Noise Health Home 

[Download PDF]
Year : 1999  |  Volume : 2  |  Issue : 5  |  Page : 1--15

Hearing protection in the military environment

A Dancer, K Buck, P Hamery, G Parmentier 
 French-German Research Institute of Saint-Louis, 5 rue du général Cassagnou, BP 34, 68301 SAINT-LOUIS Cedex, France

Correspondence Address:
A Dancer
French-German Research Institute of Saint-Louis, 5 rue du général Cassagnou, BP 34, 68301 SAINT-LOUIS Cedex


The present state of passive (linear and non-linear) and active techniques for hearing protection in the military environment is reviewed. Solutions which allow to protect the ear from large continuous and high-level impulse noises while preserving the operational abilities of the personnel (detection, localization, communication...) are emphasised.

How to cite this article:
Dancer A, Buck K, Hamery P, Parmentier G. Hearing protection in the military environment.Noise Health 1999;2:1-15

How to cite this URL:
Dancer A, Buck K, Hamery P, Parmentier G. Hearing protection in the military environment. Noise Health [serial online] 1999 [cited 2021 Dec 9 ];2:1-15
Available from:

Full Text


In the military environment, personnel may be exposed to very high-level noises: impulse noises produced by weapons can reach 190 dB peak at the ear, continuous noises in the vicinity of jet engines can exceed 130 dBA! Although these extreme exposures conditions are relatively infrequent and concern only a few people, they present serious problems as they can produce immediate cochlear lesions and hence, large Permanent Threshold Shifts (PTS) (NATO, 1987; Dancer et al., 1998). Moreover, even "moderate" intensity noises: impulse noises of 150 to 165 dB peak (such as those produced by rifles in military training) (Paakkonen and Kyttala, 1994), continuous noises of 100 to 120 dBA (such as those existing in armoured vehicles or planes), are well above the admissible exposure conditions (ISO 1999, 1990). Altogether these noises correspond to the major cause of acoustic trauma among the military personnel. In 1996, 966 soldiers suffering from acute acoustic trauma have been treated in French hospitals at a 4 million dollar cost. In 1996, France spent 60 million dollars to compensate for the Noise Induced Hearing Losses (NIHL) of veterans (in 1997, the USA spent about 270 million dollars to compensate for the NIHL of 57,993 veterans).

Personnel exposed to weapon noise must be equipped with correctly fitted Hearing Protectors (HP) offering an adequate performance. However, the use of HP (together with the pre­existing hearing conditions: TTS and/or PTS) leads to difficulties in detection, localization and identification of acoustic sources in the environment, and impedes the efficiency and the security of the soldier (Garinther et al., 1985). The HP also generally reduce the speech intelligibility (actually, the decrease in speech intelligibility is a complicated phenomenon which depends on the characteristics of the speech signal and of the interfering noise, as well as on the type of HP and of the intercom system) (Steeneken et al., 1994; Wessling, 1997). Decrease in speech intelligibility can drastically reduce the global performance of complex and expensive weapon systems. By adjusting experimentally the intelligibility of the communications in a tank, Peters and Garinther (1990) have shown that the percentage of successful missions (including navigation, reporting and gunnery) is proportional to speech intelligibility [Figure 1].

The HP must be designed and evaluated by taking into account the need for hearing protection and the consequences of their use for the operational abilities. Moreover, the HP should be fitted to each soldier station! If not, the risk is that the protection will not be worn. Keeping these requirements in mind, we shall review some aspects of the passive and active HP techniques.

Noise Reduction and Insertion Loss Measurement Techniques

Noise Reduction (NR) is the difference between the incident and the received sound pressure levels (SPLs) when a HP is worn (i.e., between the free field and the entrance to the ear canal - or the tympanum - for an earmuff; between the free field and the tympanum for an earplug). Insertion Loss (IL) is the difference between the SPLs measured at a reference point (i.e., at the entrance to the ear canal - or at the tympanum - for an earmuff; at the tympanum for an earplug) before and after a HP is put into place (Berger, 1986).

