THE INSTITUTE OF ELECTRONICS, IEICE Technical Report INFORMATION AND COMMUNICATION ENGINEERS

This article is a technical report without peer review, and its polished and/or extended version may be published elsewhere.

Copyright ©2015 by IEICE

120-8551 5 E-mail: { 13kmc05@ms kaneda@c}.dendai.ac.jp

SN

DA LPF

A Study of nonlinear error in the impulse response measurement

using swept sine signal

Noritaka SATO and Yutaka KANEDA

Graduate School of Engineering, Tokyo Denki University, 5 Senju-asahi-cho, Adachi-ku, Tokyo 120-8551, Japan E-mail: { 14kmc13@ms kaneda@c }.dendai.ac.jp

Abstract The impulse response is an important characteristic for determining the frequency characteristic or reverberation time of acoustic systems. When an impulse response is measured using an excitation signal with a high sound pressure level to improve the SN ratio, nonlinear distortion occurs. The nonlinear distortion appears as a well-known harmonic distortion when a swept sine signal is used as an excitation signal. However, unexplained error components sometimes appear. In this report, we examined these error components. We found that the error components are intermodulation distortions caused by the swept sine signal and the reversely aliased signal of the LPF of a DA converter, or are the intermodulation distortions caused by the swept sine signal and the weak noise of the power supply. On the basis of these results, we proposed methods to reduce these errors and confirmed their effectiveness. Key words impulse response measurement, measurement error, nonlinear distortion, harmonic distortion, intermodulation distortion 1.

SN

(SS: Swept Sine)

TSP [1]Log-SS( TSP

ESS: Exponentialy Sept Sine) [2]

SN

(1m 100dB

)

2.

Fig.1( )

S

Copyright ©201 by IEICE 5This article is a technical report without peer review, and its polished and/or extended version may be published elsewhere.

- 63 -

H1/S

S S HS H

1/S H Fig.2 TSP Fig.3 Log-SS

TSP (Fig.2(a))TSP (Fig.2(b):

)(Fig.2(c)) Fig.3

(a)(b)(c) 3.

Fig.4

( ) Log-SS

Fig.5, 6

Fig.5 TSP(a) (b)

Fig.6(a)(b)Log-SS

Fig.5(a) TSP

Fig.5(b)

() Fig.5(a) n (n=2,3,4 )

n ( 2 3)

Fig.5(b)Fig.6(a)(b)

Log-SSFig.6(b)

fs/2(fs )

AD

H 1/SS S H HFig.1

Fig.2 TSP

(a) (b) (c)

Fig.3 Log-SS

(a) (b) (c)

DA AD

PCFig.4

(a)

(b)

Fig.5 TSP (a) (b)

- 64 -

[3] DA(

) Fig.6(b)

TSP Log-SS

Fig.5, 6

4. 4.1.

AD DA(LPF) fs/2(fs: )

AD DALPF fs/2fs/2 [4] Fig.7 LPF ADDA fs/2 f1

fs/2 fs f1 DAfs/2

Fig.8 fs=48kHz DA TSP 96kHz

TSPfs/2 fs/2

DATSP DA f1

fs f1 2f1 fs

fs f1

[3][4] 4.2.

Fig.7,8 0.9*fs/2 fs/2 fs/2 1.1*fs/2

DA 0.9*fs/2

LPFADAD

LPF

Fig.9 TSPFig.9(a)(b) Fig.10DA LPF ADLPF LPF

Fig.6 Log-SS(a) (b)

(a)

(b)

Fig.7 DA AD LPF ( 1 fs/2 )

Fig.8 TSP

fs/2

- 65 -

Fig.10

Fig.8 26kHzDA LPF 26kHz

5. 5.1.

Fig.11 Log-SS

Fig.12(a)

Fig.13Log-SS

TSP

Fig.14(a)(b)TSP

Fig.13 Log-SSFig.15

( ( 20 )TSP )

(a)

(b)

Fig.10 TSP

Fig.12 Log-SS

Fig.13 Log-SS

Fig.11 Log-SS

(b)

(a)

Fig.9 TSP

Fig.14 TSP

- 66 -

Fig.15 0.05TSP

2.2kHz/s Fig.14(a)100Hz

100Hz

TSP

( )

TSP 2

(Fig.16(a))

Fig.16(b) TSP-70 dB

50 Hz

TSP

100 Hz 5.2.

Fig.17Fig.18 Fig.17

Fig.18 Fig.17

Fig.19 Fig.17 Fig.17Fig.19 Log-SS

DFT

Fig.20 120HzFig.20 120Hz

Fig.16

Fig.18 Fig.17

Fig.17

Fig.19 Fig.17

Fig.15 TSP

- 67 -

( :240Hz,360Hz,..)

50Hz100Hz

2 240Hz220Hz( 120Hz 100Hz) 260Hz( 3

360Hz 100Hz) 50Hz 100Hz

120Hz50Hz 100Hz

5.3.

PC PC USB(AD/DA)

BOSE101MM(45W)

Fig.21 Log-SS

Fig.21 Fig.11Fig.17

Fig.21

6.

DA/AD LPF

JSPS 15H02728

[1] Y. Suzuki, F. Asano, H. Y. Kim, and T. Sone,”An Optimum

computer-generated pulse signal suitable for the measurement of very long impulse responses,” J. Acoust. Soc. Am. 97, 2,1119-1123, 1995.

[2] , ” SNTSP ”,

, 3-P-18, 2000. [3] , , , “TSP

PC AD/DA ”, , 3-Q-13, 2006.

[4] , , , “AD/DA”,

, 1-P4-2, 2014.

Fig.20 120Hz

Fig.21 Log-SS

- 68 -