Main Index | Feedback

TETRA Transmission planning

Loudness, Shock and Stability

Background

Private digital networks should aim for a Send Loudness Rating (SLR) of +7dB and Receive Loudness Rating (RLR) of +3dB [ETS300283, Section 6]. GSM networks have nominal values for SLR of 8dB and RLR 2dB [GSM 03.50]. These differences are due to the environment in which GSM equipment is generally used where users tend to speak louder than they would on a conventional telephone. DECT telephones [ETR041] use the conventional values for SLR and RLR [ETS300283].

In the case of mobile operating environments with a high background noise the user can speak extremely loudly resulting in clipping of the audio. The solution in these cases is to reduce the sensitivity of the microphone (increase the SLR) and raise the volume in the loudspeaker (reduce the RLR). GSM [GSM03.50 Annex E] suggests a set of values for adaptive gain control in a noisy environment (see below). It does not suggest a method of establishing the background noise level but measuring the average signal power from the microphone should give a good indication of the background noise. This method should also tend to reduce the effect of loud talkers on the transmitted audio.

Measuring the SLR and RLR will be difficult using fixed tones and artificial speech [P.50] should be used when making end to end measurements. Tones should be suitable when interfacing at the linear PCM point within the handset.

Handset terminals

Portable terminals and handset mobile terminals should have the following nominal values when in duplex calls.

If there is a user controllable volume control then the nominal RLR shall be met for at least one setting on the control. With the control at maximum the RLR shall not be less than -13dB and at minimum the RLR shall not be more than 18dB.

In non duplex modes where the user holds the handset away from the head when receiving the RLR may be further lowered but care should be taken to ensure that the user acoustic shock figures cannot be exceeded in normal use.

Both sending and receiving sensitivity shall be measured at each of the 14 frequencies given in [P.79, table 1, bands 4 - 17]. The sending sensitivity is expressed in terms of dBv/Pa and the SLR shall be calculated according to [P.79, formula 2.1] over bands 4 to 17 and using m = 0.175 and sending weighting factors from [P.79 table 1].

The artificial ear sensitivity shall be corrected by the factor given in [P.79, table 2].

Handsfree terminals

Handsfree terminals should have the following nominal values when in duplex calls. These figures are different to those for handset operation because of the position of the user in relationship to the microphone and loudspeaker.

If there is a user controllable volume control then the nominal RLR shall be met for at least one setting on the control. The volume control should allow the user to raise the volume by at least 15dB (RLR drops by 15dB) from nominal to compensate for the noisy environment in a car.

In non duplex modes there are no restrictions on loudspeaker volume apart from acoustic shock.

The sending sensitivity is measured according to [GSM 03.50, Annex C.1.2] and the receive sensitivity according to [GSM 03.50, Annex C.1.4].

Headset terminals

Terminals with headsets should have the following figures

Receive volume controls should have a range of +/-6dB.

GSM [GSM 03.50] only defines measurements according to [P.38] for supra-aural earphones. Other types are yet to be defined.

Acoustic Shock

The Low Voltage Directive (LVD) [73/23/EEC] is used to set requirements for the prevention of acoustic shock. The essential requirements can be met by ensuring that the levels described below are not exceeded. These levels assume the use of an artificial ear conforming to [P.57].

Continuous signal

The handset/headset is mounted with the artificial ear and a digital signal generator fed into the Linear PCM Reference Point within the terminal. The signal generator is set to generate a digitally encoded equivalent of a square wave with a peak code equal to the maximum code which can be sent by the PSTN. Tones are generated on each of the third octave frequencies in the range 200Hz to 4KHz. For each frequency the sound pressure level in the ear is measured and shall not exceed +24dBPa(RMS).

This figure should be met for the local call arrival signalling (ringing). The user should not be subjected to a level exceeding this figure with the handset in any of its normal operating positions.

Peak Signal

Method TBD. The peak signal in the artificial ear shall be limited to +36dBPa. This test covers short intense noises such as clicks and pops.

Stability loss

If the degree of coupling between microphone and loudspeaker is too high acoustic feedback can result in howling of the terminal.

Stability loss should exceed 6dB for all frequencies in the range of 200Hz to 3.6KHz and meet the principles of the requirements as defined by [G.122 sections 2 and 3]. This requirement is in accordance with [ETS300283, section 5]. All tests should be performed at maximum loudspeaker volume and echo control switching effects should be taken into account.

Handsets shall be tested face down on a surface which shall be one of three perpendicular plane, smooth, hard surfaces forming a corner. Each surface shall extend 0.5M from the apex of the corner. The handset shall be placed 250mm from the corner as shown in the disgram below.

Handset reference position for stability loss test"

Adaptive gain control

In noisy surroundings the speaker automatically raises their voice to compensate for the noise. The sending sensitivity can therefore be reduced, giving the same voice output to the line as under quiet conditions and at the same time reducing the output line noise level.

Under the same circumstances the loudspeaker volume has to be raised to give a better perceived sound quality.

This adjustment can be performed automatically by implementing an adaptive gain control which reduces the microphone sensitivity and increases the loudspeaker volume in a symmetrical manner.

GSM [GSM 03.50] presents three charactistics for guidance.

Adaptive gain with a single threshold

Adaptive gain with double threshold

Adaptive gain with continuous variation

Copyright ©2002 Andy Fletcher andy@x31.com