For the assessment of the attenuation afforded by earplugs and earmuffs at very high levels, the classical measurements performed by means of the subjective method: Real-Ear-Attenuation-at ­Threshold (REAT) at low steady-state noise levels (according to ISO 4869-1, 1990 or ANSI S12.6-1997) are not suitable. First of all, this method does not allow the evaluation of the peak pressure of an impulse under a HP. Even if serious doubts exist about the pertinence of "peak level" measurements under HP as part of the classical DRC (Dancer and Franke, 1995), it is nevertheless essential to get this information: the ISO standard (1999) restricts the equal energy principle to peak levels below 140 dB, the American Conference of Governmental Industrial Hygienists (1988) does not allow unprotected exposure to impulses above 140 dB peak, and most Damage Risk Criteria (DRC) for weapon noises are based on the duration, the number of impulses, and the peak pressure (NATO, 1987, Pekkarinen et al., 1992; Dancer, 1992). Moreover, the behavior of the HP undergoing the action of high-level noises may exhibit non-linearities. The development of a non-linearity, its importance, the modification of its characteristics as a function of the parameters of the noise as well as its net effect (either "positive" or "negative") on the global attenuation, are generally unpredictable. For that reason the attenuation of each HP should be measured in the actual exposure conditions for which it is intended to be used! The same kind of limitations apply to the "Microphone-In-Real­Ear" (MIRE) measurement technique (Hellstrom, 1992). Moreover, MIRE is presently unsuitable for earplugs attenuation measurements and, last but not least, it is impossible to use as a routine technique with high-intensity impulses because of the security of the subjects.

Therefore, the only possibility to assess the actual behavior of the HP when exposed to high-level (impulse or continuous) noises, to characterize their non-linearity (if any), and to measure amplitude spectrum and peak pressure attenuations, is to use an Acoustical-Test-Fixture (ATF) and preferably an artificial head with an ear simulator (Parmentier et al., 1999). ATF are currently used to evaluate the physical attenuation afforded by earmuffs in steady-state noise. In these conditions the ATF must comply with standards (ISO and/or ANSI standards) (ISO TR 4869-3, 1989; ANSI 12.42-1995). However, ATF which are commercially available (Bruel & Kjaer, Knowles Electronics Manikin for Acoustic Research, Head Acoustics GmbH) are either not suited for our purpose due their mechanical configuration and/or present a poor acoustic isolation. In order to reach better performances, we designed a new ATF. The "head" was arranged to fit: (i) the HEAD Acoustics "external ear" and "circumaural region", (ii) the Bruel & Kjaer ear simulator (type 4157). The 1/2" Bruel & Kjaer microphone (type 4134) of the original ear simulator is replaced by an under-polarized (28 V instead of 200 V) 1/4" microphone (type 4136) in order to allow the measurement of peak levels up to 190 dB, [Figure 2]. The measured Transfer Function of the Open Ear (TFOE) of this ATF is in close agreement with the experimental data published by Shaw (1974) and can be considered to be linear up to a peak pressure of about 190 dB at the ear simulator microphone. Due to a special assembling method and to various suspending and damping devices, the maximum IL obtained by the ATF (the ear canal being sealed) is better than 80 dB from 0.4 to 5 kHz [Figure 3] and well above the ANSI and ISO criteria. This way, it becomes possible to study the IL of hearing protectors without any limitation (= 80 dB) of the dynamics of the measurements! This ATF is perfectly suitable for measurements with earmuffs. Although the HEAD Acoustics device provides some simulation of the "ear canal" tissues, measurements with earplugs still needs to be improved: thickness, compliance, geometry of the simulated ear canal is not perfect... and finally international standardization is desirable. Generally speaking, very large IL values which are measurable with the ISL ATF are not taken into account for hearing protection evaluation because they exceed the Bone Conduction (BC) thresholds (Berger, 1983, Schroeter and Poesselt, 1986). However, the measurements which are feasible thanks to the large dynamics of our ATF allow (i) to determine the physical performance of (almost) any HP, (ii) to know the actual pressure time history existing below a HP under (almost) any exposure conditions, and (iii) to apply any correction curve corresponding to either the BC limits, or to the Physiological Masking noise (PM) and the Occlusion Effect (OE) (see for example: Schroeter and Poesselt, 1986), and thus allow a very general approach of the determination of the HP efficiency.

