JPH02154161A - Method and device for frequency analysis - Google Patents

Method and device for frequency analysis

Info

Publication number
JPH02154161A
JPH02154161A JP30788488A JP30788488A JPH02154161A JP H02154161 A JPH02154161 A JP H02154161A JP 30788488 A JP30788488 A JP 30788488A JP 30788488 A JP30788488 A JP 30788488A JP H02154161 A JPH02154161 A JP H02154161A
Authority
JP
Japan
Prior art keywords
frequency
band
frequency analysis
power spectrum
analysis
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
JP30788488A
Other languages
Japanese (ja)
Other versions
JP2784197B2 (en
Inventor
Masao Furusawa
政生 古沢
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Yamaha Motor Co Ltd
Original Assignee
Yamaha Motor Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Yamaha Motor Co Ltd filed Critical Yamaha Motor Co Ltd
Priority to JP30788488A priority Critical patent/JP2784197B2/en
Publication of JPH02154161A publication Critical patent/JPH02154161A/en
Application granted granted Critical
Publication of JP2784197B2 publication Critical patent/JP2784197B2/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Landscapes

  • Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)

Abstract

PURPOSE:To correctly display a power spectrum by setting a band so that a peak frequency of the power spectrum resulted from a narrow band frequency analysis is not overlapped with frequencies of the upper and lower limits of the band width on a wide band frequency analysis. CONSTITUTION:The acoustic pressure of a sound source 10 is detected by a microphone 12, amplified by an amplifier 14 and stored in memory means 16. Then, data of the means 16 are analyzed for narrow band frequencies in analysis means 18. By means of using the above result, the frequency fm, the spectrum power of which is made to be maximum, is obtained in frequency components detecting means 20. This frequency fm is inputted to band decision means 22, wherein a center frequency of the band is moved so that the frequency fm is made to be the center frequency fi of a certain band in the manner of not coinciding with the frequencies fT, fB of the upper and lower limits of each bandwidth, then new band is set. Further, the spectrum power in the band width newly obtained is calculated in wide band frequency analysis means 24, and the result is displayed on display means 26.

Description

【発明の詳細な説明】 (産業上の利用分野) 本発明は、騒音や振動などの評価に用いられる周波数分
析方法と装置に関するものである。
DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a frequency analysis method and apparatus used for evaluating noise, vibration, etc.

(発明の背景) 騒音、振動などを示す圧力、速度、加速度、変位、歪み
等の入力波形を周波数分析し、これら入力波形の評価を
行うことが広く行われている。
(Background of the Invention) Frequency analysis of input waveforms such as pressure, velocity, acceleration, displacement, distortion, etc. indicating noise, vibration, etc. is widely performed to evaluate these input waveforms.

この分析方法では、まず入力波形の全周波数範囲を10
24等分、512等分などに等分した刻みによりFFT
 (高速フーリエ変換)の周波数スペクトルを求める。
In this analysis method, first the entire frequency range of the input waveform is
FFT by dividing into 24 equal parts, 512 equal parts, etc.
Find the frequency spectrum of (fast Fourier transform).

そしてこの狭帯域の各周波数成分のスペクトルパワーす
なわち周波数成分の絶対値の2乗を求め、狭帯域パワー
スペクトルを求める(狭帯域周波数分析という)。しか
しこの狭帯域パワースペクトルの周波数軸を線形で広い
周波数範囲にわたるスペクトル表示をしたのでは、低周
波部分のスペクトル構造を知るのに不適当であり、また
聴覚との関係がよ(表わせなかったり、従来から広く使
われている1/3オクターブや1オクタ一ブ分析のよう
な手法によるスペクトルとの対比ができない、などの問
題が生じる。
Then, the spectral power of each frequency component in this narrow band, that is, the square of the absolute value of the frequency component, is determined to obtain a narrow band power spectrum (referred to as narrow band frequency analysis). However, if the frequency axis of this narrowband power spectrum is linear and the spectrum is displayed over a wide frequency range, it is inappropriate to understand the spectral structure of the low frequency part, and it has a poor relationship with the sense of hearing. Problems arise, such as the inability to compare spectra using techniques such as 1/3 octave and 1 octave analysis, which have been widely used in the past.

