JPS5940217A - Method for measuring quality of inharmonic degree in multipath transmitting system - Google Patents

Abstract

PURPOSE:To measure psychological quality in a transmitting system highly accurately, by applying a testing signal to a multipath transmitting system, and measuring the inharmonic degree of the received signal at a receiving point at timing based on the shortest delay time at the receiving point. CONSTITUTION:A testing signal is supplied to a multipath transmitting system. The output of a microphone 22 provided at a receiving point is inputted to a ramp wave generating circuit 102 through a Schmitt trigger circuits 101. A ramp waveform is generated and sent to an AD converter 105 constituting s spectrum analyzer through an OR circuit 104. The data for each screen is inputted to a memory 106. When the memory 106 is filled up, an FET103 sends the spectrum data to an accumulation adder. The spectrum is displayed on a display device 109. Meanwhile, the ramp waveform sequentially triggers comparators 103-1- 103-3. A direct wave signal is sent to the AD converter 105 through the OR circuit 104. The psychological quality of the transmitting system is measured highly accurately based on the obtained spectrum.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a dissonance quality measurement method for a multipath transmission system with different delay times for each path. This is what I did to get it.

Conventionally, in order to measure the psychological quality of a certain transmission system, a basic measuring device has been used in the patent application filed in 1982-1970 by the applicant.
There is a "Fukyouwako measuring device" described in the specification of No. 836,
The present invention also uses the same basic configuration.

However, the measuring device described in the above specification is suitable for mainly measuring psychological quality related to nonlinear distortion of a normal transmission system, which can be measured by asynchronous observation using a signal of a stationary stochastic process, In general, signals arrive after a long and different delay time, are added vectorially, and exhibit extremely complex single-wave e/% characteristics. The drawback was that it lacked performance.

Therefore, what is the psychological quality of the multipath transmission system as mentioned above? In order to make this possible, (1) remove the discrepancy in causality between input and output signals regarding the shortest delay time in the input/output response of the transmission system; (2) remove the signal that exists in the multipath transmission system; It is necessary to provide new means or structures for two reasons: to use a signal that can capture as clearly as possible the changes in timbre occurring between the source and the receiving point.

The present invention aims to eliminate the above-mentioned conventional drawbacks and provide a dissonance quality measurement method for multipath transmission systems that can measure the psychological quality of multipath transmission systems having dramatically different delay times with high precision. be.

° That is, although the dissonance quality measurement method of the present invention has the same basic configuration as the dissonance quality measurement device described in the specification of Japanese Patent Application No. 55-70836 mentioned above, (a)
Type A uses a frequency swept sine wave, type B uses a signal consisting of a combination of 0)) fixed frequencies.
(C) Which of the three types of implementation means of type C uses the most common signals, such as signals in which different frequencies appear sequentially with level fluctuations? The test is performed by applying all test signals consisting of multiple frequency signals arranged over time to a multipath transmission system with different delay times for each bus.
At the receiving point of the multipath transmission system, a new overlap of frequency components is generated between the received signal delayed for the shortest time and other delayed received signals, and the shortest time r, J
The measurement timing is set based on the shortest delay time of the multipath transmission system detected from the received signal delayed by This method is characterized in that the psychological quality of the multipath transmission system is measured by measuring the dissonance quality of the signal.

However, a suitable example of the multipath transmission system and E7, which measures all psychological quality based on dissonance using the measurement method of the present invention, can be cited as a suitable example of a room exhibiting reverberation. The Hoto method is applicable not only to such a so-called multipath transmission system, but also to a long group delay i, which is similar in principle.
Satellite broadcasting system as a linear system with A, and furthermore, is there a talent for wound magnetic tape? It can also be widely applied to the measurement of the psychological quality of magnetic tape recording and reproducing devices that produce talostalk with time differences such as transfer between recording layers.

The present invention will be described in detail below with reference to the drawings.

(A) First, the A-type measurement method of the present invention using a swept sine wave as a test signal will be explained with full reference to FIG. 1, which shows its basic configuration.

In the illustrated structure 5k, the oscillation output signal ft(t)- of the frequency sweep oscillator 1 as shown in FIG. 2
As shown in FIG. 2(b), the reception point receives, for example, a direct sound signal fd(t) delayed by the minimum transmission time τd from the input signal ft(t), and the shortest transmission time τ.
Delay time longer than d τ□11r21..., τrn
For example, a synthesized sound signal with a plurality of reflected sound signals each delayed by τ arrives. Therefore, the signal waveform that reaches the receiving point generally receives different amounts of attenuation Ad and Ar0 to Arn for each frequency for both the direct sound signal and the reflected sound signal, so the received sound signal receives such attenuation. Each (N) is vectorially synthesized.

Next, the frequency of the direct sound signal is exactly FD engineering.

Figure 3 shows an example of superimposing the spectra of the received sound output signals including the reflected sound signals at the time when fdz, fda, and fd. The length of each frequency point on the frequency point according to the frequency resolution of the spectrum analyzer represents the iB value of the power of each spectrum. Among these spectra, the spectrum fa1 + fdg
Since + fda and fd4 are spectra of direct sound signals, only these direct sound spectra are observed when there is no reflected sound signal.

17 The psychological quality of a transmission system is determined by the input and output dissonance strengths determined from the spectrum of the input signal or output signal, respectively. The difference, that is, the increment of dissonance strength between input and output, ΔI=I. −■i
It can be obtained by converting the increment ΔD% regarding the degree of dissonance of the Mira logical quantity, that is, ΔD−Δtechβ,βhei4.

The measurement method of the present invention will be referred to as the all-simple method of type A, which calculates the increment ΔD regarding the dissonance question of the psychological quantity as described above. As mentioned above, this simple method is a relatively simple measurement method, but psychological quality can be easily determined by manual operation and by observing the instantaneous spectrum, which is normally observable with a spectrum analyzer. be. However, this simple method has the following problems associated with the sinusoidal frequency sweep operation (7).

(1) Problems related to the speed of frequency sweep When the difference in delay time between direct sound and reflected sound is small, or when the level of the reflected sound spectrum that can be observed by separating all frequencies is low, distortion occurs in the audible frequency range. When simultaneously observing a large number of reflected sound spectra by sweeping the entire frequency range, it is necessary to perform a high-speed frequency sweep. As shown in Fig. 4(a), during frequency sweep at such cylinder speed, the frequency sweep is synchronized with the sweep of the input sine waveform [7] and the frequency is at least twice the input sweep sine wave frequency and proportional to the input instantaneous frequency. If the received sound wave is sampled at the sampling frequency, as shown in Figure 4 (waveform 01 (a)), even if an ordinary spectrum analyzer is used, a swept sine wave can be obtained with the same resolution as the observed spectrum of a sine wave with a constant frequency. The spectrum of the received sound wave can be observed, but when the swept sine waveform is sampled at fixed time intervals as shown in Figure 4(b), the waveform (b) in Figure 4(0) As shown in
The observable spectrum of a swept sine wave received wave is wider than that of a line spectrum (8). Therefore, if such an expansion of the spectral width occurs, not only will the reflected sound spectrum be superimposed on the direct sound spectrum, but also the reflected sound spectra will be superimposed on each other. A problem arises in that the image quality is extremely low.

