[0001] Most modern portable electronic devices have relatively small physical dimensions
in order to facilitate portability. A perfect example of such a portable electronic
device is a mobile terminal, such as a mobile phone which has during the past decade
become smaller than ever.
[0002] One consequence of the relatively small size of the devices is that there is much
less space available in the device, for example, for the loudspeaker, than in the
past.
[0003] The size of the vibrating membrane and properties of other components build up the
acoustical characteristics of a loudspeaker. As discussed in German patent application
DE 195 48 149 A1, lower frequency tones cause larger membrane movements than tones
of higher frequency. Depending on the sound pressure that would be obtained, an overload
may even damage the loudspeaker. A solution proposed in DE 195 48 149 A1 is that the
loudpeaker is equipped with a high-pass filter, because the vibrations in the membrane
are larger at lower frequencies. Similar approach has been presented in EP 1 135 002
A2, where high-pass filtering is used when the volume level exceeds a medium level.
[0004] A problem of using a simple high-pass filter is its adverse effect on the sound quality.
It is an objective of the present invention to bring about a speaker driving circuit,
by using which it is possible to output audio signals where the part of the signal
spectrum including the resonance frequency of the loudspeaker are either attenuated
or not. This can be achieved by using the speaker driving circuit as described in
independent patent claim 1. An advantage of such a speaker driving circuit is that
some signal types can then be played back by using a higher volume without damaging
the loudspeaker than would be possible without the speaker driving circuit.
[0005] Another objective of the present invention is to bring about a mobile terminal comprising
such a speaker driving circuit. An advantage of such a mobile terminal is that then
in the mobile terminal it is possible to filter the signal to be feed to the loudspeaker
in different manners, i.e. to reduce the signal component including frequencies at
the resonance frequency of the loudspeaker of the mobile terminal.
[0006] An aspect of the present invention is to use a high-pass filter before outputting
a signal representing a ringing tone by a loudspeaker. The high-pass filter has a
cut-off frequency at or above a resonance frequency of the loudspeaker. The advantage
of such a solution is that the damaging of the loudspeaker can be avoided, and nevertheless
the ringing tone can be output with a relatively high volume.
[0007] A further objective of the present invention is to enable the using of some ringing
tones without damaging the loudspeaker. This can be achieved by using a high-pass
filter before outputting a signal representing a ringing tone by a loudspeaker as
descibed in independent patent claim 7.
[0008] Dependent claims describe various embodiments of the present invention.
[0009] In the following, the invention and its preferred embodiments are disclosed in more
detail by referring to the examples shown in the accompanying drawings 1 to 4, of
which:
Figure 1 shows a speaker driving circuit according to the present invention;
Figure 2 illustrates in more detail the band-pass filter and the high-pass filter;
Figure 3 shows the resulting frequency response of the speaker driving circuit when
the band-pass filter is used; and
Figure 4 shows the resulting frequency response of the speaker driving circuit when
the high-pass filter is used.
[0010] Figure 1 shows a speaker driving circuit 10 according to the present invention. The
speaker driving circuit 10 includes a band-pass filter 109 and a high-pass filter
111. The band-pass filter 109 has cut-off frequencies outside GSM voice band, i.e.
at 300 Hz and at 3.4 kHz, and it is mostly used in order to attenuate out-of-band
noise caused by the audio front end 103.
[0011] The audio front end 103 includes an amplifier 173 and three inputs (INPUT1, INPUT2,
INPUT3). Further there are a first and a second control channel (CONTROL CHANNEL A,
CONTROL CHANNEL B) which are used to give input to a central processing unit CPU 171
of the audio front end 103. The CPU 171 operates then responsively to signal obtained
through CONTROL CHANNEL A and B three switches SW1, SW2, SW3 used in selecting which
INPUT1, INPUT2, and INPUT3 is used for amplifying.
[0012] The output of the amplifier 173 is then connected to a loudspeaker 101 for the reproduction
of an audible sound.