そこで周波数軸も対数目盛とし1/3.1オクターブ毎
のスペクトル表示を求める方法が用いられるようになっ
た(定パーセント(比率)バンド幅分析)。
Therefore, a method has been used in which the frequency axis is also on a logarithmic scale and a spectrum display is obtained every 1/3.1 octave (constant percentage (ratio) bandwidth analysis).

ここにこの定比率バンド幅による従来の広帯域周波数分
析は、各バンドの中心周波数を所定の規格で決めていた
。これはLkHzを基準としてlオクターブバンドの場
合には2°、1/3オクターブバンドの場合には2°″
の関係により各バンドの中心周波数を決めるものである
In the conventional wideband frequency analysis using this constant ratio bandwidth, the center frequency of each band was determined according to a predetermined standard. This is 2° for 1 octave band and 2° for 1/3 octave band based on LkHz.
The center frequency of each band is determined by the relationship.

しかしこの場合には、入力波形の周波数パワースペクト
ルの山がバンド幅の境目に重なることがあり、この時に
は見かけ上パワースペクトルが小さくなるという不都合
が生じる。
However, in this case, the peaks of the frequency power spectrum of the input waveform may overlap the boundaries of the bandwidths, and in this case, there is a problem that the power spectrum appears to be small.

第2.3図はその様子を示す図であり、第2図は入力波
形のスペクトルaがバンド幅す。の上・下限周波数間に
入ってi/)る場合を、第3図はスペクトルaがバンド
幅bo、b+の境目に入る場合をそれぞれ示している。
FIG. 2.3 is a diagram showing this situation, and in FIG. 2, the spectrum a of the input waveform has a bandwidth. FIG. 3 shows the case where the spectrum a falls between the upper and lower limit frequencies of i/), and the case where the spectrum a falls between the boundaries of the band widths bo and b+, respectively.

ここに定比率バンド幅の広帯域周波数分析結果である第
2図のスペクトルAは狭帯域分析のスペクトルaの面積
に等しく、第3図のスペクトルA1.A、の合計面積は
スペクトルaの面積に等しい、従って第3図のスペクト
ルA 1. A 2は第2図のスペクトルAの半分とな
って見かけ上少なく表われる不都合が生じることになる
Here, spectrum A in FIG. 2, which is the result of wideband frequency analysis of constant ratio bandwidth, is equal to the area of spectrum a of narrowband analysis, and spectrum A1 . The total area of A, is equal to the area of spectrum a, so spectrum A1 in FIG. A 2 becomes half of the spectrum A in FIG. 2, causing the inconvenience that it appears less clearly.

(発明の目的) 本発明はこのような事情に鑑みなされたものであり、広
帯域周波数分析のバンドの境目に人力波形のスペクトル
が入って分析結果のスペクトルが見かけ上小さく表われ
たりするのを防ぐようにした周波数分析方法を提供する
ことを目的とする。
(Objective of the Invention) The present invention has been made in view of the above circumstances, and is intended to prevent the spectrum of the analysis result from appearing apparently small due to the spectrum of the human waveform entering the band boundary of wideband frequency analysis. The purpose of this invention is to provide a frequency analysis method that does the following.

また本発明はこの方法の実施に直接使用する装置を提供
することを第2の目的とする。
A second object of the present invention is to provide an apparatus that can be used directly for carrying out this method.

(発明の構成) 本発明によればこの第1の目的は、高速フーリエ変換に
より得た狭帯域周波数分析の結果をさらに広帯域周波数
分析する周波数分析方法において、前記狭帯域周波数分
析結果のパワースペクトルの山の周波数が、前記広帯域
周波数分析のバンド幅の上・下限周波数に重ならないよ
うにバンドを設定したことを特徴とする周波数分析方法
により達成される。
(Structure of the Invention) According to the present invention, the first object is to provide a frequency analysis method for further wideband frequency analysis of the narrowband frequency analysis result obtained by fast Fourier transform. This is achieved by a frequency analysis method characterized in that the band is set so that the peak frequency does not overlap the upper and lower limit frequencies of the bandwidth of the wideband frequency analysis.

