SIST EN ISO 18233:2006

2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.URVWRUVNHAkustik - Anwendung neuer Messverfahren in der Bau- und Raumakustik (ISO 18233:2006)Acoustique - Application de nouvelles méthodes de mesure dans l'acoustique des bâtiments et des salles (ISO 18233:2006)Acoustics - Application of new measurement methods in building and room acoustics (ISO 18233:2006)91.120.20L]RODFLMDAcoustics in building. Sound insulation17.140.01Acoustic measurements and noise abatement in generalICS:Ta slovenski standard je istoveten z:EN ISO 18233:2006SIST EN ISO 18233:2006en01-julij-2006SIST EN ISO 18233:2006SLOVENSKI
STANDARD
EUROPEAN STANDARDNORME EUROPÉENNEEUROPÄISCHE NORMEN ISO 18233June 2006ICS 91.120.20 English VersionAcoustics - Application of new measurement methods in buildingand room acoustics (ISO 18233:2006)Acoustique - Application de nouvelles méthodes de mesuredans l'acoustique des bâtiments et des salles (ISO18233:2006)Akustik - Anwendung neuer Messverfahren in der Bau- undRaumakustik (ISO 18233:2006)This European Standard was approved by CEN on 19 May 2006.CEN members are bound to comply with the CEN/CENELEC Internal Regulations which stipulate the conditions for giving this EuropeanStandard the status of a national standard without any alteration. Up-to-date lists and bibliographical references concerning such nationalstandards may be obtained on application to the Central Secretariat or to any CEN member.This European Standard exists in three official versions (English, French, German). A version in any other language made by translationunder the responsibility of a CEN member into its own language and notified to the Central Secretariat has the same status as the officialversions.CEN members are the national standards bodies of Austria, Belgium, Cyprus, Czech Republic, Denmark, Estonia, Finland, France,Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Romania,Slovakia, Slovenia, Spain, Sweden, Switzerland and United Kingdom.EUROPEAN COMMITTEE FOR STANDARDIZATIONCOMITÉ EUROPÉEN DE NORMALISATIONEUROPÄISCHES KOMITEE FÜR NORMUNGManagement Centre: rue de Stassart, 36
B-1050 Brussels© 2006 CENAll rights of exploitation in any form and by any means reservedworldwide for CEN national Members.Ref. No. EN ISO 18233:2006: ESIST EN ISO 18233:2006

Foreword
This document (EN ISO 18233:2006) has been prepared by Technical Committee ISO/TC 43 "Acoustics" in collaboration with Technical Committee CEN/TC 126 "Acoustic properties of building elements and of buildings", the secretariat of which is held by AFNOR.
This European Standard shall be given the status of a national standard, either by publication of an identical text or by endorsement, at the latest by December 2006, and conflicting national standards shall be withdrawn at the latest by December 2006.
According to the CEN/CENELEC Internal Regulations, the national standards organizations of the following countries are bound to implement this European Standard: Austria, Belgium, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, Switzerland and United Kingdom.
Endorsement notice
The text of ISO 18233:2006 has been approved by CEN as EN ISO 18233:2006 without any modifications.
Reference numberISO 18233:2006(E)© ISO 2006
INTERNATIONAL STANDARD ISO18233First edition2006-06-01Acoustics — Application of new measurement methods in building and room acoustics Acoustique — Application de nouvelles méthodes de mesurage dans l'acoustique des bâtiments et des salles
ISO 18233:2006(E) PDF disclaimer This PDF file may contain embedded typefaces. In accordance with Adobe's licensing policy, this file may be printed or viewed but shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In downloading this file, parties accept therein the responsibility of not infringing Adobe's licensing policy. The ISO Central Secretariat accepts no liability in this area. Adobe is a trademark of Adobe Systems Incorporated. Details of the software products used to create this PDF file can be found in the General Info relative to the file; the PDF-creation parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In the unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below.
©
ISO 2006 All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from either ISO at the address below or ISO's member body in the country of the requester. ISO copyright office Case postale 56 • CH-1211 Geneva 20 Tel.
+ 41 22 749 01 11 Fax
+ 41 22 749 09 47 E-mail
copyright@iso.org Web
www.iso.org Published in Switzerland
ii © ISO 2006 – All rights reserved
ISO 18233:2006(E) © ISO 2006 – All rights reserved iiiContents Page Foreword.iv Introduction.v 1 Scope.1 2 Normative references.1 3 Terms definitions and abbreviated terms.1 3.1 Terms and definitions.1 3.2 Abbreviated terms.2 4 Designations.2 4.1 Maximum length sequence method (MLS).2 4.2 Swept-sine method (SS).2 5 Theory.2 5.1 General.2 5.2 Sound in a room.3 5.3 Sound transmission between two rooms.5 5.4 Using the frequency response function.6 6 Measurement of the impulse response.7 6.1 General.7 6.2 Excitation signal.7 6.3 Measurement of the response.9 7 Measurement of the frequency response function.14 8 Precision.14 9 Test report.15 Annex A (normative)
Maximum length sequence method.16 Annex B (normative)
Swept-sine method.20 Bibliography.26
ISO 18233:2006(E) iv © ISO 2006 – All rights reserved Foreword ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies (ISO member bodies). The work of preparing International Standards is normally carried out through ISO technical committees. Each member body interested in a subject for which a technical committee has been established has the right to be represented on that committee. International organizations, governmental and non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely with the International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization. International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2. The main task of technical committees is to prepare International Standards. Draft International Standards adopted by the technical committees are circulated to the member bodies for voting. Publication as an International Standard requires approval by at least 75 % of the member bodies casting a vote. Attention is drawn to the possibility that some of the elements of this document may be the subject of patent rights. ISO shall not be held responsible for identifying any or all such patent rights. ISO 18233 was prepared by Technical Committee ISO/TC 43, Acoustics, Subcommittee SC 2, Building acoustics. SIST EN ISO 18233:2006

