JP2011085691A - Frequency characteristic measurement device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To highly precisely measure the frequency characteristics of an object to be measured. <P>SOLUTION: A frequency characteristic measurement device 100 measures frequency characteristics of an object 200 to be measured. The object 200 is, for example, an equalizer or an acoustic space. A signal creation section 12 creates a reference signal SREF in which an amplitude spectrum QA linearly continues along a frequency axis. An output section 14 outputs the reference signal SREF to the object 200. A detection section 22 detects a measurement signal SDET by adding the frequency characteristic of the object 200 to the reference signal SREF. A signal analysis section 24 analyzes frequency characteristics F of the object 200 by analysis of the measurement signal SDET. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a technique for measuring frequency characteristics of an acoustic device or an acoustic space.

  Various techniques for measuring frequency characteristics of objects to be measured such as acoustic devices and acoustic spaces have been proposed. For example, Patent Document 1 discloses a technique for measuring a frequency characteristic of an acoustic space by reproducing a reference signal such as pink noise or an impulse waveform in the acoustic space and analyzing a measurement signal collected via the acoustic space. Is disclosed.

JP-A 63-234699

  However, the amplitude value of the amplitude spectrum of the reference signal fluctuates randomly along the frequency axis, and irregular fluctuations due to fluctuations in the amplitude value of the reference signal are further superimposed on the frequency characteristics (spectrum) of the measurement signal. . Therefore, there is a problem that it is difficult to measure the frequency characteristics with high accuracy. In view of the above circumstances, an object of the present invention is to measure the frequency characteristics of a measurement object with high accuracy.

  In order to solve the above problems, the frequency characteristic measurement apparatus of the present invention includes a signal generation unit that generates a reference signal having an amplitude spectrum that is continuous along the frequency axis, an output unit that outputs the reference signal to a measurement target, A detection unit that detects a measurement signal in which a frequency characteristic of a measurement target is added to a reference signal, and a signal analysis unit that analyzes the measurement signal are provided. In the above configuration, since the amplitude spectrum of the reference signal supplied to the measurement target is continuous along the frequency axis, for example, white noise or pink noise in which the amplitude value of the amplitude spectrum varies along the frequency axis is measured. Compared to the case of supplying to the frequency, it is possible to measure a frequency characteristic close to the actual characteristic of the measurement target.

  Amplitude spectrum “continuous” along the frequency axis means that the amplitude spectrum continues smoothly along the frequency axis (regardless of whether it is linear or curved), and more specifically Means that there is no local fluctuation of the amplitude value (a point where the amplitude value fluctuates discontinuously) in the amplitude spectrum. Therefore, normal white noise and pink noise in which the amplitude value of the amplitude spectrum actually fluctuates discontinuously is excluded from the reference signal in the present invention.

  In the form in which the signal generation unit generates a reference signal in which a plurality of unit waveforms having amplitude spectra continuous along the frequency axis are arranged on the time axis, the measurement signal generated by the detection unit is a periodic signal. Therefore, a configuration in which the signal analysis means includes first synchronization addition means for synchronously adding the measurement signals (specifically, adding a plurality of synchronization intervals corresponding to integer multiples of the cycle of the measurement signal to each other) is also preferable. . According to the above configuration, since the noise of the measurement signal generated by the detection unit is reduced by the synchronous addition by the first synchronous addition unit, the frequency characteristic of the measurement target is improved even in an environment where noise occurs in the measurement signal. There is an advantage that it can be measured accurately.

  In a specific example of the embodiment including the first synchronous adder, the signal analyzing means includes a second synchronous adder for synchronously adding the reference signals, a measurement signal after the synchronous addition by the first synchronous adder, and the second synchronous adder. Characteristic analysis means for analyzing the frequency characteristic of the measurement object using the reference signal after the synchronous addition by. In the above aspect, since the measurement signal and the reference signal are used for the analysis by the characteristic analysis means, there is an advantage that, for example, a highly accurate frequency characteristic with reduced influence of the reference signal can be specified. Further, since the noise of the reference signal is reduced by the synchronous addition by the second synchronous addition means, a special effect is realized that the frequency characteristic of the measurement target can be measured with high accuracy even in an environment where noise is generated in the reference signal. .