ここに入力波形のパワースペクトルの山のうち最大の山
の周波数がバンド幅の上・下限周波数に重ならないよう
にしたり、これらのスペクトルの山の周波数がバンドの
中心周波数になるようにしてもよい。
Here, the frequency of the largest peak among the peaks of the power spectrum of the input waveform may be set so that it does not overlap with the upper and lower limit frequencies of the band width, or the frequency of these peaks of the spectrum may be set to be the center frequency of the band. .

また第2の目的は、入力波形のサンプル値を記憶するメ
モリ手段と、前記サンプル値を用いて前記入力波形を高
速フーリエ変換し狭帯域周波数パワースペクトルを求め
る狭帯域周波数分析手段と、前記パワースペクトルが山
となる周波数を求める周波数成分検出手段と、前記パワ
ースペクトルが山となる周波数がバンド幅の上・下限周
波数に重ならないように広帯域周波数分析のバンドを決
めるバンド決定手段と、この決められたバンドに従って
パワースペクトルを求める広帯域周波数分析手段とを備
えることを特徴とする周波数分析装置により達成される
A second object of the present invention is to provide a memory means for storing sample values of an input waveform, a narrowband frequency analysis means for fast Fourier transforming the input waveform using the sample value to obtain a narrowband frequency power spectrum, and a narrowband frequency analysis means for obtaining a narrowband frequency power spectrum. a frequency component detection means for determining the frequency at which the power spectrum peaks; a band determining means for determining a band for broadband frequency analysis so that the frequency at which the power spectrum peaks does not overlap with the upper and lower limit frequencies of the bandwidth; This is achieved by a frequency analysis device characterized in that it includes a broadband frequency analysis means for determining a power spectrum according to the band.

(実施例) 第1図は本発明の一実施例のブロック図である。この実
施例は自動二輪車などの音源lOの騒音の分析を行うも
のである。音源10の音圧はマイク12により検出され
、このマイク12が出力するアナログ信号は増幅器14
で増幅されて入力波形iとなる。この入力波形iはA/
D変換すなわち所定のサンプリング間隔で量子化されて
メモリ手段16にメモリされる。例えば512(8ビツ
ト)や1024 (16ビツト)のサンプリング値をメ
モリできる。
(Embodiment) FIG. 1 is a block diagram of an embodiment of the present invention. This embodiment analyzes noise from a sound source 10 such as a motorcycle. The sound pressure of the sound source 10 is detected by the microphone 12, and the analog signal output from the microphone 12 is sent to the amplifier 14.
The input waveform i is amplified by the input waveform i. This input waveform i is A/
The data is D-converted, that is, quantized at predetermined sampling intervals, and stored in the memory means 16. For example, 512 (8 bits) or 1024 (16 bits) sampling values can be stored.

18は狭帯域周波数分析手段であり、FFT(高速フー
リエ変換)装置で構成される。この分析手段18のFF
Tでは離散するフーリエ変換の計算をすることになるが
、入力波形iの周期がサンプル数Nの整数倍にならない
時にはひずみが生じる。そこで時間窓をかけて波形を切
取ったり、時間窓の初めと終りの不連続を解消するため
にパニング窓などによる荷重関数を用いてもよいのは勿
論である。
18 is a narrowband frequency analysis means, which is composed of an FFT (fast Fourier transform) device. FF of this analysis means 18
At T, a discrete Fourier transform is calculated, but distortion occurs when the period of the input waveform i is not an integral multiple of the number of samples N. Therefore, it is of course possible to cut out the waveform by applying a time window, or to use a weight function such as a panning window to eliminate discontinuity between the beginning and end of the time window.