Assessment of the Hearing Protection Efficiency

There are two main methods to decide whether the hearing protection afforded by a HP is sufficient: (i) by measuring the signal close to the head of the subject and using the IL characteristics of the protector in order to calculate the equivalent dose of acoustic energy to which the subject would be exposed unprotected (Berger, 1986), (ii) by measuring the pressure-time signature under the protector and introducing the peak pressure and duration into the classical criteria for weapon noises.

Which method is the most representative of the actual hearing protection afforded by a protector?

The second method, which is sometimes used for impulse noises, consists of an untested extension of the use of the weapon noise criteria (DRC) which have been primarily designed to apply to free field pressure-time signatures and to unprotected ears (Dancer and Franke, 1995). Moreover, peak pressure attenuation measurements grossly underestimate the protection afforded by the HP when used in conjunction with the classical DRC (Patterson, 1998). Actually, the risk corresponding to the exposure to a signal with a long rise time (as recorded under a HP: low-pass filtering) is much lower than the risk corresponding to the exposure to a shock wave with an instantaneous rise time (with the same peak pressure).

The LAeq8 attenuation based on IL measurements (method 1) gives in most exposure conditions a good evaluation of the auditory hazard but in some other cases it still underestimates the efficiency of the HP. In some instances, a still better prediction can be achieved by using a weighting function corresponding to the curve of hearing sensitivity at threshold instead of the A-weighting function (Patterson, 1998).

On the other hand, it could well be that the very high protection afforded by ordinary hearing protectors (standard earmuffs are able to fully protect the ear against 100 impulses of 187 dB peak pressure simulating artillery noise) (Johnson and Patterson, 1992; Patterson et al., 1993; Patterson and Johnson, 1996), is due to the non-linearity of the middle ear. Price and Kalb (Price, 1974; Price and Kalb, 1991; Price, 1992) emphasised the limited tympano-ossicular chain displacements due to the non-linear mechanical characteristics of the annular ligament when exposed to large impulses. If important for unprotected exposures, this effect could be essential in order to understand the surprisingly small damages induced by large but slow-rising impulses as those existing under HP.

Insertion Loss of Hearing Protectors in High-Level Impulse Noise

The attenuation afforded by earplugs (E-A-R foam, E-A-R Ultrafit, E-A-R Ultratech, E-A-R Link Series 3 eartips, RACAL Airsoft, perforated earplugs: RACAL Gunfender, ISL plug...), and earmuffs (WILLSON SB 258, E-A­R Ultra 9000...) was measured during exposure to Friedlander waves of -150, 170 and 190 dB peak pressures (A-durations: - 0.2 and z 2 ms) under normal and grazing incidences (Parmentier et al., 1994). Some HP behave linearly: no significant modification of the IL is observed when the peak pressure of the impulse changes. [Figure 4] presents the IL of the RACAL Airsoft earplug as a function of frequency (1/3 octave bands) for impulses of - 150, 170 and 190 dB peak (-z 2 ms A-duration, normal incidence). It is interesting to note that the attenuation of the RACAL Airsoft earplug as measured by REAT methods (ISO 4869 and ANSI S 12.6-1997) at low steady-state noise levels by the Berufsgenossenschaftliches Institute for Arbeitssicherheit (quoted by Kusy, 1991) is comparable to our results obtained with high-level impulses. This indicates clearly (i) that this earplug behaves linearly, (ii) that the ATF reproduces reasonably well the average behaviour of the human ear. Some HP behave nonlinearly. In some instances the non-linearity is unfavourable (i.e., because large earmuff motions cause the cup to momentarily break its seal against the head), and the higher the peak pressure, the lower the IL [Figure 5]. In some other instances the non-linearity is favourable (i.e., because non-laminar flow causes an increase resistance to sound penetration through an orifice at high sound levels), and the higher the peak pressure of the impulses, the higher the IL [Figure 6]. From those examples, it is particularly clear that the actual attenuation provided at high levels cannot be inferred from REAT attenuation values.