ISO 18233:2006(E) © ISO 2006 – All rights reserved vIntroduction Stochastic signal analysis methods for the measurement of sound transmission phenomena started to be developed around 1960, but lack of available computing power excluded the use of these methods outside the best equipped research laboratories. The development of digitizing circuitry, powerful personal computers and the use of digital signal processing components in sound measuring equipment for field use, have made the application of measuring equipment based on extended digital signal analysis readily available. Dedicated instruments, as well as specialized software used on general computers, currently apply such methods and are already widely used. The new methods bring a number of advantages compared to the well-established classical methods, such as suppression of background noise and extended measurement range. However, there is also risk of unreliable results if certain guidelines are not followed. The new methods may demonstrate larger sensitivity to time-variations and change in the environmental conditions than the classical methods. This International Standard is developed to give requirements and guidelines for the use of new measurement methods in building and room acoustic measurements, but can also be used in the construction of measuring equipment for the implementation of the methods. As even an experienced user of equipment based on classical methods may be unaware of the difficulties and limitations for some applications of the new methods, the user is encouraged to develop a deeper understanding of the theoretical bases for the new methods. Instrument manufacturers are also encouraged to give further guidelines for applications and to make it an objective to design instruments that give warnings when results are not reliable. This International Standard gives guidelines and requirements for the application of new methods for the measurement of sound insulation in buildings and building elements and for the measurement of reverberation time and related quantities. Reference is made to the standards for the classical methods regarding what to measure, the number and the selection of measurement points, and the conditions for measurements. SIST EN ISO 18233:2006