(2) Problems related to the method of displaying the instantaneous spectrum in an expanded manner As mentioned above with reference to Fig. 3, the instantaneous spectrum of the received sound wave was observed exactly at the frequencies fd□, fd, , f(18 and fd+) of the received direct wave designated in advance. However, depending only on the configuration shown in the first diagram, the instantaneous sowing frequency fd□~f of the received direct wave to be expanded and displayed as described above
d. The function of arbitrarily specifying and observing is not biased, and there is no automatic display that simultaneously displays the spectral distribution at each point in time of the instantaneous frequencies fclt to fd+ in parallel.

That is, in the configuration shown in FIG.
Furthermore, it is difficult to observe the entire spectrum distribution as shown in FIG. 3 by repeating the same manual operation sequentially over the entire measurement frequency range.

Therefore, one of the objects of the present invention is to provide a dissonance quality measurement method having a function of automatically and reliably displaying an instantaneous spectrum over a whole area at a desired sine wave frequency.

Among the above-mentioned problems with the simple method of the A-type measurement method of the present invention, (2) To solve the problem of the global expansion of the instantaneous spectrum, we need to establish the necessary temporal relationship for the global expansion of the instantaneous spectrum. An example of the configuration of the measuring device shown in FIGS. 5(a) to 5(d), which sequentially develops and displays instantaneous spectra based on their temporal relationships, will be described with reference to FIG. 6.

Microphone 2 provided at the receiving point in the configuration shown in Figure 6
2 is supplied to the Schmitt trigger circuit 101, the signal level of the direct sound component that reaches the earliest in the microphone output is set to the threshold level of the Schmitt trigger circuit 101, which is set higher than the noise level of the transmission system. , the output of the Schmitt trigger circuit 101 triggers the swept ramp wave generator 102 which generates a ramp waveform identical to the ramp waveform of the swept sine wave oscillator in the configuration shown in FIG. In addition to starting the ramp waveform generation as shown in (c), the OR circuit]04'
i, activate the analog-to-digital converter 1.05 that constitutes the spectrum analyzer. The analog-to-digital converter 105 is determined by the frequency resolution of the spectrum analyzer;
When the sampling data memory]06 is satisfied, the conversion operation is completely stopped. Then, the Fourier transformer (FFT) 107 that reads the digital data finally converts the digital data into spectral data, stores it in the spectral data cumulative adder 108, and sequentially stores the stored data. The spectrum is read out and supplied to the spectrum display 109, and the spectrum corresponding to one screen is sequentially displayed as shown in FIG. 5(d). Therefore, as shown in FIG. 5(0), The ramp waveform is shown in the same figure (
The direct wave signal with the waveform shown at is sequentially sampled by spectrum analysis processing as shown in FIG. Each comparator 108-1~
108-8, its voltage E.

The comparator 108-1, which takes the total comparison voltage, generates a trigger pulse and activates the analog-to-digital converter 105 via the OR circuit 104, and performs the above-mentioned spectrum analysis in response to the trigger pulse from the Schmitt trigger circuit 01. The same data processing as the processing is repeated, and the data shown in Fig. 5 ((
In addition to displaying the spectrum as shown in 1),
Spectrum data cumulative addition memo IJ 1. Accumulate and store 1* at different frequency positions.

The ramp waveform shown in FIG. 5(0) corresponds to the frequencies f and f of the direct wave signal shown in FIG. 5(a), respectively.
la (14 As the voltage E 8 , E 4 is reached, the comparator 1.0
2) When similar spectral data are obtained for all observed frequencies fdx to f (14) of the wetted IG, the spectrum for each frequency is mostly mutually exclusive, as shown in Figure 5(d). Displayed without superimposition, a residual spectrum of swept frequency components is obtained over the entire frequency domain.

'' When the rung waveform shown in Figure 5 (0) reaches the final voltage value, the memory contents of the spectral data accumulation memory 108 are cleared and returned to the initial state at the start of measurement. Return.

The method for measuring psychological quality by supplying the spectrum data from the spectrum analyzer 3 having such a configuration to the psychological quality measuring instrument is based on the basic configuration shown in the first diagram.
The measurement is performed for each received sound spectrum at each instantaneous frequency of the received direct wave.9 Also, when determining the psychological quality, the dissonance strength of the input is determined by using a single sine sweep wave as the test signal. 1□,=0.

Next, regarding the A-type analog-to-digital conversion synchronization method according to the present invention, which solves the frequency sweep speed problem that was a problem with the simple method P and the instantaneous spectrum expansion display method, we will discuss the observation results shown in FIG. This will be explained with reference to a specific example. This analog-to-digital conversion synchronization method simultaneously solves the problems of performing accurate observation over the entire band and displaying measurement results over the entire band.

First, regarding the observation display example when the measurement of the frequency residual spectrum is over, shown in Figure 7, series A is the fundamental wave 55 H
21 times z (i=0, 1, -・-, n-1; n-8)
shows the spectrum of the sine wave and its residual component, and
Series B is 21 times the fundamental wave 55 x %/'T Hz (
1 wealth 0,1. ..., n-]; n=8) and its residual component. Note that instead of 9 for V] in series B, a rational number having the number of significant digits necessary to sufficiently approximate v'2 may be used. In the following explanation, when dealing with irrational numbers, rational numbers with similar precision will be used.

The characteristic of the above-mentioned specific observation example is that even when the frequency of the input sinusoidal wave is being swept, the sweep completely stops +J-
However, since the input sine wave is synchronized with the sampling frequency of the analog-to-digital converter, its frequency axis
The position of - is stationary, and a constant ratio value with respect to the sampling frequency is obtained on the frequency axis, and -! In addition, the input frequency range is set so that the spectra of the residual components do not overlap and contain large aliasing errors even when using a sampling filter with a fixed cutoff frequency, and the function of the measuring instrument is It is configured.

Accordingly, the necessary conditions for realizing the above-mentioned specific observation method' (i7) are as follows.

(1) Even when the frequency of the input sine wave is being swept or after one sweep is stopped, the sampling frequency and the input surface sine wave frequency must always be synchronized at a constant ratio.

(2) In order to be able to directly read the absolute values of each of the frequencies 1 to 1 above from the constant ratio values mentioned above, 1. The relationship between the input sine wave frequency and the sampling frequency must be set to match the numerical relationship for signal processing of the spectrum analyzer.

(8) Reduce the ratio of the upper and lower ends of the sweep frequency range,
The cutoff frequency of the sampling filter is fixed at 2h.
Rigata. j-) Consider measures other than the simple method described above in order to reduce the aliasing error that occurs and to observe the entire frequency range without exception.