この分析手段18はFFTの計算により狭帯域周波数分
析を行い、その結果を用いて次にスペクトルパワーが最
大となる周波数で1を周波数成分検出手段20によって
求める。この周波数f、はバンド決定手段22に入力さ
れる。この手段22は各バンド幅の上・下限周波数ft
、f−に一致しないように、例えばこの周波数f、、が
あるバンドの中心周波数f、になるようにバンド中心周
波数を移し、新たなバンド幅を設定する。例えばこの中
心周波数f、に対して1オククーブバンドの場合には2
nオクターブ毎に、また1/3オクターブバンドの場合
には2n/3オクターブ毎に異なるバンドの中心周波数
flnが決められ、この時の各バンドの上限周波数fT
と下限周波数fsとは1オクターブバンドの場合には、 fT=21/2 ・fln iI B= 2−1/2・f I+’11/3オクター
ブバンドの場合には f1=2176 ・f、。
This analysis means 18 performs narrowband frequency analysis by FFT calculation, and using the result, the frequency component detection means 20 determines 1 at the next frequency where the spectral power becomes maximum. This frequency f is input to the band determining means 22. This means 22 has an upper and lower limit frequency ft for each band width.
, f-, for example, the band center frequency is shifted so that this frequency f, , becomes the center frequency f of a certain band, and a new bandwidth is set. For example, in the case of 1 occube band for this center frequency f, 2
The center frequency fln of a different band is determined every n octaves, or every 2n/3 octaves in the case of a 1/3 octave band, and the upper limit frequency fT of each band at this time
In the case of a 1-octave band, the lower limit frequency fs is: fT=21/2 ・fln iI B= 2-1/2・f I+'11/3 In the case of an octave band, f1=2176 ・f.

t、+=2−”’・ f で決められる。広帯域周波数分析手段24はこの新たに
求めたバンド幅に入るスペクトルパワーを計算し、それ
をCRTなどの表示手段26に表示させる。なおこの表
示手段26では、聴覚に対応した周波数補正(例えばJ
IS  C1502,1505などで定められた周波数
補正特性A)を行うようにしてもよいのは勿論である。
t, +=2-'''·f.The broadband frequency analysis means 24 calculates the spectral power that falls within this newly found bandwidth, and displays it on the display means 26 such as a CRT. The means 26 performs frequency correction corresponding to hearing (for example, J
Of course, the frequency correction characteristic A) defined in IS C1502, 1505, etc. may be used.

この実施例ではバンド決定手段22は周波数成分検出手
段20で求めた最大周波数成分子、がバンドの中心周波
Rt、になるようにバンドを決めているが、本発明は最
大周波数以外のパワースペクトルの山を選んでこの山が
バンドの境目に入らないように中心周波数を決めてもよ
く、また中心周波数を手動によって決めるようにしても
よい。
In this embodiment, the band determining means 22 determines the band so that the maximum frequency component obtained by the frequency component detecting means 20 becomes the center frequency Rt of the band. The center frequency may be determined by selecting a peak so that the peak does not fall on the boundary between bands, or the center frequency may be determined manually.

(発明の効果) 本発明の第1の発明は以上のように、狭帯域周波数分析
の結果からパワースペクトルの山を調べ、この山の周波
数がバンドの上・下限周波数にかからないようにバンド
を決めて広帯域周波数分析を行うから、パワースペクト
ルの山がバンドの境目に入って広帯機器周波数分析の結
果が見掛は上小さく表われるという不都合が生じること
がない。
(Effects of the Invention) As described above, the first invention of the present invention examines the peak of the power spectrum from the results of narrowband frequency analysis, and determines the band so that the frequency of this peak does not fall within the upper and lower limit frequencies of the band. Since the wideband frequency analysis is performed using the method, there is no problem that the peak of the power spectrum falls on the boundary between bands and the result of the wideband instrument frequency analysis appears smaller than it appears.

また第2の発明によれば、この第1の発明の方法を実施
するために直接使用する周波数分析装置を得ることがで
きる。
Further, according to the second invention, it is possible to obtain a frequency analyzer that can be used directly to implement the method of the first invention.

