed, and a reference signal of a burst wave is transmitted. The radiated emission measurement system repeats the measurement of the electric field strength a sufficient number of times. Radiation emissions are calculated by calculating the measured value of the capture rate from the number of times the reference signal is captured by the receiver 5 and comparing it with the burst period T of the reference signal and the expected value calculated from the observation time D of the receiver 5. It is possible to verify whether the entire measurement system is properly controlled.

According to this embodiment, the following effects can be obtained.

(1) When the observation time of the receiver 5 included in the EMC measurement system 10 is D, a pulse having a pulse period t ₁ that is the inverse of the desired spectral peak interval 1 / t ₁ from the comb generator 21 that generates the reference signal. Bursts containing a plurality of bursts are continuously transmitted in a burst period T such that T> D, the reference signal is measured a plurality of times by the EMC measurement system 10, and the number of times the reference signal is received with respect to the number of times the measurement is performed. It is possible to determine whether or not the EMC measurement system normally captures long-period noise from the ratio of. Therefore, it is possible to evaluate the capture performance of long-period noise by actual measurement.

(2) In the reference signal, a pulse having a pulse width of 1 ns or less and a burst pulse containing an integer number of 0.1 × D / t ₁ + 1 or less per burst are repeatedly and continuously transmitted, and the reference signal is received. When the burst length is determined by comparing the ratio with the expected value in the range of 1.1 × D / T or more at D / T or more to determine whether the EMC measurement system 10 normally captures long-period noise. B becomes shorter than the observation time D, and the fluctuation of the expected value due to the setting of the threshold level can be reduced.

(3) The comgenerator 21 transmits a burst wave as a reference signal, and is set so that short-period pulses having a pulse period of t ₁ are included in an integer number of (T / t ₁ ) -1 or less per burst. Therefore, it is possible to provide a period during which no pulse of at least one pulse cycle is transmitted between bursts, and it becomes possible to divide the signal in the time region for each burst, and one burst has one cycle of long-cycle noise. Long-period noise can also be simulated by regarding it as noise generated in the meantime.

Although the present invention has been described above by taking the embodiment as an example, it is understood by those skilled in the art that various modifications can be made to each component and each processing process of the embodiment within the scope of the claims. By the way. Hereinafter, a modification example will be touched upon.

The present invention can be applied to various EMC measurement systems by changing the pulse width τ, period t ₁ and burst period T of short-period pulses according to the measurement frequency range.

As the circuit configuration of the comb generator according to the present invention, any configuration can be adopted as long as the target burst wave can be generated.

1 Turntable 2 Antenna positioner 3 Antenna 4 Mechanism control unit 5 Receiver 7 EUT
8 Measurement control program 10 ECM measurement system 20 Long-period noise capture performance evaluation system 21 Comgenerator 22 Received data judgment device 31 Burst wave generation circuit unit 32 Waveform shaping circuit unit

Claims (6)

A long-period noise capture performance evaluation system in an EMC measurement system having an antenna and a receiver to which a received signal of the antenna is input.
It is equipped with a com generator that generates a reference signal and a reception data determination device that inputs the output signal of the receiver.
When the observation time of the receiver is D (s), the comb generator has the spectral peak interval 1 / t ₁ (Hz) of desired equal intervals as the reference signal. A burst containing a plurality of pulses having a pulse period t ₁ (s), which is the reciprocal of t ₁ , is continuously transmitted in a burst period T (s) such that T> D.
The burst is a pulse having a pulse period of t ₁ (s) and a pulse width of 1 (ns) or less per burst.
k × D / t ₁ +1 (However, the coefficient k is a positive value less than 1)
Contains the following integers,
The EMC measurement system measures the reference signal a plurality of times, and in the reception data determination device, the EMC measurement system normally produces long-period noise from the ratio of the number of times the reference signal is received to the number of times the measurement is performed. A long-period noise capture performance evaluation system that determines whether or not it is capturing. The reception data determination device compares the ratio of receiving the reference signal with the expected value in the range of (1 + k) × D / T or less at D / T or more, so that the EMC measurement system normally has a long cycle. The long-period noise capture performance evaluation system according to claim 1 , which determines whether or not noise is being captured. The long-period noise capture performance evaluation system according to claim 1 or 2 , wherein the coefficient k is a numerical value greater than 0 and 0.1 or less. This is a method for evaluating long-period noise capture performance in EMC measurement systems.
When the observation time of the receiver included in the EMC measurement system is D (s),
A pulse with a pulse period t ₁ (s) that is the inverse of the spectral peak spacing 1 / t ₁ so as to have the desired equidistant spectral peak spacing 1 / t ₁ (Hz) from the comb generator that generates the reference signal. Bursts containing a plurality of bursts are continuously transmitted in a burst period T (s) such that T> D, and the burst is a pulse having a pulse period t ₁ (s) and a pulse width of 1 (ns) or less. Per burst
k × D / t ₁ +1 (However, the coefficient k is a positive value less than 1)
Include the following integers,
Whether or not the EMC measurement system normally captures long-period noise is determined from the ratio of the number of times the reference signal is received to the number of times the reference signal is measured by the EMC measurement system a plurality of times. Long-period noise capture performance evaluation method to judge. Whether or not the EMC measurement system normally captures long-period noise by comparing the rate of receiving the reference signal with the expected value in the range of (1 + k) × D / T or less at D / T or more. The long-period noise capture performance evaluation method according to claim 4 . The long-period noise capture performance evaluation method according to claim 4 or 5 , wherein the coefficient k is a value greater than 0 and 0.1 or less.

Images