Non-linear Earplugs

The perforated earplugs present an attenuation which increases with the peak pressure (the acoustic resistance through the orifice(s) increases with the peak level). The former non­linear plugs (RACAL Gunfender) acted nonlinearly only beyond 140 dB and the IL increased by about 5-10 dB for each 20 dB increase of the peak pressure of the impulse (NR peak attenuation increased by 10-15 dB from 130 to 190 dB: [Figure 7] (Forrest and Coles, 1970; Shaw, 1982). New designs by ISL (Hamery et al., 1997, Dancer and Hamery, 1998) allow the non-linearity to begin around 110 dB (i.e., below the potentially dangerous levels for this kind of impulses), and to get very large IL values for the highest peak pressures [Figure 6] (NR peak attenuation increases by 20-25 dB from 110 to 190 dB: [Figure 7]. Actually, to protect the ear against impulse noises, non-linear (perforated) earplugs are a very attractive solution. They are light, cheap, easy to clean and to maintain, they work without any energy supply and without intervention of the subject, and are compatible with other head equipment. Unlike classical plugs, they present a limited insertion loss at the low and moderate levels (especially at low frequencies where it may be near zero up to 1 kHz), and so allow speech communication, detection and localization of acoustic sources in about the same conditions as an unprotected subject [Figure 8]. All the same, they afford a protection adapted to occasional exposures to impulse noises such as those produced during training or combat (Dancer et al., 1992). Moreover, recent human studies by Johnson et al. (unpublished data) have demonstrated that these new earplugs are efficient (no significant TTS) for repeated exposures up to 187 dB peak (Friedlander waves, 1.5 ms A-duration, free field, normal incidence, 100 rounds), when they are well fitted. Some work is presently in progress in order to determine the best design to ensure a good and easy seal, as well as a good comfort when the earplug is worn during a long time. Such non-linear plugs will be soon used by the French armed forces. A good seal and a very satisfactory comfort index can be achieved with the help of custom moulded earplugs. The higher cost and the time required to take impressions and manufacture the custom ear-mould is obviously a major drawback but their use by long term volunteers might be favourably considered. Finally, to allow protection against continuous noise (during transportation in APC for example), the non-linear earplugs can be easily adapted to improve the passive attenuation characteristics at "moderate" levels.

Double Hearing Protection and Speech Intelligibility

The efficiency of a double protection is usually limited at the low frequencies by the coupling between the protectors and by the compliance of the skin of the external auditory canal and, at the high frequencies, by bone conduction (Schroeter and Poesselt, 1986). According to Gierke (1956), a double protection improves only by 10 dB on an average the IL from 125 to 8000 Hz. As a good protection can be afforded by (well-fitted) ordinary HP, the use of a double hearing protection (earmuff and earplugs) is not necessary in case of exposure to impulse noise (Dancer et al., 1992; Patterson and Johnson, 1996). Actually, the main interest of the double hearing protection is to protect the ear against loud continuous noise and to improve speech intelligibility.