INTERNATIONAL STANDARD ISO 18233:2006(E) © ISO 2006 – All rights reserved 1Acoustics — Application of new measurement methods in building and room acoustics 1 Scope This International Standard gives guidelines and specifies requirements for the application of new methods for the measurement of the acoustic properties of buildings and building elements. Guidelines and requirements for selection of the excitation signal, signal processing and environmental control are given, together with requirements for linearity and time-invariance for the systems to be tested. This International Standard is applicable to such measurements as airborne sound insulation between rooms and of façades, measurement of reverberation time and other acoustic parameters of rooms, measurement of sound absorption in a reverberation room, and measurement of vibration level differences and loss factor. This International Standard specifies methods to be used as substitutes for measurement methods specified in standards covering classical methods, such as ISO 140 (all parts), ISO 3382 (all parts) and ISO 17497-1. 2 Normative references The following referenced documents are indispensable for the application of this document. For dated references, only the edition cited applies. For undated references, the latest edition of the referenced document (including any amendments) applies. IEC 61260, Electroacoustics — Octave-band and fractional-octave-band filters IEC 61672-1, Electroacoustics — Sound level meters — Part 1: Specifications 3 Terms definitions and abbreviated terms 3.1 Terms and definitions For the purposes of this document, the following terms and definitions apply. 3.1.1 classical method conventional method of measurement where the resulting sound pressure levels or decay rates are determined directly from the recorded responses to random noise or impulse signals 3.1.2 new method measurement method in which various deterministic signals can be used to first obtain the impulse response of the system under test and from which the required sound pressure levels and decay rates can be obtained NOTE The new methods may have additional, intentional features such as giving results under situations where no result is obtained by the classical method. The new methods may, for instance, be more immune to noise from other sources. SIST EN ISO 18233:2006

ISO 18233:2006(E) 2 © ISO 2006 – All rights reserved 3.1.3 effective signal-to-noise ratio signal-to-noise ratio ten times the logarithm to the base 10 of the ratio of the mean-square value of the signal part caused by the excitation and obtained by the new method, to the mean-square value of the unwanted part of the signal obtained by the same method and caused by sources other than the excitation NOTE 1 The effective signal-to-noise ratio is expressed in decibels. NOTE 2 The effective signal-to-noise ratio is used as a substitute for the normal signal-to-noise ratio when establishing procedures for the new method based on a classical method. 3.1.4 peak-to-noise ratio ten times the logarithm to the base 10 of the ratio of the squared peak value of the signal part caused by the excitation and obtained by the new method, to the mean-square value of the unwanted part of the signal obtained by the same method and caused by other sources than the excitation NOTE The effective peak-to-noise ratio is expressed in decibels. 3.1.5 fractional-octave band frequency range, in hertz, from lower to higher band edge frequency for a fractional-octave-band filter as specified in IEC 61260 NOTE Both full-octave- and fractional-octave-band filters are designated fractional-octave-band filters. 3.2 Abbreviated terms MLS Maximum length sequence method SS Swept-sine method 4 Designations 4.1 Maximum length sequence method (MLS) An MLS method in accordance with this International Standard shall be designated as “ISO 18233–MLS”. 4.2 Swept-sine method (SS) An SS method in accordance with this International Standard shall be designated as “ISO 18233–SS”. 5 Theory 5.1 General The transmission of sound within a room as well as the transmission of sound between rooms may normally be considered as a close approximation to a linear and time-invariant system. The general theory applicable to such systems may therefore be used to establish the relationship between excitation and response for the sound transmission. The impulse response is the basis of all measurements. The methods are applicable to the velocities measured on structures as well as to sound pressures measured in rooms. SIST EN ISO 18233:2006

ISO 18233:2006(E) © ISO 2006 – All rights reserved 35.2 Sound in a room The scope of Parts 3 to 5 of ISO 140 and of Parts 9 to 12 of ISO 140 is to specify methods to measure the airborne sound insulation for building elements and the insulation between dwellings. ISO 3382 (all parts) specifies the measurement of reverberation time. In order to measure these quantities, the sound pressure level and the reverberation time in rooms by the application of noise excitation shall be measured. For the measurement of reverberation time, the noise source is switched on for a time sufficient to obtain a steady level. The source is thereafter switched off, and the decay of the sound in the room is observed. The time for switching the noise off is set to t = 0 in this International Standard. A recording of the sound pressure level versus time will, in general, contain information on the obtained stationary sound pressure level in the room as well as the reverberation time. A typical level versus time diagram is shown in Figure 1. The stationary sound pressure level before the sound source is switched off is given by the recording for t < 0, and information about the decay will be given for t W 0. The decay may be further processed to obtain the reverberation time.
Key L0 stationary noise level before the excitation is switched off LN background noise level t time NOTE The excitation is switched off at time t = 0. Figure 1 — Typical level versus time curve The classical methods for the measurement of airborne sound in rooms, defined in the ISO 140 and ISO 3382 series, specify a stochastic signal for the excitation. Although the room in most cases may be described as a deterministic system, statistical spread from the random excitation will lead to a certain stochastic variation in the result, which may be characterized by a standard deviation. Therefore, averaging of more measurements is normally needed to obtain results close to the stochastically expected values. Such averaging may for the classical method be combined with the spatial averaging needed to obtain a mean value for the room. The methods described in this International Standard intend to obtain measurement values in fractional-octave bands. Requirements and guidelines are selected accordingly. It has been shown (Reference [6]) that the expected decay in one particular observation point may be obtained without averaging, by processing the impulse response between the excitation signal (loudspeaker) SIST EN ISO 18233:2006