The above-mentioned requirement (1) can be achieved by using the circuit configuration described later, and the requirement (2) can be achieved by using the circuit configuration described later.
) can be achieved as described below.

In order to realize the measurement method of the present invention at a general level of digital technology, K has all the following practical numerical relationships in terms of performance. In other words, the upper limit of the measurement frequency range is set to 1 for all referenced frequencies, and the referenced frequency of the input sine wave is set to 0.
．． 64, and the normalized sampling frequency -i 2.56
If the ratio between the two is set to l:4, 120”B
Using a sampling filter with a slope characteristic of /octave, the frequency range where the aliasing error is '&-70 (within iB, θ~ with the above frequency scale value) is measured with an accuracy of Nuvector measurement error of 0.5 dB or less. Make it possible to measure with.

In the specific observation example shown in Figure 7, the highest frequency of the input sine wave is 19956 Hz, and the frequency 1 on the normalized frequency scale mentioned above is 10 KI (Z, based on the so-called frequency range setting of 1.0 kHz). Sampling filter and final frequency 25.6 KHz
In this case, the absolute frequency at the end of the frequency sweep can be directly read by multiplying by the same wave number of the frequency range.

Next, the above-mentioned necessary condition (8) is such that the sampling frequency at the start of the frequency sweep is Q lower than the sampling frequency at the end of the frequency sweep, and the cutoff frequency of the sampling filter is a value that matches the highest sampling frequency. This is necessary to reduce aliasing errors caused by

In the specific observation example shown in Figure 7, in the case of only the A series, first, as the basic series, a harmonic series, That is, the frequency ratio 21 (i=o
, 1 , 2 , . . . , n-1), and set their frequency sweep interval to a frequency range with a frequency ratio of 2:] to cover the entire measurement frequency range. In addition, a series B having the geometric mean frequency of the frequencies of series A is separately provided, and the sweep frequency ranges of both series A and B are made into a frequency range with a ratio of 1:v'2 to further narrow the sweep frequency range. Shi, Shi may be, both series A.

The assigned region of B is shifted so that the sweep frequency range covers the entire required observation frequency region.

As a specific example of dividing the sweep frequency region at that time, 1
The frequency range of an octave is called zlsllr in musical tones:
Using this, the frequency at the starting end of each sharing series is set to the following ratios of 48, 12, 12, 1.2, . The frequencies corresponding to the second term and subsequent terms of each series in 2 are set, and in the basic series and each division series, these are set as the starting frequencies, and the frequency series forms a monium.

The frequency ratio of the upper and lower limits in the sweep of each series is
When the number of sequences is 1.2, 8, 4, and 6.12, the ratio is 1:2, 1:, respectively.

An example of the test signal for the A-type analog-to-digital conversion synchronization method can be realized as a sum of sine waves having M wave numbers of this configuration.

(19) Next, the test signal having such a frequency configuration is not only useful in measuring physical signals as described above, but also has psychoacoustically significant properties.

The series A and B each have a harmonic frequency configuration as described above, and are the most harmonious frequency configuration in terms of psychoacoustics. Next, when series A and B are added, a frequency relationship of 1 : V'T-7, or 1 : 212, is created.If we talk about this relationship as a musical scale, it becomes 4. , for example, the relationship between C and F+ or C and Gshi, which forms one of the chords used on the upper deck -cn
b, -9 humble measure] : 2"; Even if we assume a frequency relationship of 212, we have a major third, that is, 1: 1.26, 0:
E and minor sixth i.e.]: 1.59, O:
It has a scale relationship of "G" and "A", which constitutes chords that are more commonly used in music.

Therefore, from a psychological point of view, these chords have psychoacoustic harmonic effects that are very similar to real music, unlike the impulse and band (20) band noises that are usually used to measure transmission characteristics. . The spectrum of the receiving point actually measured using the power/karu test signal is shown in Figure 7.
However, since the input cassette vector component due to the reflected wave remains, the sound exhibits a consonance window or a sense of harmony that is audibly different from the input.

Therefore, as mentioned above, the transmission system that causes the difference between the input sound quality and the output sound quality does not have the above-mentioned specifications described in Tokusho Ro No. 1983-70836, which makes it possible to compare the input sound quality and the output sound quality. Consonance (psychological quality)? It becomes possible to measure and display the acoustic psychological product η between the transmitting point and the receiving point using the X-11 II determination device. Also, as will be described later, such a signal configuration simplifies the configuration of the signal generator. It is also useful.

Therefore, in the measuring force equation of the present invention, when using a multi-frequency sweep signal consisting of the sum of multiple harmonic sequences as described above, the dissonance of the sound source is reduced to 0 when the frequency of one sine wave is swept. and becomes 1, whereas k i + O
becomes. However, when using multiple sequences, the ratio relationship of all frequencies is constant regardless of the presence or absence of sweeping.
The larger the number of sequences, the smaller the frequency change ratio due to sweeping, so the ratio of the amount of change in dissonance intensity caused by sweeping to the dissonance intensity in the case without sweeping is large, and therefore tl is 11. = Can be treated as constant, ■, 'f
- You can measure the dissonance strength of the output of the system under test by measuring it only once in advance. The psychological quality of the transmission system can be determined only by measuring the .

Next, FIG. 8 shows a configuration example of a test signal generating circuit having the frequency relationship shown in FIG. 7 in the dissonance quality measuring apparatus according to the present invention. Therefore, in order to generate the A-series test signal shown in Figure 7, the upper limit frequency of the sampling frequency sweep a is set to 25.6 KH2, and the lower limit frequency is set to 25°6/V/TKH2, that is, 18,102 KH2.
As KH2, the oscillation frequency total rational number frequency divider 210- of the sweep synchronized sampling pulse oscillator 200 whose sweep is controlled by the output ramp wave of the sweep ramp wave generator 102 from the outside.
70 by a,' OH2/25600 I (Z=0.
2750 times 1. , 202-a(n-i) in sequence, and further,
Low-pass filter 208-al, 208-a2°・
. . , 208-a(n-1) to form each harmonic frequency of the A series. On the other hand, in order to generate the B-series test signal shown in FIG.
In the case of the cA series, the frequency division ratio of the rational number frequency divider 210-b is 'k 9956''/25600 Hz frequency divider 202-b
l, 202-b2. ..., 202-b(n-1) and low-pass filter 208-bl.

2 o 3-b+a, ---, 203-b(n-x)
Form each harmonium frequency of the Ite B series for L.

A.

Each such frequency of both B series is added by an adder 204 to form a final test signal. In addition, each low-pass filter 208-al, 203-a2, --.

208-a(n-1) and 208-bl, 208-b
z, --., 208-b(n-z) To L, Ten, The level fluctuation of the fundamental wave extracted within the sweep frequency range is as small as, for example, within 0.5 dB, and the harmonics are For example, it is necessary to have sufficient attenuation performance of -40 (iB) or less.