Levels as high as 120-125 dBA can be measured in armoured vehicles (100-110 dBA in turboprop planes) [Figure 9]. As these noises contain a lot of low frequencies, a simple earmuff is not sufficient to achieve acceptable protection of the ear (Kusy, 1991): inside the Bradley vehicle the noise is about 114 dBA and the sound level at ear is 100 dBA with the current tanker helmet, therefore allowing only a daily exposure time of 20 minutes! Actually, most of the current earmuffs/headsets present a very poor attenuation at low frequencies. On [Figure 10] we can compare the passive attenuation obtained by two military headsets used in armoured vehicles: LE 132 (former French HP) and BOSE CVC (new US HP) when measured with the ATF under the same experimental conditions. The newly designed HP (BOSE CVC) is much better: for low frequencies, this improvement is mainly due to a new type of cushion. However, we must note that these attenuation curves correspond to optimal values. When used on soldiers, the attenuation values will drop because of fitting problems and of the possible use of glasses. Therefore, in practice for low frequencies the muff will show very little attenuation and the speech signals will be masked. The masking ruins the performances of the communication system and forces the subject to increase the speech level. It is then possible to measure a higher level under the muff than in the vehicle itself when the communication system is keyed [Figure 9]. Under these conditions the speech intelligibility is no more determined by the global attenuation of the HP but by its performance at the low frequencies only. The low frequencies mask medium and high frequencies and this masking effect is non-linear (the higher the level, the larger the spread and the amplitude of the masking). [Figure 11] allows a better understanding of this phenomenon. We can observe that it is not the attenuation performance of the HP at medium frequencies and hence the noise level under the pilot's headset at those frequencies (dashed line) which determines the speech level as adjusted by the listener, but the masking effect (solid thin line) due to the very low frequencies of the turboprop noise which enter the pilot's headset (Wessling, 1997). A better HP, as far as the medium and high frequencies are concerned, would not perform better with respect to the communications. The only solution is to improve the low frequency attenuation characteristics of the HP. As discussed before, it is possible to increase this attenuation but the performance of the system in real life will always fall short of the values measured in the laboratory.

In those circumstances, the use of a double hearing protection was recommended by Kryter long time ago (1946, 1955). The level difference between the speech signal and the noise is not modified over the whole frequency range but the earplugs worn under the earmuff bring both signals (noise and speech) down to a level at which the ear is not saturated and at which the masking by the low frequencies is less effective. In [Figure 12] we can see that it is always the masking effect due to the very low frequencies which determines the speech level but (i) this effect is smaller and (ii) the overall level at ear is reduced. We have shown that in this configuration the same speech intelligibility (CVC test) can be achieved with a reduction of 20 dB of the global level (speech + noise) at the level of the ear, therefore allowing a long term exposure and an improvement of the communication performance. This very simple and cheap method is usually well received by the subjects and gives results with are comparable to those obtained with Active Noise Reduction (ANR) techniques.

ANR Techniques and Speech Intelligibility

For the same reasons (limited attenuation at the low frequencies by the ordinary HP and speech masking problems), ANR devices are used in the military environment (McKinley et al., 1996). Nixon et al. (1992) summarized the main characteristics and possible applications of ANR. The present ANR earmuffs improve the attenuation by a maximum of 20 dB at the low frequencies (between 50 and 300 Hz) and so get an insertion loss comparable to a double hearing protection in this frequency range [Figure 13]. It is interesting to note that the HP which offer the BEST overall attenuation are not those with the active noise reduction! This emphasizes once more the importance of the passive characteristics of the HP at the low frequencies. [Figure 14] shows that in those circumstances and unlike what was observed with the double hearing protection, the characteristics of the HP at medium and high frequencies is the limiting factor for the speech intelligibility (the masking curve is then at a lower level). As a conclusion, one should say that a good ANR HP must be first a very good passive HP! However, all devices show an amplification up to 10 dB at frequencies near 1 kHz. This amplification can be quite disturbing because it amplifies residual noise in an area where the passive attenuation is not yet maximum and where the speech may be disturbed. As that effect is always situated just after the effective area of the active attenuation, this bandwidth has to be extended in order to shift the amplification area to a spectral part where it cannot impede the speech signal. The best way to achieve that , is to reduce the volume of the device and this leads directly from an earmuff to an earplug.