  A typical example of the reference signal is a signal whose amplitude spectrum continues linearly along the frequency axis. Specifically, the signal generation means, for example, a reference signal in which the amplitude value at each frequency of the amplitude spectrum becomes a constant value or a reference signal in which the amplitude value of the amplitude spectrum changes at a predetermined change rate with respect to the frequency. Generate. Regardless of which reference signal is applied, it is possible to measure the frequency characteristics of the measurement object with high accuracy.

  The frequency characteristic measuring apparatus according to each aspect described above is realized by hardware (electronic circuit) such as a DSP (Digital Signal Processor) dedicated to the measurement of frequency characteristics, and a general-purpose such as a CPU (Central Processing Unit). This is also realized by cooperation between the arithmetic processing unit and the program. The program according to the present invention adds a signal generation process for generating a reference signal whose amplitude spectrum is continuous along the frequency axis, an output process for outputting the reference signal to the measurement target, and a frequency characteristic of the measurement target to the reference signal. A computer executes detection processing for detecting a measurement signal and signal analysis processing for analyzing the measurement signal. According to the above program, the same operation and effect as the frequency characteristic measuring apparatus according to the present invention are realized. The program of the present invention is provided to a user in a form stored in a computer-readable recording medium and installed in the computer, or provided from a server device in a form of distribution via a communication network and installed in the computer. Is done.

It is a block diagram of the frequency characteristic measuring device concerning a 1st embodiment. It is a conceptual diagram for demonstrating the production | generation of a reference signal. It is a block diagram of a signal analysis part. It is a conceptual diagram for demonstrating the effect of 1st Embodiment. It is a conceptual diagram for demonstrating the production | generation of the reference signal of 2nd Embodiment. It is a conceptual diagram for demonstrating the effect of 2nd Embodiment. It is a block diagram of the signal analysis part in 3rd Embodiment. It is a conceptual diagram for demonstrating the effect of 3rd Embodiment.

<A: First Embodiment>
FIG. 1 is a block diagram of a frequency characteristic measuring apparatus 100 according to the first embodiment of the present invention. The frequency characteristic measuring apparatus 100 is a measuring instrument that measures the frequency characteristics of the measurement target 200. The measurement target 200 is an element (acoustic device or acoustic space) that adds a specific frequency characteristic to voice or a signal. In the first embodiment, an equalizer that outputs a signal supplied from the outside with a predetermined frequency characteristic added is assumed as the measurement target 200.

  As shown in FIG. 1, the frequency characteristic measuring apparatus 100 includes a signal generation unit 12, an output unit 14, a detection unit 22, a signal analysis unit 24, and a result output unit 26. Each element of the frequency characteristic measuring apparatus 100 is realized by, for example, a general-purpose arithmetic processing unit (CPU) executing a program. A configuration in which each element of the frequency characteristic measuring apparatus 100 is realized by a dedicated electronic circuit (DSP), or a structure in which each element of the frequency characteristic measuring apparatus 100 is distributed in separate integrated circuits may be employed.

  The signal generator 12 generates a reference signal SREF. The reference signal SREF is a time domain periodic signal in which a plurality of unit waveforms W are arranged along the time axis. Part (A) in FIG. 2 is the amplitude spectrum QA of each unit waveform W, and part (B) in FIG. 2 is the phase spectrum QP of each unit waveform W. As shown in part (A) of FIG. 2, the amplitude spectrum QA of each unit waveform W of the reference signal SREF continues linearly along the frequency axis. Specifically, the amplitude value of the amplitude spectrum QA is maintained at a constant value over the entire band. On the other hand, as shown in part (B) of FIG. 2, the phase value of the phase spectrum QP fluctuates irregularly along the frequency axis.