【図面の簡単な説明】[Brief explanation of the drawing]

第1図は本発明の一実施例のブロック図、第2.3図は
パワースペクトルが見かけ上小さく表われる理由を説明
するための図である。 16・・・メモリ手段、 18・・・狭帯域周波数分析手段、 20・・・周波数成分検出手段、 22・・・バンド決定手段、 24・・・広帯域周波数分析手段、 26・・・表示手段。 特許出願人 ヤマハ発動機株式会社
FIG. 1 is a block diagram of an embodiment of the present invention, and FIGS. 2 and 3 are diagrams for explaining the reason why the power spectrum appears small in appearance. 16... Memory means, 18... Narrowband frequency analysis means, 20... Frequency component detection means, 22... Band determination means, 24... Wideband frequency analysis means, 26... Display means. Patent applicant Yamaha Motor Co., Ltd.

Claims (6)

【特許請求の範囲】[Claims] (1)高速フーリエ変換により得た狭帯域周波数分析の
結果をさらに広帯域周波数分析する周波数分析方法にお
いて、 前記狭帯域周波数分析結果のパワースペクトルの山の周
波数が、前記広帯域周波数分析のバンド幅の上・下限周
波数に重ならないようにバンドを設定したことを特徴と
する周波数分析方法。
(1) In a frequency analysis method in which narrowband frequency analysis results obtained by fast Fourier transform are further subjected to wideband frequency analysis, the peak frequency of the power spectrum of the narrowband frequency analysis result is above the bandwidth of the wideband frequency analysis. - A frequency analysis method characterized by setting bands so that they do not overlap the lower limit frequency.
(2)前記請求項(1)に記載の周波数分析方法であっ
て、入力波形のパワースペクトル最大の周波数が、前記
広帯域周波数分析のバンド幅の上・下限周波数に重なら
ないようにバンドを設定したことを特徴とする周波数分
析方法。
(2) The frequency analysis method according to claim (1), wherein the band is set so that the maximum frequency of the power spectrum of the input waveform does not overlap with the upper and lower limit frequencies of the bandwidth of the wideband frequency analysis. A frequency analysis method characterized by:
(3)パワースペクトルの山の周波数がバンド幅の中心
周波数になるようにバンドを決めたことを特徴とする請
求項(1)または(2)に記載の周波数分析方法。
(3) The frequency analysis method according to claim (1) or (2), wherein the band is determined so that the peak frequency of the power spectrum becomes the center frequency of the bandwidth.
(4)入力波形のサンプル値を記憶するメモリ手段と、
前記サンプル値を用いて前記入力波形を高速フーリエ変
換し狭帯域周波数パワースペクトルを求める狭帯域周波
数分析手段と、前記パワースペクトルが山となる周波数
を求める周波数成分検出手段と、前記パワースペクトル
が山となる周波数がバンド幅の上・下限周波数に重なら
ないように広帯域周波数分析のバンドを決めるバンド決
定手段と、この決められたバンドに従ってパワースペク
トルを求める広帯域周波数分析手段とを備えることを特
徴とする周波数分析装置。
(4) memory means for storing sample values of the input waveform;
narrowband frequency analysis means for performing fast Fourier transform on the input waveform using the sample values to obtain a narrowband frequency power spectrum; frequency component detection means for determining frequencies at which the power spectrum peaks; a band determining means for determining a band for wideband frequency analysis so that the frequency does not overlap the upper and lower limit frequencies of the bandwidth; and a wideband frequency analysis means for determining a power spectrum according to the determined band. Analysis equipment.
(5)前記バンド決定手段は、狭帯域パワースペクトル
が最大となる周波数が広帯域周波数分析のバンドの中心
周波数となるようにバンドを決めることを特徴とする請
求項(4)に記載の周波数分析装置。
(5) The frequency analysis device according to claim 4, wherein the band determining means determines the band such that the frequency at which the narrowband power spectrum is maximum becomes the center frequency of the band for wideband frequency analysis. .
(6)バンド決定手段は、手動によってバンドの中心周
波数および上・下限周波数を設定可能とした請求項(4
)または(5)に記載の周波数分析装置。
(6) Claim (4) wherein the band determining means is capable of manually setting the center frequency and upper and lower limit frequencies of the band.
) or the frequency analyzer according to (5).
JP30788488A 1988-12-07 1988-12-07 Frequency analysis method and apparatus Expired - Fee Related JP2784197B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP30788488A JP2784197B2 (en) 1988-12-07 1988-12-07 Frequency analysis method and apparatus