[0013] Typically, for a mobile terminal, speech information SPCH is received via a radio
modem 151. The modem 151 passes the speech information SPCH received to a processor
unit 105. The speech information SPCH is usually coded using some coding, mostly for
bandwidth efficiency and error recovery purposes. The speech information SPCH has
then to be decoded in a codec 121 which corresponds to a soft- or hard-coded algorithm
for obtaining a suitable sampling which can be used for speech reproduction. This
signal SIG1 which will be used for speech reproduction will be outputted from the
codec 121 to a preamplifier 107.
[0014] If there is a mobile terminating call coming, the mobile network under which the
mobile terminal is roaming starts paging the mobile terminal. The mobile terminal
then receives a paging signal RNG via the same radio modem 151. This is handled in
the processor 105 in a similar manner as the speech information SPCH but instead of
passing the paging signal RNG to a codec 121 it activates a ringer 123. The ringer
123 gives a signal SIG2 representing ringing tone which is then further passed to
the preamplifier 107.
[0015] The signal SIG2 representing a ringing tone is preferably a signal having a substantially
square waveform in the time-voltage space. This corresponds to connecting the output
of a voltage source on and off for a short period in the manner usually done in a
buzzer. Also other signal forms for the signal SIG2 representing a ringing tone can
be used.
[0016] The preamplifier 107 gives from its output (PREAMP OUTPUT) a preamplified signal
SIG. Typically, for a mobile terminal there will be generated no signal SIG2 representing
a ringing tone when there is an ongoing call so that speech information SIG1 is fed
to the preamplifier 107.
[0017] According to one aspect of the present invention, the preamplified signal SIG is
passed to the band-pass filter 109 and to the high-pass filter 111. The band-pass
filter 109 is connected to a first input INPUT1 and the second high-pass filter is
connected to a second input INPUT2 of the audio front end 103. The signal SF1 received
from the band-pass filter 109 is different from the signal SF2 received from the high-pass
filter 111 in the sense that because now the cut-off frequency of the high-pass filter
111 is above the resonance frequency of the loudspeaker 101.
[0018] The main purpose of the high-pass filter 111 is to remove the lowest frequency part
of the signal. This is particularly advantageous when the signal to be amplified in
the amplifier 173 is such a signal as the signal SIG2 representing a ringing tone,
especially when the ringing tone is created by a buzzer or has otherwise a substantially
square waveform in time-voltage space.
[0019] The lowest frequency part carries most energy of the signal. Further, it causes the
membrane of the loudspeaker 101 to vibrate too much therefore possibly damaging the
loudspeaker especially when the ringer volume is high.
[0020] One reason for the listener'not observing the deterioration of the sound quality
can be explained by the results of psychoacoustic theory, also known from the example
of the missing fundamental, which was observed by Harvey Fletcher already in 1924.
For a sound with harmonic partials to be heard as a musical tone, its spectrum must
include three successive harmonics of a common frequency. The pitch is given by that
common frequency, whether or not it is present in the musical tone. (Reference on
page 58 in Music, cognition, and computerized sounds: an introduction to psychoacoustics,
editor Perry R. Cook, ISBN 0-262-03256-2).
[0021] Because the interval relationship of the upper harmonic remains the same regardless
of the high-pass filter 111, the tone can still be perceived as having the original
pitch.
[0022] In the example of Figure 1 audio front end 103 comprises the switches SW1, SW2 and
SW3 used to select the correct input. However, according to another aspect of the
present invention, instead of selecting between filtered signals SF1, SF2, i.e. selecting
after the band-pass filter 109 and high-pass filter 111, a selection can be made earlier
in the signal path. In other words, one can select between the filters 109 and 111.
[0023] In both embodiments, the selection can be made based on the signal type or class.
For example, all ringing tones or signals from a specific signal source, such as an
FM radio, are all led to the high-pass filter 111 whereas all other signals are led
to the band-pass filter 109. Also such a solution where a digital sound processing
apparatus is used, based on the low-frequency content of the signal (such as SIG1)
the selecting means (SW1, SW2) are operated to select the most suitable high-pass
filter.
[0024] The selection can also be made responsively to the state of the portable electronic
device. Examples of such states are "ongoing call", and "paging".