FIG. 9 shows an example of the overall configuration of a type A intensification quality measuring device according to the present invention method using the test signal formed as described above. The illustrated configuration is almost the same as the configuration using regular time interval sampling shown in FIG. 6, but there are differences and singularities in the following points.

(]) The sampling method in the analog-to-digital converter 105 that processes the received signal is shown in Fig. 4(a).
The synchronous sampling signal in the configuration shown in FIG. 9 for processing the received signal is the same as the sweep ramp wave generator 102 of the circuit for generating the test signal shown in FIG. It is generated using a swept ramp wave generator 102'i whose performance has been calibrated. The ramp waveform is similar to the ramp waveform shown in FIG. 5(0).

When the ramp waveform voltage of the sweep ramp wave generator 102 in the configuration shown in FIG. 9 becomes the same as the ramp waveform voltage of the sweep ramp wave generator 102 in the configuration shown in FIG. Calibrate so that they are the same.

Note that in order to realize this function, as mentioned above, it is not necessary that the relationship between the absolute value of the ramp waveform voltage and the frequency be the same between the two; the key point is that the relationship between the absolute value of the ramp waveform voltage and the frequency is On the other hand, it is sufficient to realize only the relationship in which the absolute value of the frequency is reduced all times. This can also be achieved by a configuration in which the absolute value of the sampling frequency is the same for the elapsed time measurement value by the digital counter.

(2) In the spectrum analysis process shown in FIG. 5), a pause period Tp was provided, and the analog φ digital converter 105 was activated at multiple voltage points in the ramp waveform, but the configuration shown in FIG. 9 In this case, the analog-to-digital converters 105 are operated in the same manner, but the rest period Tp = 0, and the analog-to-digital converters 105 are sequentially operated freely, and instantaneous spectral data are sequentially stored in the spectral data sequential memory 110 without wasting time. For this purpose, it is provided with an AD converter continuous start signal generator 112 for generating an analog-to-digital converter start pulse, and a switch 111 for switching its output and the output of the OR circuit 104.

(8) The configuration shown in Figure 9 is the spectral data sequential memory 1.
A spectrum display 1097 reads and displays the instantaneous spectrum data stored in the frame 10 one screen at a time.The spectrum display 1097 functions as a monitor and sequentially transfers it to the psychological quality measuring device 4 to measure psychological quality.

Next, the main feature of B is that multiple sine waves whose amplitude changes over time at a constant frequency are used as test signals.
The method for measuring molds according to the present invention will be explained.

Therefore, the operating principle of the A-type measurement method of the present invention explained above utilizes the property that the various delay times of direct waves and reflected waves enable the emission time of the sound source wave to be identified by the frequency of the swept wave. Because it can be measured by calculation, and because the spectra of direct sound and reflected sound are separated and expanded on the frequency axis at the same time at the sound receiving point, and are different from the spectrum of the source sound, the auditory effect is This difference was used to measure the psychological quality of an acoustic transmission system consisting of a linear system. That is, while the type A measurement method of the present invention uses a test signal consisting of a sinusoidal wave whose amplitude is -W and whose frequency changes over time, the measurement method of the type B described above uses a test signal consisting of a sinusoidal wave whose amplitude is -W and whose frequency changes over time. The investigation measurement method is
In contrast to type A, this test uses a test signal consisting of a sine wave whose frequency is constant and whose amplitude changes over time.

In other words, a phenomenon occurs in which the own device and the reflected sound of the source sound emitted before the direct sound is emitted overlap with different frequencies at the time of sound reception, and in order to measure this phenomenon, the test signal must be Two fixed frequencies are used, and the signals of these two types of frequencies are sent out at different times, and the two different frequencies are transmitted only at the receiving point, without the frequencies appearing overlapping in the source sound. Therefore, signals of frequencies of 1 to 1 may overlap each other based on the delay time difference. In this case, at the sound receiving point, a whole spectrum different from that of the signal source is formed, so that the psychological quality between them can be counted (27). The above is the basic principle of the B-type measurement method of the present invention.

As described above, two types of sine wave signals with different fixed frequencies are generated at different times, each gated with appropriate rise and decay time characteristics, and this process is repeated to transmit continuously. An example of a mold test signal is shown in FIG. 10(a). At the receiving point, these two types of sine wave signals appear in the form of overlapping envelope waves, as shown in FIG. 10(b), due to a time delay due to the path length. For details, please refer to Figure 10 (a3
When emitting two types of sine wave signals with frequencies f0 and f2 shown in , the reflected wave of the sine wave with frequency f0 and the direct wave and reflected wave of the sine wave with frequency f2 appear simultaneously at the sound receiving point. A situation may arise.

Therefore, when there is no reflected wave and a single frequency sine wave is received, the dissonance strength at the receiving point is O if there is no background noise, but as above = two frequency sine wave as described above. If both exist simultaneously, the dissonance strength at the receiving point will generally be a positive value (28).

10,000, the dissonance strength on the signal source side is 2 frequencies f11
Since the sine wave of f2 does not exist at the same time, the frequency f□
, f2, the dissonance strength is 0 at any time of occurrence.
It is.

Therefore, as shown in Fig. 10 (C), the psychological quality of the transmission line originating from the increment of the dissonance intensity is determined by the fact that two frequency components overlap at the receiving point, similar to the dissonance intensity at the receiving point. Takes a positive value in some parts. It's a turtle! Psychological quality based on 11 parts of dissonance
The measurement can be performed by installing the obscurity measuring device described in the specification of No. 70886 at the receiving point.

The above test signal was constructed by alternately connecting two types of sine waves with different fixed frequencies, but in general, even if the spectrum distribution is complex, the frequencies between adjacent spectra are If they are arranged in series without being shared, and l1=constant, psychological quality can be measured by the same measuring method as described above.

That is, in both types A and B of the measurement method of the present invention explained so far, in both types, the case where the dissonance strength i of the transmitted signal is Ii = O and the case where it is constant By using this property, we can manipulate the dissonance strength of the receiving point. It was possible to easily measure psychological quality only by measuring .

However, when determining the general psychological quality of a multipath transmission system, it is possible to measure the psychological quality using a test signal in which the dissonance strength Ii on the input side changes over time, without complicating the system so much. A measurement method that can obtain the desired results is desired.

The Type 0 measurement method that enables such psychological quality measurement will be described next.

FIG. 11 shows a configuration example of a type 0 dissonance quality measuring device that uses a test signal whose dissonance intensity changes over time as described above. In the illustrated configuration, the psychological quality measuring device 100 receives two types of input signals, that is, an input signal of a sound source and two types of output signals generated corresponding to the input signals, that is, a It is necessary to handle human output signals from the system under test with clear relationships. In order to realize two types of signals having a clear causal relationship directly at the same time at the two input terminals of the psychological quality measuring device 400, in the configuration example shown in FIG. In addition to the test signal generator 800-1 that inputs the test signal to the psychological quality measuring device 4.Oo, there is another test signal generator that generates the exact same signal as the input signal delayed by the direct wave delay time τd. It is fully equipped with aoO-2. In addition, in order to compensate for the delay time τd of the direct wave, it is also possible to provide a delay circuit that delays the input signal by the delay time τd.
In that case, it is difficult to automatically obtain a delay time corresponding to the distortion of the delay circuit and the delay time τd, and moreover, at the same time for each frequency.