An experimental active earplug has been made by ISL. The size of this plug is about 2x2 cm. The "loudspeaker" is a specially designed PVDF or piezo-ceramic transducer. The microphone is a standard miniature microphone. Such ANR earplugs operate up to 3 kHz. In that way, the area where the residual noise is amplified is shifted to higher frequencies where the passive attenuation is very good. ANR earplugs represent a significant improvement especially for the intelligibility of speech (which is only marginally improved by the present ANR earmuffs at least because of their frequency restricted ANR performances) (Buck and Parmentier, 1997; Zimpfer and Buck, 1999) [Figure 15].

Most of the ANR-HP work as well for continuous as for impulse noise. However, their efficiency is limited by the output level of the electro-acoustic system (120 - 130 dB) and they are of little use for large impulses (Buck and Parmentier, 1996).

Last but not least, the use of digital filtering (Digital Signal Processor) (Zimpfer and Buck, 1999) will help to adjust the ANR systems to each user and/or to each noise exposure condition. ANR-HP would then be compatible with high quality speech intercom system (adaptive critical band filtering, neural networking) and even include a three-­dimensional virtual reality (Pellieux et al., 1992).

Up to now, only tank crews, helicopter and fighter pilots benefit of the ANR, but this technique which will be soon an integrated part of the much more sophisticated and comprehensive head equipment of the soldier.


This short review should help to draw the attention on the very specific problems of noise and hearing protection in military life, to avoid misinterpretations and/or negligence which are the cause of many hearing damages, and to indicate some solutions to reduce greatly the noise hazard while preserving the operational capabilities of the soldier.[47]