ISO 18233:2006(E) 4 © ISO 2006 – All rights reserved and the observation point (microphone) directly. This holds for the decay curve and the stationary levels as long as the system is linear and time-invariant. The theory may be extended and applied to sound in the source room, to sound in the receiving room, and to the transmission from the source to the receiving room. The measured response in the classical method based on noise excitation may, in theory, be described as a convolution between the excitation signal and the impulse response of the room. However, in the classical method with noise excitation, the response is recorded directly and information about the impulse response is normally not known. According to the new methods described in this International Standard, the results may be obtained from processing of the impulse response itself. NOTE 1 The impulse response is normally the combined impulse response of the system, consisting of amplifiers, transducers, applied filters, and the enclosure between the transmitting and the receiving points. Several methods may be applied to obtain the impulse response or the frequency response function, which is linked to the impulse response by Fourier transformation. All such methods may be used if they are able to demonstrate reliable results within normal measurement conditions. When a room has been excited by stationary white noise for a time sufficient to obtain stationary conditions and the noise is thereafter switched off at the time t = 0, the expected level at any time t W 0 will be [6]: 20ref()10lg()ddBtWLthttC∞⎡⎤⎢⎥=⎢⎥⎣⎦∫ (1) where 0W is a constant specifying the signal power per unit bandwidth of the excitation signal; ()ht is the impulse response; refC is an arbitrary selected reference value for the level calculation. The decay corresponds to the expected decay based on the classical method, which conventionally is approximated by a straight line. NOTE 2 Due to the fact that the running time, t, is the lower start point for the integration, the operation in Equation (1) may be described as backwards integration. In an alternative form of the formula, the integral starts at +∞ and runs backwards to the actual time. Historically, this was achieved using analog technology by playing a tape with the recorded response in the reverse direction. Equation (1) does not consider the extraneous noise normally accompanying a measurement. When a fractional-octave-band filter is a part of the measured system, Equation (1) will describe the expected decay according to the classical method for the applied filter band. Equation (1) may be used to compute the expected level at any time after the signal source was switched off. It may also be used to obtain the expected mean level before the excitation was switched off, 0L. The level may be obtained from Equation (1) by setting t = 0: 200ref010lg()ddBWLhttC∞⎡⎤⎢⎥=⎢⎥⎣⎦∫ (2) Figure 2 illustrates how the level versus time function is obtained by the classical and the new method. SIST EN ISO 18233:2006

ISO 18233:2006(E) © ISO 2006 – All rights reserved 5
a)
Classical method b)
New method Key L sound level h impulse response t running time NOTE In the classical method, an approximation, Lm(t), of the expected decay is found by averaging (ensemble) a number of individual decays, L1(t), L2(t), … LN(t), based on noise excitation. By application of the new method, the expected decay, L(t), is found by processing the impulse response h(t). Figure 2 — Illustration of the difference between classical and new method 5.3 Sound transmission between two rooms If a noise source is placed in a source room and the sound pressure level is measured at a point S, the expected level, 1L, may according to Equation (2) be obtained from the impulse response between the excitation point and the point S: 1().ht 2011ref010lg()ddBWLhttC∞⎡⎤⎢⎥=⎢⎥⎣⎦∫ (3) SIST EN ISO 18233:2006