  The signal generation unit 12 in FIG. 1 generates the unit waveform W of the portion (C) in FIG. 2 by performing inverse Fourier transform on the amplitude spectrum QA and the phase spectrum QP, and mutually converts the plurality of unit waveforms W. To generate a reference signal SREF. Therefore, the unit waveform W corresponds to one cycle of the reference signal SREF. A method of generating the reference signal SREF by connecting the unit waveforms W so that the unit waveforms W that follow each other partially overlap on the time axis is also suitable.

  The output unit 14 in FIG. 1 outputs the reference signal SREF generated by the signal generation unit 12 to the measurement target 200. On the other hand, the detection unit 22 detects the reference signal SREF passed through the measurement object 200 as the measurement signal SDET. That is, the measurement signal SDET is a periodic signal obtained by adding the frequency characteristic of the measurement target 200 to the reference signal SREF.

  The signal analysis unit 24 specifies the frequency characteristic F of the measurement target 200 by analyzing the measurement signal SDET detected by the detection unit 22. The frequency characteristic F is typically a power spectrum. Therefore, an element that converts the measurement signal SDET in the time domain into the frequency characteristic F in the frequency domain by, for example, Fourier transform can be adopted as the signal analysis unit 24. The result output unit 26 outputs the frequency characteristic F measured by the signal analysis unit 24. For example, a display device that displays the frequency characteristic F and a printing device that prints the frequency characteristic F are employed as the result output unit 26.

  FIG. 3 is a block diagram of the signal analysis unit 24. As shown in FIG. 3, the signal analysis unit 24 includes a frequency conversion unit 32 and an averaging unit 34. The frequency converter 32 sequentially generates a spectrum (power spectrum) P of the measurement signal SDET for each frame on the time axis. A known technique (for example, fast Fourier transform) is arbitrarily adopted for generation of the spectrum P. The averaging unit 34 generates a frequency characteristic F by averaging (smoothing) the plurality of spectra P generated by the frequency conversion unit 32.

  Part (A) of FIG. 4 is the spectrum P of the measurement signal SDET (before processing by the averaging unit 34). In FIG. 4B, the spectrum P_WT of the measurement signal when normal white noise is supplied to the measurement object 200 is also shown for comparison with the spectrum P when the reference signal SREF is used. ing. As shown in part (D) of FIG. 2, white noise is non-periodic noise in which the amplitude value of the amplitude spectrum QA_WT fluctuates irregularly within a predetermined range depending on the frequency. In FIG. 4, an equalizer (peaking equalizer) having a frequency characteristic that suppresses a 1 kHz component by 10 dB is assumed as the measurement target 200.

  As shown in part (B) of FIG. 4, the spectrum P_WT when white noise is supplied to the measuring object 200 includes irregularities caused by fluctuations in the amplitude value of the amplitude spectrum QA_WT (part (D) of FIG. 2). Changes occur. On the other hand, since the amplitude value is maintained at a constant value over a wide range in the amplitude spectrum QA of each unit waveform W constituting the reference signal SREF, as shown in part (A) of FIG. 4, it is caused by the amplitude value of the amplitude spectrum QA. Such variation hardly occurs in the spectrum P of the measurement signal SDET. That is, compared with the case where white noise is supplied to the measuring object 200, there is an advantage that the spectrum P (and also the frequency characteristic F) close to the actual frequency characteristic of the measuring object 200 can be specified.

  Even when normal white noise is supplied to the measurement object 200, if a sufficient number of spectra P_WT averaged by the averaging unit 34 is ensured, measurement errors caused by fluctuations in the amplitude value of the amplitude spectrum QA_WT can be reduced. It is possible to reduce. However, since it is necessary to average a large number of spectra P_WT, there is a problem that the processing load by the averaging unit 34 is large (measurement takes a long time). On the other hand, according to the first embodiment using the reference signal SREF, since the fluctuation caused by the amplitude spectrum QA of the reference signal SREF hardly occurs in the spectrum P of the measurement signal SDET, the spectrum P to be averaged by the averaging unit 34 is obtained. It is possible to reduce the load on the average unit 34 (ideally, the average by the average unit 34 is omitted).