1American Conference of Governmental Industrial Hygienists, "TLV's threshold limit values and biological exposures indices for 1988-1989", Cincinnati, 1988.
2ANSI S12.42-1995, "Microphone-in-real-ear and acoustic test fixture methods for the measurement of insertion loss of circumaural hearing protection devices", American National Standards Institute, New York, 1995.
3ANSI S12.6-1997, "Methods for measuring the real-ear attenuation of hearing protectors", American National Standards Institute, New York, 1997.
4Berger, E.H., "Laboratory attenuation of earmuffs and earplugs both singly and in combination", Am. Ind. Hyg. Assoc. J., 44(5), 1983, pp 321-329.
5Berger, E.H., "Methods of measuring the attenuation of hearing protection devices", J. Acoust. Soc. Am., 79, 1986; pp 1655-1687.
6Buck, K. and Parmentier, G., "Evaluation of active hearing protectors with high level impulse noise", Proceedings Inter-Noise 1996, pp. 989-992.
7Buck, K. and Parmentier, G., "Active hearing protectors new developments and measurement procedures", Agard Conference Proceedings 596, 1997, pp 10-1/10-5
8Dancer, A.L., "Status and Shortcomings of Military Noise Standards", in "Noise-Induced Hearing Loss", Dancer, A.L., Henderson, D., Salvi, R.J., Hamernik, R.P., eds., Saint-Louis, Mosby Year Book, 1992, pp 513-520.
9Dancer, A., Grateau, P., Cabanis, A., Barnabe, G., Cagnin, G., Vaillant, T., Lafont, D., "Effectiveness of earplugs in high-intensity impulse noise", J. Acoust. Soc. Am., 91, 1992, pp 1677-1689.
10Dancer, A.L., Franke, R., "Hearing Hazard from Impulse Noise: a comparative study of Two Classical Criteria for Weapon Noises (Pfander Criterion and Smoorenburg Criterion) and the LAeq8 Method", Acta Acustica, 3, 1995, pp 539-547.
11Dancer, A., Buck, K., Parmentier, G. and Hamery, P., "The specific problems of noise in military life", Scand. Audiol. Supplementum, vol. 27, 1998, suppl. 48, pp. 123-130
12Dancer, A. and Hamery, P., "A new non-linear earplug for use in high-level impulse noise environment", J. Acoust. Soc. Am., 103, 5, 1998, pp 2878
13Forrest, M.R., Coles, R.R.A., "Problems of communication and ear protection in the royal marines", J. R. Nav. Med. Serv., 56, 1970, pp 162-169.
14Garinther, G., Kalb, J., Hodge, D., Price, G., "Proposed aural nondetectability limits for army materiel", US Army Human Eng. Lab 1985, Techn. Memorandum, pp 3-85.
15Gierke, H.E., "Personal protection", Noise Control, 2(1), 1956; pp 37.
16Hamery, P., Dancer, A. and Evrard, G., "Etude et realisation de bouchons d'oreilles perfores nonlineaires", ISL Report, 1997, R-128/97
17Hellstrom, P., "Objective methods for evaluating conventional, non-linear, and active hearing protector attenuation, in "Noise-Induced Hearing Loss", Dancer, A.L., Henderson, D., Salvi, R.J., Hamernik, R.P., eds., Saint-Louis, Mosby Year Book, 1992, pp 401-412.
18ISO TR 4869-3: 1989 (E), "Simplified method for the measurement of insertion loss of ear-muff type protectors for quality inspection purposes", International Organization for Standardisation, Geneva, 1989.
19ISO 1999, "Acoustics Determination of Occupational noise exposure and estimation of noise induced hearing impairment", International Organization for Standardisation, Geneva, 1990.
20ISO 4869-1, "Subjective method for the measurement of sound attenuation", International Organization for Standardisation, Geneva, 1990.
21Johnson, D.L., Patterson, J.H., "Rating of hearing protector performance for impulse noise", in "Proceedings Hearing Conservation Conference", University of Kentucky ed., 1992, pp 103-106.
22Kryter, K.D., "Effects of ear protective devices on the intelligibility of speech in noise", J. Acoust. Soc. Am., 18, 1946, pp 413-417.
23Kryter, K.D., "Speech Communication in Noise", Operational Applications Laboratory, Air Force Cambridge Research Center, Bolling Air Force Base, Washington, D.C., Rpt. AFCRC TR 54-52, 1955.
24Kusy, A., "Efficacite et confort des protecteurs individuels contre le bruit", Travail et Securite (INRS), 1991, pp 215­243.
25McKinley, R.L., Steuver, J.W., Nixon, Ch.W., "Estimated reductions in Noise-Induced Hearing Loss by application of ANR headsets", in "Scientific Basis of Noise-Induced Hearing Loss", Axelsson, A., Borchgrevink, H.,