[0025] Figure 2 illustrates the band-pass filter 109 and the high-pass filter 111. Both
filters 109, 111 are connected to the preamplifier 107 output.
[0026] The band-pass filter 109 is made up by a high-pass filter and a low-pass filter and
has resistors R
1, R
5 and R
6, and capacitances C
1, C
2, C
3 and C
4 for selecting the cut-off frequency.
[0027] The high-pass filter 111 has a higher cut-off frequency than the band-pass filter
109. The cut-off frequency is higher than the resonance frequency of the loudspeaker.
The high-pass filter 111 comprises resistors R
2, R
3, and R
4, and capacitances C
5 and C
6. Possible values for a loudspeaker 101 (type = SK MS605) used in one mobile terminal
model can easily be calculated. A person skilled in the art understands how to modify
these values if the loudspeaker 101 is changed to another one.
[0028] Figure 3 shows the resulting frequency response of the speaker driving circuit 10
when the band-pass filter 109 is used; and Figure 4 shows the resulting frequency
response of the speaker driving circuit 10 when the high-pass filter 111 is used.
One sees readily the effect of the high-pass filtering performed by the high-pass
filter 111: the lower part of the voice band of GSM which in Figure 3 was more or
less linear is now highly attenuated.
[0029] The band-pass filter 109 and the high-pass filter 111 can be implemented as a digital
filter as well. In this case the filtering will most conveniently be performed in
the processor unit 105.
[0030] Although the invention has been described above with reference to the examples shown
in the appended drawings, it is obvious that the invention is not limited to these
but may be modified by those skilled in the art without difference from the scope
and the spirit of the invention. Instead of having a GSM terminal, a CDMA, WCDMA,
GPRS, UMTS or any other suitable mobile terminal can be used. In addition to mobile
terminals, the invention can be included in any other portable electronic device,
such as a portable radio, CD- or MP3 player.
1. A speaker driving circuit (10) for processing an audio signal (SOUT) to be outputted
by a loudspeaker (101) having a resonance frequency,
comprising:
- means (107) for receiving a signal from a signal source (121, 123);
- a band-pass filter (109) and a high-pass filter (111) for filtering the signal received
to a first and a second filtered signal, whereby the band-pass filter (109) has a
lower cut-off frequency at or lower than the resonance frequency of the loudspeaker,
and the high-pass filter (111) has a cut-off frequency higher than the resonance frequency
of the loudspeaker; and
- means (105) for selecting the first or the second filtered signal for amplifying
the filtered signal or for selecting the band-pass filter (109) or the high-pass filter
(111).
2. A speaker driving circuit (10) of claim 1, wherein: said means (105) for selecting the first or the second filtered signal for amplifying
the filtered signal or for selecting the band-pass filter (109) or the high-pass filter
(111) are adapted to select the first filtered signal for a signal from a first signal
source (121) and the second filtered signal for a signal from a second signal source
(123).
3. A speaker driving circuit (10) of claim 2, wherein: the first signal source (121) is a voice codec and the second signal source (123)
is a ringing tone generator.
4. A speaker driving circuit (10) of claim 1, 2, or 3, wherein: said means (105) for selecting the first or the second filtered signal for amplifying
the filtered signal or for selecting the band-pass filter (109) or the high-pass filter
(111) are responsive to a type of the signal received from the signal source (121,
123).
5. A speaker driving circuit (10) of claim 4, wherein: at least one of said signal types include a ringing tone.
6. A speaker driving circuit (10) of claim 5, wherein: said ringing tone has a substantially square-like waveform in time-voltage space.
7. The use of a high-pass filter (111) before outputting a signal (SIG2) representing
a ringing tone by a loudspeaker (101), characterized in that: said high-pass filter (111) has a cut-off frequency at or above a resonance frequency
of the loudspeaker (101).
8. The use of a high-pass filter (111) according to claim 7, wherein: said high-pass filter causes an attenuation of over 3 dB at 300 Hz.
9. The use of a high-pass filter (111) according to claim 7 or 8, wherein: said signal (SIG2) has a substantially square waveform in time-voltage space.
10. A mobile terminal comprising: a speaker driving circuit (10) according to any one of the claims 1 to 6.