<B: Second Embodiment>
Next, a second embodiment of the present invention will be described. In addition, about the element which an effect | action and function are equivalent to 1st Embodiment in each following example, each detailed description is abbreviate | omitted suitably using the same code | symbol as the above.

  Part (A) of FIG. 5 is an amplitude spectrum QA of each unit waveform W constituting the reference signal SREF of the second embodiment. As shown in part (A) of FIG. 5, the amplitude spectrum QA of the unit waveform W has a frequency such that the amplitude value changes at a predetermined change rate with respect to the frequency (for example, decreases at a slope of −3 dB / Oct). It continues in a straight line along the axis. The phase spectrum QP shown in part (B) of FIG. 5 is the same as that of the first embodiment (part (B) of FIG. 2). The signal generator 12 generates the reference signal SREF by periodically arranging unit waveforms W (part (C) in FIG. 5) generated by inverse Fourier transform on the amplitude spectrum QA and the phase spectrum QP.

  Part (A) of FIG. 6 is a spectrum P of the measurement signal SDET when the reference signal SREF of the second embodiment is supplied to the measurement object 200 having the same characteristics as the first embodiment, and the part (B) of FIG. ) Is a spectrum P_PK of a measurement signal when normal pink noise is supplied to the measurement object 200. As shown in part (D) of FIG. 5, pink noise is aperiodic noise that fluctuates irregularly so that the amplitude value of the amplitude spectrum QA_PK is roughly inversely proportional to the frequency. As understood from the comparison between the part (A) and the part (B) in FIG. 6, the same effect as that of the first embodiment is realized in the second embodiment. That is, compared with the case where pink noise is supplied to the measuring object 200, there is an advantage that the spectrum P (and further the frequency characteristic F) close to the actual frequency characteristic of the measuring object 200 can be specified.

<C: Third Embodiment>
FIG. 7 is a block diagram of the signal analysis unit 24 of the frequency characteristic measuring apparatus 100 according to the third embodiment. As shown in FIG. 7, the signal analysis unit 24 includes a first synchronization addition unit 41, a second synchronization addition unit 42, and a characteristic analysis unit 44. The second synchronization adding unit 42 adds (synchronizes and adds) a plurality of synchronization sections obtained by dividing the reference signal SREF generated by the signal generation unit 12 of the first embodiment or the second embodiment on the time axis. A time domain reference signal XREF is generated. Each synchronization section of the reference signal SREF is a section that is synchronized with each other (section in which the phases match). Specifically, a section corresponding to a predetermined number of unit waveforms W in the reference signal SREF (a section corresponding to an integer multiple of the period) is selected as the synchronization section. For example, the second synchronization adding unit 42 adds (averages) the plurality of synchronization sections obtained by dividing the reference signal SREF supplied from the signal generation unit 12 for each unit waveform W (for each period) to each other, thereby averaging the reference signal XREF. Is generated. The second synchronous adder 42 sequentially generates the reference signal XREF by repeating the synchronous addition of the reference signal SREF.

  The first synchronous adder 41 generates a measurement signal XDET by adding (synchronously adding) a plurality of synchronous intervals of the measurement signal SDET detected by the detector 22 to each other. The measurement signal SDET detected by the detection unit 22 is a periodic signal (a signal having a period corresponding to the unit waveform W of the reference signal SREF as one period) reflecting the periodicity of the reference signal SREF. Therefore, a section corresponding to a predetermined number of unit waveforms W (a section corresponding to an integral multiple of the period of the measurement signal SDET) in the measurement signal SDET is selected as the synchronization section. For example, the first synchronization adder 41 generates the measurement signal XDET by adding (averaging) a plurality of synchronization sections obtained by dividing the measurement signal SDET for each unit waveform W (cycle). The first synchronous adder 41 sequentially generates the measurement signal XDET by repeating the synchronous addition of the measurement signal SDET.