ISO 18233:2006(E) 6 © ISO 2006 – All rights reserved In a similar way, if the sound level is measured in an adjacent receiving room at a point R, the expected level, 2L, may be obtained from the impulse response between the excitation point and the point R: 2()ht. 2022ref010lg()ddBWLhttC∞⎡⎤⎢⎥=⎢⎥⎣⎦∫ (4) The expected sound level difference, D, between the source and the receiver room may therefore be computed as: 21012220()d10lgdB()dhttDLLhtt∞∞⎡⎤⎢⎥⎢⎥⎢⎥=−=⎢⎥⎢⎥⎢⎥⎣⎦∫∫ (5) The variable describing the excitation power, 0W, is eliminated in the result for the level difference as the arbitrary chosen reference refC. NOTE The new methods specified in this International Standard can also be applied to the measurement of sound insulation of façades. In this context one of the measurement positions will be an outdoor position. 5.4 Using the frequency response function A sinusoidal signal has a unique position in the theory of signals and linear time-invariant systems. If the transients formed when signals are switched on and off are disregarded, the response for such a system to a sinusoidal excitation will always be sinusoidal with the same frequency. The amplitude (gain) and the phase may, however, change. Information about the change of amplitude and phase between input and output as a function of frequency is called the frequency response function of the system. The frequency response function will, as the impulse response, give full information about the response to any input signal. The frequency response function may be obtained from the impulse response by Fourier transformation. Equation (2) may be transformed by the application of Parseval's theorem: 22000()d()d2WWhttHωω∞∞−∞=π∫∫ (6) where ω=is=the=ang×lar=∞req×ency;=()Hω is the frequency response function obtained by the Fourier transformation of the impulse response():ht {…j()()()edtHhthttωω∞−−∞==∫F (7) where j1=− NOTE In Equation (6), it is assumed that h(t) = 0 for t < 0, which will be the case for a physical, causal system. SIST EN ISO 18233:2006

ISO 18233:2006(E) © ISO 2006 – All rights reserved 7Equation (6) shows that only the modulus of the frequency response function may be used for the level calculation. This is in contrast to the measurement of reverberation time, where both the phase and modulus of the frequency response function are required. By combining Equations (5) and (6), the expected sound level difference, D, between the source room and the receiving room may be computed from the frequency responses for the rooms. The expected sound level difference for a fractional octave band with lower band edge frequency 112fωπ= and upper band edge frequency 222fωπ= will be given by: 2121211222()d10lgdB()dHDLLHωωωωωωωω⎡⎤⎢⎥⎢⎥⎢⎥=−=⎢⎥⎢⎥⎢⎥⎢⎥⎣⎦∫∫ (8) 6 Measurement of the impulse response 6.1 General The impulse response for a room will typically be an oscillatory signal with a large number of periods. The envelope of the signal will be irregular but typically have a fast attack-time and an exponential decay. The impulse response may be measured as the response of the room to a very short acoustic pulse. However, it will in most cases where sources other than a loudspeaker are used, be difficult to have sufficient control of the spectral content and the directional characteristics of the excitation. To obtain the required control of the excitation signal, the impulse response is in most practical cases obtained by digital signal processing. The room is excited by a known signal for a certain time and the impulse response is calculated from the response to the excitation. The excitation signal is distributed over a longer period of time to increase the total radiated energy. This procedure will enhance the achievable dynamic range and reduce the influence of extraneous noise. Several methods for the measurement of the impulse response are described in the literature, see References [6] to [8] and [13] to [15] in the Bibliography. For measurements of the impulse response, movement of the source or the microphones is not acceptable as it will violate the requirement for time-invariance. The impulse response of a room is formed by a complex interaction of sound waves reflected between the floor, ceiling and walls of the room. Between the reflections, the air in the room influences the transmission. Movement of the air or change in the speed of sound (temperature) may also violate the requirement for time-invariance. 6.2 Excitation signal 6.2.1 General In the classical methods, random noise or an impulse with a bandwidth at least equal to the bandwidth of the measurement channel is required. The random nature of the noise will give a stochastic distribution of the measured levels and will limit the repeatability of the measurement. The new methods apply deterministic excitation signals, i.e. they can be accurately reproduced, and thereby enhance the repeatability of the measurement. SIST EN ISO 18233:2006