With the above configuration, the input signal from the test signal generator 800-2 on the left side of the figure to the psychological quality measuring device 400 is transmitted from the test signal generator 300-1 to the psychological quality measuring device 2 via the multipath transmission system 2. The psychological quality measuring device 400 is delayed by the delay time τd due to the input signal to the device 400, accurately corresponds to the output of the receiving point, and can compare the sound quality at the same time.
Since two human power signals can be obtained, its psychological quality can be measured.

What should be noted in particular about this C-type measurement method is that, as is clear from the above explanation, the configuration shown in Figure 1 is not suitable for the signals to be generated from the test signal generators 300-1 and 300-2. In general, even if a complex test signal, such as the sound of a real-life program, is used, the psychological quality of the test string can be measured. The same applies to each illustrated configuration. In addition,
In that case, a method for starting all of the test signal generators 300-2, or a configuration of the test signal generator 500 to compensate for the delay time τ in the configuration shown in FIG. 19 can be done as described below. Of course. Furthermore, these test signal generators 800-1 and 300-2
．． 500, there is a method of decomposing and combining an actual signal into an audible frequency component signal (1) with a constant level and a level fluctuation signal W (t), and using one of these symbols as a pattern. Of course, for both the method and the method using the actual signal itself, the allocation to the read-only memory and the delayed readout described later can be used.

However, in addition to such general signals, there is a program sound model signal as a test signal suitable for the C-type measurement method, which allows detailed physical observation and provides accurate measurement results. The entire structure of this signal generator is shown in Figure 2, and examples of the signal waveforms of each part are shown in Figure 1.8 (a), ■).
, respectively shown in (a).

Therefore, as described in Japanese Patent Publication No. 55-20289 filed by the client, the program sound model signal generation 'a is caused by a level fluctuation representing a fluctuation in the strength of the audio frequency signal V(t). Signal W (U(t) obtained by multiplying tl = v
(t)・W(t) In the configuration example shown in FIG. 12, first, a normally distributed noise signal is Instead, a pseudo-normal distribution noise signal generator 801 is provided that repeatedly generates a part of the signal cut out from the normal distribution noise signal, and an output signal having a signal waveform as shown in FIG. 13(a) is leveled. The signal is supplied to a fluctuation signal generator 308 to generate a level fluctuation signal W(1,) as shown in FIG. 13(C) as described in the above-mentioned publication.

On the other hand, FIG. 13 (
The output signal with the signal waveform shown in a) is supplied to the zero cross square wave generator 302, and as shown in FIG. An upper polarity gate circuit 308 that generates all waveform gate signals and opens when the square wave gate signal takes a positive voltage value, and a negative polarity gate circuit 308 that opens when the square wave gate signal takes a negative voltage value; (041 and When the positive polarity gate circuit 303 is opened, the audio frequency signal V,(t) from the audio frequency signal generator 805 passes through the positive polarity gate circuit 303, and the negative polarity gate circuit 303 is opened. When the circuit 304 is open, the audio signal V, (t
) are all passed through the negative polarity gate circuit 304I, and both are supplied to the adder 807, and are alternately combined and successive to each other to form a series of audio frequency signals v(t)=v1(t)+v
2(t). Audio frequency 4M component V (t) of the desired program sound model signal
is supplied to the multiplier 309, and the level fluctuation signal generator 3
By multiplying the level fluctuation signal W(t) supplied from 08, the program sound model signal U(t) (= V(
a·W(t)).

The program sound model No. 1 generator with the above-mentioned configuration uses two types of audio frequency signals V, (t) and vQ (t), which have different frequencies, over time. The audio frequency signal V(t) is constructed by combining the above-mentioned two-building audio frequency signals Vl(tl and V 2 (t)), which are used in the above-mentioned type A measurement method of the present invention. signals of series A and series B forming a harmonium, or sine wave signals of frequencies f0 and fg used in the above-mentioned fcB type measurement method of the present invention; It is preferable to use a signal, that is, a signal that does not share a frequency and has a general dissonance strength.

Another feature of the partially C-type measurement method of the present invention is the use of the pseudo-normal distribution noise signal generator 301 as described above. In other words, the important point in the C-type measurement method of the present invention is to compensate for the delay time τd of the direct wave, so the arrival time of the direct wave must be known as accurately as possible, and the input test signal can be determined from the arrival time. The key point is to generate a level fluctuation signal w(t)' that is exactly the same as that used for the generation.
As shown in the figure, each of the following mountings (1
) to (8).

(1) In the configuration shown in FIG. 14, data stored in the read-only memory 350 can be read out from the earliest address by a start pulse.

(2) As the data value of the earliest address mentioned above, the first
As shown in the amplitude values at both ends shown in Figure 3(a), the Schmitt circuit 10 in the configuration example of the psychological quality measuring device using instantaneous spectrum measurement shown in Figure 9] should have a value large enough to operate accurately. , it shall be possible to accurately determine the time of synchronization start by receiving 100 waves.

(3) If the normally distributed noise signals for generating the level fluctuation signal have exactly the same waveform, the level fluctuation signal W(t) formed from the normally distributed noise signal with the same waveform will have the same shape. A part of the true normally distributed noise signal is cut out and stored in the analog read-only memory.
All digital conversion data is repeatedly extracted during measurement,
Used in place of a true normally distributed noise signal. If the synchronization start time is accurate through such a procedure, a level fluctuation signal W(tl) having exactly the same waveform as the level fluctuation signal W (t) on the sound source side is obtained after that time.

Next, sample data k W of the true normal distribution noise signal
Using the output clock pulse of the clock pulse generator 856, the address specification data generated from the address generator 854 is generated from the data stored in the read-only memory, which is stored in the register 351, and the specified data is digitally stored. The signal is converted into an analog signal by an analog-to-digital converter 352, and the above-mentioned true normal distribution noise signal is reproduced by a low-pass filter 858.

In the above description, the address counter 355 is
After the start, the read data address is counted, and after all the data a on the read-only memory are successively output, address generation is controlled so that the data can be read again from the start address.

Note that when reading data, the start and stop of the address counter's operation is controlled by an external start/stop signal.

Next, regarding other performance requirements for the pseudo-normally distributed noise signal generator, the normally distributed noise signal is normally 0.1 to It is assumed that the spectral distribution is obtained by adding a bandpass filter having a resonant center frequency in the range of 10 Hz. Therefore, the sampling frequency used in the pseudo-normal distribution noise signal generator needs to be at least about 50 H2. In order for the waveform near T-10 and IH2 to be random, the length of the signal waveform to be stored as digital data in the read-only memory must have a period of 100 seconds or more,
Therefore, when performing 8-bit quantization, the memory capacity needs to be approximately 8×50×]00=40,000 pits and 5000 bytes.