26Hamernik, R.P., Hellstrom, P-A., Henderson, D., Salvi, R.J., eds., Thieme, New York, 1996, pp 347-360.
27NATO Research Study Group on the Effects of Impulse Noise, "Effects of Impulse Noise", NATO Document 1987; AC/243, Panel 8/RSG.6, D/9, pp 1-84.
28Nixon, Ch.W., McKinley, R.L., Steuver, J.W., McCavitt, A.R., "What can active noise reduction headsets do for you", in "Proceedings of the Hearing Conservation Conference", university of Kentucky ed., Cincinnati, University of Kentucky, BU156, 1992, pp 107-110.
29Paakkonen, R., Kyttala, I., "Attenuation of rifle and pistol noise with the use of suppressors and subsonic bullets", Acta Acustica, 2, 1994, pp 29-36.
30Parmentier, G., Franke, R., Buck, K., Kronenberger, G. and Evrard, G., "Tete artificielle pour l'evaluation de l'efficacite des protecteurs auditifs vis-a-vis de bruits de niveau eleve. II Application a differents types de bouchons d'oreilles et de serre-tete", ISL Report, 1994, R-113/94
31Parmentier, G., Dancer A., Buck, K., Kronenberger, G.and Beck C.., "Artificial head for evaluation of hearing protectors", Acta Acustica, Submitted
32Patterson, J.H., Mozo, B.T., Johnson, D.L., "Actual effectiveness of hearing protection in high level impulse noise", in "Noise as a public health problem", Vallet, M., ed., Bron, France, INRETS, vol. 3, 1993, pp 122-127.
33Patterson, J.H., Johnson, D.L., "Temporary Threshold Shifts produced by High-Intensity Free-Field Impulse Noise in Humans Wearing Hearing Protection", in "Scientific Basis of Noise-Induced Hearing Loss", Axelsson, A., Borchgrevink, H., Hamernik, R.P., Hellstrom, P-A., Henderson, D., Salvi, R.J., eds.,. Thieme, New York, 1996, pp 313-320.
34Patterson, J.H., "The effects of exposure of intense free­field impulse noise on humans wearing hearing protection: Implications for new criteria", J. Acoust. Soc. Am., 103, 5, 1998, pp 2877-2878.
35Pekkarinen, J.O., Starck, J.P., Ylikoski, J.S., "Hearing protection against high-level shooting impulses in relation to hearing damage risk criteria", J. Acoust. Soc. Am., 91, 1992, pp196-202.
36Pellieux, L., Piedecocq, B., Quandieu, P., "Stereophonic presentation of information: space simulation under earphones", Journal de Physique IV, Colloque C1, Suppl III, 2, 1992, pp 395-398.
37Peters, L., Garinther, G., "The effects of speech intelligibility on crew performance in an M1A1 tank simulator", US Army Human Eng. Lab, 1990, Techn. Memorandum, pp 11-90.
38Price, G.R., "An upper limit to stapes displacement: implications for hearing loss", J. Acoust. Soc. Am., 56, 1974, pp 195-197.
39Price, G.R., "Modeling the ear's response to intense impulses and the development of improved damage risk criteria", in "Proceedings Hearing Conservation Conference", University of Kentucky ed., 1992, pp 93-96.
40Price, G.R., Kalb, J.T., "Insights into hazard from intense impulses from a mathematical model of the ear", J. Acoust. Soc. Am., 90, 1991, pp 219-227.
41Schroeter, J., Poesselt, C.H., "The use of acoustical test fixtures for the measurement of hearing protector attenuation. Part II: Modeling the external ear, simulating bone conduction, and comparing test fixture and real ear data", J. Acoust. Soc. Am., 80, 1986, pp 505-527.
42Shaw, E.A.G., "The external ear", in "Handbook of Sensory Physiology/Auditory System" Keidel, W.D., Neff, W.D., eds., Berlin, Springer Verlag, vol. 1, 1974, pp 455­490.
43Shaw, E.A.G., "Hearing protector design concepts and performance limitations", in "Personal Hearing Protection in Industry", Alberti, P.W., ed., New York, Raven, 1982, pp 51-68.
44Steeneken, H.J.M., Verhave, J.A. and Houtgast, T., "Description of the STI-measuring method", 1994, TNO­Report TNO-TM 1994 I-2
45Wessling, T., "Erweiterung der Methode nach Houtgast und Steeneken zur Prognose der Sprachverstandlichkeit (sog. STI) fur Falle tieffrequenzen Larms hohen Pegels", Ruhr-Universitat Bochum, Lehrstuhl fur allgemeine Elektrotechnik und Akustik, Arbeitsbericht, June 16, 1997
46Zimpfer, V. and Buck, K., "Digital filters used in the feedback loop of an ANR earplug", Proceedings of the ASA-EAA meeting, Berlin 1999, J Acoust. Soc. Am., 105: 973
47Zwicker, E. and Feldkeller, R., "Das Ohr als Nachrichtenempfanger", Hirzel Verlag, Stuttgart, 1967