ISO 18233:2006(E) 8 © ISO 2006 – All rights reserved 6.2.2 Spectral requirements 6.2.2.1 General The effective frequency range of the excitation signal shall at least cover the actual fractional-octave band being measured. If a broad-band measurement covering the whole audio range is being performed, the aim is to approximate the shape of the spectrum of the excitation signal, as captured at the receiver position, to that of the ambient noise prevailing there. By this, a frequency-independent signal-to-noise ratio will be obtained. The typical sources of background noise (air-conditioning, traffic, etc.) tend to have a spectral distribution that increases with decreasing frequency. For this reason, the excitation signal should feature an emphasis at lower frequencies when room impulse responses are to be measured. In many of these cases, a pink excitation signal (with constant energy per fractional-octave band) is suitable to obtain a sufficient signal-to-noise ratio. In sound insulation measurements, however, the sound reduction index normally increases at higher frequencies, making it necessary to increase the energy of the excitation signal in this range. The most sophisticated emphasis scheme would involve compensating the acoustical power response of the measurement loudspeaker and adapting to the spectral distribution of the background noise. A smoothed version of the latter, multiplied by the inverted speaker response, confined to the intended frequency range, can be used as a template for the generation of a suitable excitation spectrum. 6.2.2.2 Repetitive excitation If a repetitive excitation signal is used, the spectrum of the excitation will consist of narrow spectral lines where the distance between adjacent lines, ∆f, will be given as the inverse of the time for one repetition period REPT: REP1fT∆= (9) In order to ensure that all modes of the room are excited, the repetition period shall not be shorter than the reverberation time, T, for the room being measured. This requirement applies to the measurement of reverberation time as well as level differences: REPTTW (10) NOTE Each room mode may be approximated by a second-order band-pass function with a certain quality factor (Q-factor). A larger Q-factor means a narrower frequency response and a longer decay after excitation has ceased. For a second-order function with bandwidth (–3 dB), B, in hertz, the virtual reverberation time will be approximately (2,2/B). The requirement for the repetition time ensures that at least two spectral lines of excitation fall within the bandwidth of any room mode. 6.2.2.3 Non-repetitive excitation A non-repetitive excitation signal may be of any suitable length. However, the excitation shall be succeeded by a period of silence in order to allow the decaying response to be properly recorded. The decay shall be recorded over a period equal to at least half of the reverberation time. For a sweep excitation from a low to a high frequency as described in Annex B, the required length of the period of silence will normally be determined by the reverberation time for the higher frequencies. 6.2.3 Level and linearity The sound power in the excitation shall be sufficiently high to obtain an effective signal-to-noise ratio satisfying the requirements given in the International Standard specifying the applicable classical method. Methods involving deterministic excitation signals are generally more efficient at suppressing extraneous noise than the classical method. Enhancement of the signal-to-noise ratio by 20 dB to 30 dB or more compared to the classical method may be obtainable. SIST EN ISO 18233:2006

ISO 18233:2006(E) © ISO 2006 – All rights reserved 9The use of loudspeakers typically introduces non-linear distortion in the system. Distortion violates the requirement for linearity in this method. Distortion due to the loudspeaker increases with the excitation level. The user shall be aware of the problem and experiment with the excitation level to obtain the optimum signal-to-noise ratio. Sometimes the signal-to-noise ratio may be increased by reducing the excitation level. This needs special consideration with the MLS method as described in Annex A. When properly established, the swept-sine method described in Annex B allows elimination of artefacts in the measurement result caused by harmonic distortion. The region where the impulse response decays into the noise floor will usually be the most affected by non-linearities. This makes the measurement of reverberation more vulnerable to the effect of distortion than the measurement of level differences. 6.2.4 Directivity The directivity for the source shall comply with the requirements given in the applicable classical method. 6.2.5 Number of source positions The number of source positions shall comply with the requirements given in the applicable classical method. 6.3 Measurement of the response 6.3.1 Transducers for measurement Transducers for the measurement, normally measurement microphones, shall comply with the requirements given in the applicable classical method. 6.3.2 Frequency weighting The methods in Annex A and Annex B describe the measurement of broadband impulse responses. The broadband impulse responses shall be further processed to obtain the fractional-octave-band weighted impulse response for the required range of frequency bands. Although Equations (1) to (5) are general, the impulse responses in these formulas shall be fractional-octave-band weighted in order to produce results valid for fractional-octave bands. In principle, the fractional-octave-band weighted impulse response is obtained as the output from a fractional-octave-band filter, as specified in IEC 61260, to the response of the broadband impulse response. When selecting methods to perform the requested frequency weighting, precautions shall be taken to ensure that the tolerances for the frequency weighting are within the requirements stated in IEC 61260 for the appropriate class of filters as specified in the classical method. Sampling frequency shall be selected and precautions against effects from frequency aliasing shall be taken accordingly. For excitation with repetitive signals, the response is recorded with the time and frequency resolution set by the requirements for the excitation signal and with a length equal to one or more periods of the excitation signal. For non-repetitive excitation and measurement of level, the recorded part of the response shall cover the time from the start of excitation to the time where the response in each fractional octave band has decayed by more than 30 dB. For the measurement of reverberation time with non-repetitive excitation, the record shall at least cover the part of the decay required in the applicable classical method. 6.3.3 Level linearity and dynamic range The signal processing shall have sufficient resolution and dynamic range to comply with the requirements for level linearity as specified in IEC 61672-1. Measurement equipment made for obtaining the result by the new method can normally not be tested as conventional sound measuring equipment. In general, the microphone signal is digitized and the result is SIST EN ISO 18233:2006