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP30788488A JP2784197B2 (en) 1988-12-07 1988-12-07 Frequency analysis method and apparatus

Publications (2)

Publication Number Publication Date
JPH02154161A true JPH02154161A (en) 1990-06-13
JP2784197B2 JP2784197B2 (en) 1998-08-06

Family

ID=17974323

Family Applications (1)

Application Number Title Priority Date Filing Date
JP30788488A Expired - Fee Related JP2784197B2 (en) 1988-12-07 1988-12-07 Frequency analysis method and apparatus

Country Status (1)

Country Link
JP (1) JP2784197B2 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0526721A (en) * 1991-07-19 1993-02-02 Hitachi Ltd Detecting device for knocking
JP2009179314A (en) * 2008-01-30 2009-08-13 Eurocopter Method of optimizing ducted anti-torque rotor of rotorcraft, in particular helicopter, to minimize acoustic annoyance, and ducted anti-torque rotor obtained thereby
JP2017020961A (en) * 2015-07-14 2017-01-26 住友ゴム工業株式会社 Tire noise display method, and noise performance evaluation method using the same

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0526721A (en) * 1991-07-19 1993-02-02 Hitachi Ltd Detecting device for knocking
JP2009179314A (en) * 2008-01-30 2009-08-13 Eurocopter Method of optimizing ducted anti-torque rotor of rotorcraft, in particular helicopter, to minimize acoustic annoyance, and ducted anti-torque rotor obtained thereby
JP2017020961A (en) * 2015-07-14 2017-01-26 住友ゴム工業株式会社 Tire noise display method, and noise performance evaluation method using the same

Also Published As

Publication number Publication date
JP2784197B2 (en) 1998-08-06

Similar Documents

Publication Publication Date Title
Tolonen et al. A computationally efficient multipitch analysis model
Karjalainen A new auditory model for the evaluation of sound quality of audio systems
US5245665A (en) Method and apparatus for adaptive audio resonant frequency filtering
US6675114B2 (en) Method for evaluating sound and system for carrying out the same
US20050147262A1 (en) Method for decreasing the dynamic range of a signal and electronic circuit
EP3540734A1 (en) Spectral-dynamics of an audio signal
US8223990B1 (en) Audio noise attenuation
JPH0877683A (en) Inspecting device for rotating device and method of inspecting rotating device
Cabrera et al. Increasing robustness in the calculation of the speech transmission index from impulse responses
JPH02154161A (en) Method and device for frequency analysis
US6629049B2 (en) Method for non-harmonic analysis of waveforms for synthesis, interpolation and extrapolation
Cabrera et al. PsySound3: a program for the analysis of sound recordings
JP2000069597A (en) Method for measuring impulse response
Abramov et al. Increasing the Accuracy of Sound Signal Spectral Estimation According to the Properties of Hearing Analyzer
Abramov et al. Frequency Analysis of Sound Broadcasting Signal Parameters
Sottek et al. Sound quality evaluation of fan noise based on advanced hearing-related parameters
JP2841257B2 (en) Reverberation device
MacDonald et al. Ultrasonic vocalizations, their recording, and bioacoustic analysis
Sucic et al. Optimisation algorithm for selecting quadratic time-frequency distributions: performance results and calibration
US7505858B2 (en) Method for analyzing tone quality of exhaust sound
Sottek et al. Sound quality evaluation of fan noise based on hearing-related parameters
Sottek et al. Progress in Roughness Calculation
JP4590545B2 (en) Acoustic evaluation method and system
JPH05127668A (en) Automatic transcription device
EP4362496A1 (en) System and method for switching a frequency response and directivity of microphone

Legal Events

Date Code Title Description
R250 Receipt of annual fees

Free format text: JAPANESE INTERMEDIATE CODE: R250

LAPS Cancellation because of no payment of annual fees