Therefore, the C-type psychological quality measuring device W according to the present invention is based on the method described in the above-mentioned Japanese Patent Application No. 55-70836.
The structure is similar to that of the "Strong Dissonance Measuring Device", but the structure of the main part is shown in FIG. The main parts of the illustrated psychological quality measuring device are input signals of a multipath transmission system, such as speaker input signals from a test signal generator, and
A dissonance strength measuring device 4]0-a and a dissonance strength measuring device 4]'O-b which input all output signals of a multipath transmission system, for example, microphone output signals of a psychological quality measuring device, respectively, and a dissonance strength measuring device 410 for these -a, b are spectrum analyzers 401-a, b, minimum devices 408-a, b, respectively;
, two-frequency dissonance strength calculation devices 404-a, b, and dissonance strength averaging devices 405-a, b.

The transmission system input dissonance strength Ii and the transmission system output dissonance strength ■o obtained from the dissonance strength measuring devices 410-a and 410-b are supplied to the total subtractor 406 to calculate the difference therebetween,
Furthermore, the psychological quality (Ii) β (β ki 0.25) is determined by the dissonance strength-dissonance degree converter 407 and displayed on the display 412.

The key point in measuring psychological quality using a device with such a configuration is to exclude the delay time τd of the direct wave as the two-manpower signal of the psychological quantity 5N measuring device, just as humans compare the two-manpower by hearing. As mentioned above, the method consists in using two human-powered signals that match causal correspondence at the same time.

The characteristic feature of the type 0 psychological quality measurement method of the present invention is that the test signal is a dynamic signal whose amplitude, frequency, and phase vary over time, such as an AM signal.
The point is that an FM signal or the like can be used, and the psychological quality of measurement obtained by the type 0 measuring device is the instantaneous psychological quality regarding such a dynamic signal. Therefore, in order to know the psychological quality in all operating states of the system to be measured, it is necessary to find the statistical average value of such instantaneous quality. however,
The correct psychological quality obtained by the above-described psychological quality measuring device shown in FIG.

Therefore, in order to find the average quality, for the dissonance strength that has algebraic additivity as an intermediate quantity, find the instantaneous dissonance strength increment brought about by the thread to be measured, find the average value, and then convert it to psychological quality. Therefore, it is necessary to find the average psychological quality. FIG. 16 shows an example of the configuration of a psychological quality measuring apparatus according to the present invention, which is capable of measuring the dynamic average psychological quality of a system to be measured through such a procedure. In the illustrated configuration, the dissonant strength measuring devices 4.10-a and 4]0-b and the dissonant strength averaging device are used for the same transmission system input signal and transmission system output signal as described above with reference to FIG. 15, respectively. 411-a and 4]1-b are provided to calculate the average dissonance strength Ii and Ii of the input signal and the output signal. The increment of the input-output dissonance strength obtained by the subtractor 406 is guided to the dissonance strength-dissonance degree converter 407 to obtain the average psychological quality, and the difference is calculated separately from the actually measured reference data by 1-. , the average de-excitation of the input signal and the output signal 1
11 Strength work, and work. are guided to the dissonance strength-dissonance degree converters 407-a and 407-b, respectively, to obtain the average dissonance degrees of the input and output of the measured port system, respectively, and calculate the average dissonance degrees as psychological quantities. Displayed on all displays 412.

Therefore, in the psychological quality measuring device having the configuration shown in FIG. 15 and FIG.
0-a and 4]0-b, respectively, to improve data processing speed, these dissonance intensity measuring instruments each include a spectrum analyzer, making them significantly more expensive. Therefore, it is preferable that the immobility f11 strength on the input side and the output side can be measured by one dissonance strength measuring device. The basic configuration of the C-type psychological quality measuring device according to the present invention, which is a simplified version of the Dissonance 9 MII measuring device in preparation for such use, is shown in FIG. 17, and the time chart of the operation of each part thereof is shown in FIG. 18. A detailed configuration example is shown in FIG. 19.

In the basic configuration shown in FIG. 17, the test signal 300 having the shape shown in FIG.
is supplied to the input side speaker 21 of the transmission system under test 42 through a delay circuit 420 having
A received sound signal with a signal waveform shown in FIG. 18(b) obtained from the output side microphone 22 after =τ+τd, and a test signal with a signal waveform shown in FIG. 18(a) from the sending point of the transmitted signal 300. are alternately switched by a switch 422 and supplied to a single psychological quality measuring device 4. In the detailed configuration example shown in FIG. 19 that embodies this basic configuration,
The test signal delayed by the aforementioned delay time τ from the delayed test signal generator 500 (48), which is driven by the reset start signal, is supplied to the transmission system under test 508 in the same manner as described above, and the obtained signal is output to the output side of the test signal. The above-mentioned received sound wave signal, reset start signal, and timing information from the psychological quality measuring device 4 are supplied to the counter 501 to perform time measurement as shown in FIG. 18 ((1)). The resulting control signal controls the switching of the switch 502 that operates in the same manner as described above, and the psychological quality measuring device 4 measures the dissonance strength of both the input and output in the manner shown in FIG. 18(C). Do the following.

Therefore, if the delay time τd is short, even if the spectrum analyzer constituting the psychological quality measuring device 4 operates at high speed, it will still complete the spectrum processing for one screen within the direct wave delay time 76. However, a case may arise in which it is not possible to alternately measure the dissonance strength on the input side and the output side of the transmission system under test.

Therefore, in order to be able to complete the required measurement by sequentially switching the input side and the output side, it is necessary to use a direct C44) wave delay time τdJ: sufficiently long, which is compatible with the signal processing time of the spectrum analyzer. It is necessary to add a certain delay time TS to the human signal. The additional delay time τ necessary for this purpose is provided by the delay circuit 420.

1. First, when switching the switch 422 to the input side and starting measurement of the test signal, the analog-to-digital conversion of the test key signal, the time TM required for storage, and the FF
If the total spectral data storage time Tg is shorter than the total delay time Ts = τ 0 τd, then if the analyzed spectral data is stored, i will automatically switch 422.
If you switch to the full output side, will the output received sound signal of the transmission system under test 421 be of psychological quality? It reaches the H11 constant device 4, and at that point the measurement of the delay time Ts also ends. This delay time T
s is set for the time measurement end time for the next time measurement. In addition, in the case where the output received sound signal arrives at the point in the spectrum data, the additional delay time τ (total increase) is added so that the output received sound signal reaches , so that the time relationship shown in Figure J8 described above can be obtained.