ISO 18233:2006(E) 10 © ISO 2006 – All rights reserved obtained by digital processing on the samples representing the microphone signal. Proper operation of the microphone and digitizing circuitry can often be verified by conventional testing, but not the entire calculation process. The accuracy of the digital processing is considered to be determined by the design of the equipment and not affected by ageing or changes in the environmental conditions of operation as long as valid results are presented. It is recommended to validate the design and operation of the system by making measurements where the results can be compared to results obtained by the classical method. Measurements at fixed positions in rooms where the acoustical conditions are well controlled may be used. However, a time-invariant system with electrical input and output signals may be more convenient. Such a system may be a digital reverberator without time modulation. The validation should be performed with the range of possible reverberation times. The performance of the measuring equipment in reduced signal-to-noise ratios may be investigated by adding broadband random noise to the analog input or output signals. It is recommended to test the microphone and digitizing circuitry as well as the excitation generator at regular intervals, as appropriate (periodic verification). 6.3.4 Crosstalk The application of deconvolution measurement techniques allows measurements with large dynamic ranges, often including levels that extend below the level of extraneous noise. Even levels below the inherent noise levels in microphones and the measurement system may be measured. Care shall therefore be taken to eliminate influence from unwanted signal paths, such as electrical crosstalk. Cables for the excitation, such as loudspeaker cables, shall be located far away and screened from microphone cables. Even internal crosstalk in the instrumentation, normally buried in the self-noise, may show up. Sufficient immunity from crosstalk may be demonstrated by substituting the normal transducer (microphone) with a dummy device having very low sensitivity to the signal to be measured. A display of the impulse response, if available, may indicate a possible crosstalk problem. Sound signals are normally delayed, even the direct sound, due to the speed of sound and distance between the source and the receiver. Crosstalk signals, being electrical signals, are normally not delayed. To exclude the influence of any residual crosstalk, windowing may be applied at the beginning of the impulse response to attenuate any non-acoustical components. 6.3.5 Limits for the time integration 6.3.5.1 Measurement of level Equation (2) specifies an infinitely long integration period. This is neither possible nor wanted. The length of the recorded impulse response will give the maximum value for the upper integration limit. Measured impulse responses will always be accompanied by unwanted noise from extraneous sources and from self-noise in the instruments. Effects from violation of the requirement for linearity and time-invariance may add to the noise. The contribution from the noise in the integral will increase with an increasing length of the integration interval. If the integration is performed between 0and 1t, the level will be given by: 11220ref00110lg()d()ddBttLWhttttCε⎧⎫⎡⎤⎪⎪⎢⎥=+⎨⎬⎢⎥⎪⎪⎣⎦⎩⎭∫∫ (11) where ()tε is the background noise signal. In Equation (11), the cross term is neglected because it is assumed that there is no correlation between ()ht and ()tε. Too low a value for the upper integration limit will give too low a value for the integral. Figure 3 shows a sketch of how the value of 1taffects the calculated level. SIST EN ISO 18233:2006