  The characteristic analysis unit 44 in FIG. 7 uses the measurement signal XDET after the synchronous addition by the first synchronous addition unit 41 and the reference signal XREF after the synchronous addition by the second synchronous addition unit 42 to use the frequency characteristic F of the measurement target 200. Is analyzed. As shown in FIG. 7, the characteristic analysis unit 44 includes a frequency conversion unit 52, a signal processing unit 54, and an averaging unit 56.

  The frequency conversion unit 52 sequentially generates the spectrum PDET of each measurement signal XDET generated by the first synchronous adder 41 and the spectrum PREF of each reference signal XREF generated by the second synchronous adder 42. The signal processing unit 54 generates a spectrum PX by a predetermined process for the spectrum PDET and the spectrum PREF. For example, the signal processing unit 54 generates the spectrum PX of the component in which the influence of the reference signal SREF is suppressed from the measurement signal SDET by reducing the component of the spectrum PREF from the spectrum PDET. The averaging unit 56 generates the frequency characteristic F by averaging a plurality of spectra PX that are sequentially generated by the signal processing unit 54. The frequency characteristic F generated by the averaging unit 56 is output from the result output unit 26.

  The part (A1) and the part (A2) in FIG. 8 are the frequency characteristics F (parts (A)) output by the signal analysis unit 24 (characteristic analysis unit 44) when the reference signal SREF of the second embodiment is supplied to the measurement target 200. A1)) and the coherence characteristics (portion (A2)) of the output of the signal analysis unit 24. On the other hand, the part (B1) and the part (B2) in FIG. 8 are a frequency characteristic (part (B1)) and a coherence characteristic (part (B2)) output by the signal analysis unit 24 when pink noise is supplied to the measurement object 200. ). Since normal pink noise has no periodicity (regularity) and synchronous addition cannot be performed, the first synchronous addition unit 41 and the second synchronous addition unit 42 are omitted in the portions (B1) and (B2) of FIG. Assumes that. The coherence characteristics of the part (A2) and the part (B2) in FIG. 8 are indices (numeric values from 0 to 100) indicating the degree of influence of noise generated in each element of the frequency characteristic measuring apparatus 100, and the coherence is high. This means that the influence of noise is reduced.

  As can be understood from the part (A1) and the part (B1) in FIG. 8, the frequency characteristic F close to the actual characteristic of the measurement object 200 can be specified also in the third embodiment. Further, as understood from the comparison between the part (A2) and the part (B2) in FIG. 8, the synchronous addition of the reference signal SREF by the first synchronous adder 41 and the synchronization of the measurement signal SDET by the second synchronous adder 42. By performing the addition, the coherence characteristic is improved as compared with the configuration in which each synchronous addition is omitted (part (B2) in FIG. 8). That is, there is an advantage that the influence of noise generated in each element of the frequency characteristic measuring apparatus 100 is reduced.

<D: Modification>
Each of the above forms can be variously modified. Specific modifications are exemplified below. Two or more aspects arbitrarily selected from the following examples can be appropriately combined.

(1) Modification 1
The measurement object 200 is arbitrary. Specifically, the above embodiments are similarly applied to the case where the frequency characteristic F is measured using an acoustic device other than the equalizer (for example, an amplifier, a speaker, or a microphone) as the measurement target 200. It is also possible to measure the frequency characteristic F using the acoustic space as the measurement object 200. When the acoustic space is set as the measurement target 200, a sound emitting device that emits a sound wave corresponding to the reference signal SREF is employed as the output unit 14, and the frequency characteristics of the acoustic space during propagation (reflection / diffusion) in the acoustic space are adopted. A sound collecting device that picks up the added sound wave and generates the measurement signal SDET is employed as the detection unit 22. As can be understood from the above examples, the output unit 14 of each form is included as an element that outputs the reference signal SREF to the measurement target 200, and the reference signal SREF is generated in any form of an electric signal and reproduced sound (sound wave). It does not matter whether the data is output to the measurement object 200.