Next, similar spectrum analysis and data storage are performed on the output received sound signal of the transmission system to be measured, and then immediately switching is made to measurement of the input test signal, and spectrum analysis and data storage are performed again. Furthermore, spectrum analysis and data storage for the next output received sound signal are performed by the counter 50.
The measurement starts after the time measurement value 1 reaches the end time T8 of the previous measurement.

According to the data processing procedure described above, the input test signal analysis spectrum ψ data and the output received sound signal analysis spectrum data have a cause-effect correspondence relationship that is synchronous in measurement, and the desired result is achieved. Measurements can be taken. Moreover, if the additional delay time τ that is most similar to the shortest delay time Ts value such that the total storage time Tg≦total delay time Ts is selected,
If the time required for spectral analysis and data processing is a bottleneck, spectral data for the input test signal and the output received sonic signal can be obtained from multiple sources.

J: The delay circuits 4 and 20 that give the additional delay time τ of A to the test signal are constructed by preparing two pseudo-normal distribution noise signal generators as described above with reference to FIG. The data read address from the constituent read-only memory is shifted by an offset address (=T/sampling period) from the address of the test signal itself, and the data is output with a delay of time τ to achieve the desired function. be able to.

Note that the glue-only memories for the two pseudo-normal distribution noise signal generators should be equipped with a data read function that can read all data from each other according to the two types of read addresses while holding the above-mentioned offset address. For example, a single read-only memory can be used for two pseudo-normal distribution noise signal generators.

As described above, when two systems of pseudo-normally distributed noise signals having a time difference τ were obtained, these pseudo-normally distributed noise signals were used as the outputs of the pseudo-normally distributed noise signal generator shown in FIG. 12 and were supplied respectively. Two pre-program model signal generators are used to generate the input test signal and speaker input signal of the psychological quality measuring device having the basic configuration shown in FIG. 17, or the test signal/delayed test signal generator 50 shown in FIG. 19.
Two outputs of 0 can be formed.

The effects of the present invention that are clear from the explanations below are as follows, enumerated and briefly described by appropriate items.

(A, l Regarding the A-type measurement method of the present invention (1) The simple method has a simple configuration of the measuring device and can be carried out manually using an existing measuring device,
It is extremely convenient to use it experimentally at the beginning to get an overview of the psychological quality of the system being measured.

(2) Among the synchronous methods, even in the synchronous method, which uses asynchronous input sampling pulses and controls only when the measurement frequency is specified, the developed instantaneous spectra can be displayed in parallel in a snap manner, making it extremely suitable for routine repeated measurements. It is.

In addition, in practical terms, it is considered to have the greatest utility value because clear measurement results can be obtained despite the relatively simple configuration of the measuring device.

(8) The A-D conversion synchronization method is suitable for full-scale precision acoustic measurements, and is especially suitable for estimating the propagation path of reflected sound by improving frequency resolution, or for changing room acoustic characteristics, using reflectors and sound absorbers. It is suitable for understanding the effects of countermeasures depending on the material physically, psychologically, accurately, accurately, and quickly.

(B) Regarding the measurement method of the present invention for type B In the measurement method of the present invention for type A, the residual component on the frequency axis is measured acoustically and psychoacoustically, whereas
In the measurement method of the present invention, the residual components that appear on the time axis are weighted (2), and when a sine wave signal of an identified frequency is used as the test signal, the spectrum is assumed to be in a relatively stationary state. Therefore, the psychoacoustic effect is simple, and highly accurate psychological quality measurement results can be expected.

((About the measurement method of the present invention for Type 330) The measurement force formula of the present invention for Types A and B improved physical accuracy, but on the other hand, the psychoacoustic quantities handled were relatively simple;
Although the measurement method of the present invention has slightly increased complexity due to changes in the configuration of the measuring device, both physical and psychological phenomena have become more complex, making it possible to more practically measure physical and psychological qualities. can be done.

(D) Scope of application of the method to the present invention The subject of psychological quality measurement using the method of the present invention is typically a linear multipath transmission system, but other linear systems with time delays, such as tape recorders, can also be used. Any of these can be applied to the measurement method of the present invention, and communication systems having long-distance propagation paths such as satellite broadcasting can also be applied.

(11 Regarding time measurement, for example, when using a swept sine wave signal whose frequency changes linearly with time as a test signal, the frequency difference between the signals is proportional to the delay time, so the developed spectral distribution The delay time of each frequency component can be measured from the frequency difference.

(2) Regarding spatial acoustic characteristic measurement, by multiplying the delay time measured in the previous section by the speed of sound, the path length of the frequency component can be measured. Knowing the length of the deflected path limits the number of estimated reflection points.

Therefore, by using a directional microphone that has only one direction of zero sensitivity and knowing the direction of sensitivity where the output at a certain frequency is zero, it is possible to determine not only the path length mentioned above but also the direction of the reflected wave. can.

Furthermore, by measuring the psychological quality in the above-mentioned state, it becomes clear how the reflected wave component contributes to sound quality inhibition. In addition, when measuring the psychological quality in an indoor sound field with reflected waves using speakers and microphones with various directional characteristics, the contribution of these speakers and microphones to the psychological quality can be clarified with respect to the reference system. be able to. Note that the same applies to the case of dummy head sound reception.

(8) Regarding crosstalk psychological quality, for example,
In the losstalk system as shown in Figure 0, two different
Determining the psychological quality between two output terminals using human signals;
Alternatively, the C-type measurement method of the present invention is extremely useful for determining the degree to which the output sound quality obtained by linearly adding two human input signals differs from the input sound quality. However, if the spectra do not overlap with each other and crosstalk is added to the two signals S0 and 82, which are connected over time, the components with the same frequency as the input signal will not cause a large change in timbre, but will differ from the input signal. Frequency components make a big difference to the timbre. Such psychological quality can be measured using the O-type measuring devices shown in Figures 1 and 17.

(4) With regard to magnetic tape interlayer transfer, when signal components other than the original signal are transferred to the magnetic tape, the sound quality deteriorates, but changes in sound quality due to such transfer can be measured quantitatively in terms of physical quantity and psychological quality using the type 0 side treasure method. Measurements can be taken. For example, as an input signal, the frequencies are f□, fg
When using a type B test signal, which is a sine wave of Since simultaneous phenomena occur, the psychological quality corresponding to such transcription has the signal relationship shown in FIG. 21, and can be quantitatively measured by the measuring device having the configuration shown in FIGS. 11 and 17. can.

(5) Regarding the measurement of the amount of psychological contribution of distributed sound sources, when the sound emitted from n sound sources is received, the total dissonance @ degree It (nl, the sound emitted from the n+1th sound source is If we add the total dissonance strength factor t(n+,),
The amount of psychological influence due to receiving the sound from the n+1th sound source is ΔI(n+x)=It(n+i) It
(n) The increment in dissonance intensity determined by the formula can be converted into psychological quality and measured. This increment ΔI
(n+1) l'I n+ This is different from the strong received dissonance FI (n+1) of the first sound source.