ISO 18233:2006(E) © ISO 2006 – All rights reserved 11 a)
Sketch of the envelope of the impulse response, the background noise and the combination of impulse response and noise
b)
Sketch of the calculated levels as a function of the upper integration limit t1 Key L sound level (dB) tr time relative to reverberation time t1r time for integration limit relative to reverberation time S signal from impulse response N signal from background noise CSN signal from combined impulse response and noise NOTE 1 The effective signal-to-noise ratio is only 10 dB in this example in order to display the contribution from the background noise. Time is relative to the reverberation time. NOTE 2 Time is relative to the reverberation time. The levels are as calculated according to Equation (11). The figure shows the first part, S, the second part, N, and the complete integral, CSN. 0 dB indicates the correct level corresponding to an infinite integration period without any noise contribution. Note that the difference between the noise and the maximum envelope of the impulse response is only 10 dB in this example. Figure 3 — Integration limits SIST EN ISO 18233:2006

ISO 18233:2006(E) 12 © ISO 2006 – All rights reserved Although the reverberation time may not be known, the upper integration limit 1t in Equation (11) for a level measurement shall be selected from a rough estimate of the reverberation time T for the appropriate fractional-octave band in question: 13TtW (12) This implies that the integration should at least be made until the –20 dB point on the squared impulse response curve. The optimum value of 1t will depend on the signal-to-noise ratio. If the background noise is low, higher accuracy is obtained by increasing the integration interval. The aim is to obtain an effective signal-to-noise ratio exceeding or equal to the required signal-to-noise ratio in the classical method. Classical methods referred to in this International Standard describe procedures to correct measured levels when the signal-to-noise ratio is low. The new measurement methods may be used to measure the effective signal-to-noise ratio and thus compensate automatically for the influence from the noise as a part of the method. If the noise compensation is part of the method, no further noise compensations shall be applied, even if stated in the description of the classical method. The integration limit may be selected individually for each fractional-octave band or as a common limit based on the highest value of T. Equation (11) shall be used for the calculation of level difference as defined by Equation (5). The integration limit, 1t, from Equation (12) may be selected independently for the two rooms or the highest value may be used. 6.3.5.2 Measurement of reverberation As for level measurements, the upper limit for the time integral in Equation (1) shall be limited in order to reduce the contribution from unwanted noise: 22220ref1()10lg()d()ddBttttLtWhttttCε⎧⎫⎡⎤⎪⎪⎢⎥=+⎨⎬⎢⎥⎪⎪⎣⎦⎩⎭∫∫ (13) It is recommended to set 2t to be the time where the envelope of the exponential decay in the impulse response 2()ht intersects with the tail of the measured response determined by the extraneous background noise. Different methods are described in the literature to compensate for the noise and the truncation of the integration interval (see Reference [10]). 6.3.6 Response averaging Averaging more impulse responses before the final impulse response is further processed may enhance the effective signal-to-noise ratio. The impulse response for a room will ideally be determined by a deterministic process and will give a repeatable signal. The extraneous noise, however, will typically be a stochastic signal uncorrelated with the impulse response. The effective signal-to-noise ratio will then increase by 3 dB for each doubling in the number of measurements in the averaged response. Violation of the requirements for time-invariance and non-linearity will reduce the enhancement of the effective signal-to-noise ratio and will set a limit for the achievable effective signal-to-noise ratio. 6.3.7 Number of measurement points and spatial averaging The number of measurement points (combinations of source/microphone positions) shall comply with the requirement in the applicable classical method. The combination of levels or level differences shall comply with the requirements in the applicable classical method. If it is required that the spatial averaged level differences be obtained by first making the average of levels in each room, a similar procedure shall be followed. SIST EN ISO 18233:2006

ISO 18233:2006(E) © ISO 2006 – All rights reserved 136.3.8 Stability and time-invariance All parts of the signal chain, from the excitation to the received signal, shall be time-invariant. This is particularly important with the MLS method described in Annex A, whereas the swept-sine method described in Annex B is more robust to such variations. The need for time-invariance improves the gain (amplitude) and particularly the phase stability. Normally, electronic components used in analog and digital signal processing are suffici
...