(2) Modification 2
The specific configuration of the signal generation unit 12 is arbitrary. For example, a configuration in which the signal generator 12 generates the reference signal SREF by acquiring and connecting the unit waveform W from a storage circuit (not shown) that stores the unit waveform W, or a signal generation from the storage circuit that stores the reference signal SREF A configuration in which the unit 12 acquires the reference signal SREF can also be adopted. An element that obtains the reference signal SREF from the outside via a communication line (communication network) is also employed as the signal generation unit 12.

(3) Modification 3
The waveform of the reference signal SREF can be appropriately changed from the above examples. For example, the periodicity of the reference signal SREF is not essential for the present invention. In each of the above embodiments, the case where the phase spectrum QP of the unit waveform W (reference signal SREF) fluctuates irregularly along the frequency axis is exemplified, but the phase spectrum QP of the unit waveform W is arbitrarily changed. . For example, a configuration in which the signal generation unit 12 generates the reference signal SREF from a unit waveform W (impulse waveform) in which the phase value of the phase spectrum QP is maintained at a constant value over the entire band may be employed. The linearity of the amplitude spectrum QA is not essential for the present invention. That is, even when the reference signal SREF in which the amplitude spectrum QA continues in a curve along the frequency axis is used, the same effects as those of the above embodiments can be realized. That is, the reference signal SREF that can be employed in the present invention is included as a time-domain signal in which the amplitude spectrum QA continues along the frequency axis (that is, smoothly continues along the frequency axis without local amplitude fluctuations). The presence or absence of periodicity or linearity and the phase spectrum QP are not questioned.

(4) Modification 4
The configuration of the signal analysis unit 24 is changed as appropriate. For example, a configuration in which the averaging unit 34 in FIG. 3 is omitted (a configuration in which the spectrum P generated by the frequency conversion unit 32 is output as the frequency characteristic F of the measurement target 200), or a configuration in which the second synchronous addition unit 42 in FIG. 7 is omitted. (A configuration in which the frequency characteristic F is analyzed only from the result of synchronous addition of the measurement signal SDET) may also be employed.

DESCRIPTION OF SYMBOLS 100 ... Frequency characteristic measuring apparatus, 200 ... Measurement object, 12 ... Signal generation part, 14 ... Output part, 22 ... Detection part, 24 ... Signal analysis part, 26 ... Result output part, 32, 52 ...... Frequency conversion unit, 34, 56 ... Average unit, 41 ... First synchronization addition unit, 42 ... Second synchronization addition unit, 44 ... Characteristic analysis unit, 54 ... Signal processing unit, SREF, XREF ... ... Reference signal, SDET, XDET ... Measurement signal, F ... Frequency characteristics, QA ... Amplitude spectrum, QP ... Phase spectrum.

Claims (5)

An apparatus for measuring frequency characteristics of a measurement object,
Signal generating means for generating a reference signal having an amplitude spectrum continuous along the frequency axis;
Output means for outputting the reference signal to the measurement object;
Detection means for detecting a measurement signal obtained by adding the frequency characteristic of the measurement object to the reference signal;
A frequency characteristic measuring apparatus comprising: signal analyzing means for analyzing the measurement signal. The signal generating means generates the reference signal in which a plurality of unit waveforms whose amplitude spectra are continuous along the frequency axis are arranged on the time axis,
The frequency characteristic measuring apparatus according to claim 1, wherein the signal analyzing means includes first synchronous adding means for synchronously adding the measurement signals. The signal analysis means includes
Second synchronous addition means for synchronously adding the reference signals;
The characteristic analysis means which analyzes the frequency characteristic of the said measuring object using the measurement signal after the synchronous addition by the said 1st synchronous addition means and the reference signal after the synchronous addition by the said 2nd synchronous addition means Frequency characteristic measuring device. The frequency characteristic measuring apparatus according to any one of claims 1 to 3, wherein the signal generation unit generates the reference signal in which an amplitude value at each frequency of an amplitude spectrum is a constant value. The frequency characteristic measuring apparatus according to any one of claims 1 to 3, wherein the signal generation unit generates the reference signal in which an amplitude value of an amplitude spectrum changes at a predetermined change rate with respect to a frequency.

Images