(6) Distortion measurement (a) Regarding the measurement of AM-PM conversion distortion, in a linear system where the peaks and troughs of the amplitude frequency characteristics change rapidly, such as the indoor transmission system, the B-type measurement method of the present invention is used. When an AM wave signal is used as one of the test signals that can be used,
The sidebands of the waves are no longer equal and a PM wave component is generated. Therefore, if an AM wave modulated at an appropriate modulation frequency, for example 50 Hz, is used, an output with a sound quality different from that of the input will appear, and if the input modulation frequency and modulation degree are fixed, the input dissonance strength will be constant. Therefore, the dissonance level psychological quality can be measured only by measuring the dissonance strength of the output.

(b) Regarding the measurement of FM-AM conversion distortion, when the FM signals, which are one of the test signals that can be used in the above-mentioned type 0 measurement method, are added together via a multipath transmission system, the AM produces signal components.

Therefore, psychological quality can be measured in the same manner as described above for item (a).

(c) When a general signal is used, psychological quality can be measured in exactly the same way for a system to be measured in which nonlinear distortion occurs in addition to linear distortion as described above.

In addition, in the A-type A-D conversion synchronization method, when using the I-monium sequence as the basic sequence of the test signal,
To avoid the occurrence of a case where the amount of linear distortion and nonlinear distortion cannot be separated because a second harmonic is generated as nonlinear distortion and the second harmonic overlaps the input second harmonic in the received sound wave. There is a measurement method in which a frequency corresponding to the second harmonic is thinned out in the multi-sequence configuration described above.

That's naive.Another method is to use the rational number ratio splitter described above to slightly smooth each harmonic frequency of the input monium series to avoid harmonic nonlinear distortion from superimposing on the input frequency. A method can be used.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an example of the basic configuration of a type A dissonance quality measuring device 1 according to the present invention, and FIGS. Waveform diagrams each showing an example of the waveform of a sound wave, Figure 8 is a diagram showing an example of the power spectrum distribution of the received sound wave, and Figures 4 (a), (b) and (C) are the same test results. Diagrams showing examples of signal waveform sampling and examples of differences in measured spectra due to different sampling methods, respectively. Figure 6 is a block diagram showing an example of the configuration of the instantaneous spectrum expansion display type measuring device, and Figure 7 is an example of the instantaneous spectrum expansion using the A-D conversion synchronization method. 8 is a block diagram showing an example of the configuration of the sweep test signal generator, and FIG. 9 is a block diagram showing an example of the configuration of the measuring device based on the A-D conversion synchronization method. Figure 10 (a) to (C
) are waveform diagrams sequentially illustrating the measurement principle of the B-type dissonance-concealed quality measurement system of the present invention, FIG. Figure 12 is a block diagram showing an example of the configuration of the program sound model signal generator, Figures 13 (a) to (C) are waveform diagrams showing the signal waveforms of each part, and Figure 14 is the pseudo-normal diagram. FIG. 15 is a block diagram showing a configuration example of a distributed noise signal generator, FIG. 15 is a block diagram showing a configuration example of the psychological quality measuring device, and FIG. 16 is a block diagram showing a configuration example of the average psychological quality measuring device. Figure 17 is a block diagram showing the principle of construction of the simple psychological quality measurement device, Figures 18 (a) to (dl are time charts showing the operation of each part, respectively), and Figure 19 is a block diagram showing the principle of construction of the simplified psychological quality measuring device. Similarly, FIG. 20 is a block diagram showing an example of a simplified configuration of the room psychoacoustic quality measuring device, FIG. 20 is a block diagram showing an example of a circuit configuration in which the measurement method of the present invention is applied to crosstalk measurement, and FIG. It is a block diagram showing an example of a circuit configuration in which the measurement method of the present invention is applied to measurement of a tape transfer system. 1... Frequency sweep oscillator, 2... Multipath transmission system,
8... Spectrum analyzer, 4... Psychological quality measuring device,
21... Transmission point (speaker), 22... Sound receiving point (microphone),] O]...] Schmitt trigger circuit 102... Sweep ramp wave generator, 103... Comparator , 104... OR circuit, 105... Analog-to-digital converter, 106... Sampling-data memory, 107... FFT spectrum converter,
108... Spectral data cumulative addition 5.109.
... Spectrum display device, 110 ... Spectrum data 111st memory, 111 ... AD converter start signal switch, 112 ... AD converter start signal generator, 200 ... Sweep synchronous sampling pulse generator, 202... Frequency divider, 208... Low pass filter, 204... Adder, 210... Rational number divider, 800... Test signal generator, 301... Pseudo-normal distribution noise signal Generator, 802... Zero cross square wave generator, 803... Positive polarity gate circuit, 304...
・Negative polarity gate circuit, 305, 306... Audio frequency signal generator, 807... Adder, 308... Level fluctuation No. 6 generator, 309... Multiplier, 350... Read only Memory, 351...Read register, 85
2... DhK converter, 35 B... Low pass filter, 354... Address generator, 855... Address e counter, 356... Clock pulse generator, 40
0...2 Manual psychological quality measuring instrument, 40]... Spectrum analyzer, 4.02... Cut-off circuit below the minimum audible level, 403... Masking device, 404...2
Frequency dissonance strength calculation device, 405...Summing device, 40
6...Dissonance strength measuring device, 407...Dissonance strength-
Dissonance degree converter, 410... dissonance strength measuring device, 41
DESCRIPTION OF SYMBOLS 1... Dissonance intensity averaging device, 420... τ time delay circuit, 421... Transmission system under test, 422... Changeover switch, 500... Test signal/delay test signal generator,
501... Counter, 502... Selector switch, 50
3...Transmission system to be measured, 601.602...Crosstalk system, 603...Transfer system. Patent Applicant Japan Broadcasting Corporation Figure 13 Figure 14 Figure 15 Figure 16 ~ 4 (, J ”CI

Claims (1)

[Claims] L A test signal consisting of a plurality of frequency signals arranged over time is applied to a multipath transmission system having different delay times for each path, and the shortest delay time is applied to the multipath transmission system at the receiving point of the multipath transmission system. Generate new overlap of multiple frequency components between the time-delayed received signal and other delayed received signals, and measure based on the shortest delay time of the multipath transmission system detected from the received signal delayed by the minimum time. The timing is set, and at the measurement timing, the reception at the reception point regarding the overlap of the plurality of frequencies that has occurred is determined.
The psychological quality a' of the multipath transmission system is measured by measuring the total dissonance of the signal.ff: Dissonance quality measurement method of the multipath transmission system characterized by: In the measurement method described in claim 1,
A dissonance quality measurement method for a multipath transmission system in which the test signal is composed of a signal whose frequency changes over time within a required frequency range. & A dissonance-required quality measurement method for a multipath transmission system, in which the test signal is constructed by time-divisionally arranging a plurality of frequency signals within a required frequency range in the measurement method according to claim 1.