Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
37
PATENT
IMPROVEMENT IN AN APPARATUS AND METHOD
FOR HANDS FREE TELEPHONIC COMMUNICATION
BACKGROUND OF THE INVENTION
FIELD OF THE INVENTION
The invention relates to the field of telephonic
communication and, in particular, to hands-free operation
wherein a speaker phone is utilized.
DESCRIPTION OF THE PRIOR ART
Conventional telephonic conversations occur
through telephone handsets or headsets wherein the audio
transmitter and audio receiver are acoustically decoupled.
In the case of a conventional telephone handset, the
acoustical coupling between the earpiece and mouthpiece,
given the respective audio volumes produced at the earpiece
or mouthpiece, are such that the amount of feedback
introduced into the communicational loop is de minimis. In
other words, in a conventional handset the audio volume
received at the earpiece is not picked up at the mouthpiece
at a magnitude great enough to form an interfering feedback
loop in the two-way telephone conversation. In the case of
a conventional headset, the acoustical separation between
the earpiece and mouthpiece is even greater since the
earpiece generally is tightly fitted to the wearer~s ear and
~I Z~3~31~7
connected to the mouthpiece only through a thin plastic
tube.
However, in the case of a hands-free or speaker
phone operation, the audio signal intensity produced by the
speaker is generally within the same order of magnitude as
the audio signals produced by the user at the microphone
pick-up. This is true even in the case where the speaker
and microphone are physically separable, since the intensity
of the audio signal from the speaker must be great enough to
be easily heard by the user, and the microphone must be
sensitive enough to pick up audio signals at reasonable
audio voice levels. In order to avoid the inherent feedback
which would otherwise occur, hands-free speakers contain
automatic voice muting circuitry. Thus, when an audio
signal is being produced at the speaker, the microphone is
disabled. In addition, in some units, if an audio signal
beyond a predetermined threshold is picked up at thé
microphone from the user, the audio signal to the speaker is
cut off. Therefore, it is generally impossible in an hands-
free operation to have any degree of overlap in
conversations. Furthermore, the switchover or muting
operation generally takes a few tenths of seconds to
operate, which makes rapid interchange of conversation
between the caller and receiver difficult.
More significantly, the use of hands-free
operation must occur in a relatively quiet environment. Any
background noise of significant magnitude will serve to
~ 7
activate the muting operation, thereby substantially
interfering with, or disabling, the conversation of one of
the parties. Hands-free speaker phone operation cannot
therefore be used in noisy environments.
what is needed then is some means whereby hands-
free operation of a telephone may be practiced in a noisy
environment wherein loud distant noises can be distinguished
from closer sounds, such as the speaker's voice.
BRIEF SUMMARY OF THE INVENTION
The invention is an improvement in a method for
practicing hands-free two-way communication comprising the
steps of modulating audio signals at a first carrier
frequency to produce a modulated upper sideband and
modulated lower sideband output; removing the upper sideband
output; demodulating the lower sideband output~at a second
carrier frequency to produce a demodulated upper sideband
output and demodulated lower sideband output, the second
carrier frequency offset from the first carrier frequency by
a predetermined amount: and removing the demodulated upper
sideband output. As a result of this combination of steps
interferins feedback by multiple transmissions through the
two-way communication of an audio signal is substantially
eliminated.
The improvement further comprises the step of
limiting the audio signal to frequencies below the first or
~ 7
second carrier frequencies or both prior to the step of
modulating.
The method further comprises the steps of
receiving an acoustic audio signal at a pair of
approximately adjacent microphones to generate a
corresponding pair of electrical audio signals, and
differentially amplifying the electrical audio signals to
generate the input audio signal. As a result the acoustic
audio signals originating near the pair of microphones are
distinguished from acoustic audio signals generated
distantly rrom the pair of microphones and the distant
acoustic audio signals are substantially attenuated.
The step of receiving the acoustic audio signals
at the pair of microphones is such that the signals are
received preferentially along a predetermined direction.
The steps of modulating and demodulating the first
and second carrier frequencies are offset relative to each
other by an amount in the range of O to 500 Hz.
The invention can also be characterized as an
apparatus for hands-free operation in two-way communication
comprising a modulator for receiving an input audio signal
and a first carrier oscillator coupled to the modulator.
The first carrier oscillator generates a first carrier
frequency. The input audio signal is mixed with the first
carrier frequency to produce a modulated upper sideband
output and lower sideband output of the modulator. A first
filter removes the modulated upper sideband output. A
demodulator receives the modulated lower sideband output. A
second carrier oscillator generates a second carrier
frequency. The second carrier frequency is coupled to the
demodulator. The demodulator generates a demodulated upper
sideband output and demodulated lower sideband output. The
first and second carrier frequency are relatively offset
from each other by a predetermined amount. A second filter
removes the demodulated upper sideband output. As a result
interfering feedback within the hands-free two-way
communication is eliminated.
The apparatus further comprises a third filter for
removing all frequencies above the first and second carrier
frequencies. The third filter has its output coupled to the
input of the modulator. The input audio signal appears at
the output of the third filter.
The apparatus further comprises a pair of
approximately adjacent microphones for producing a
corresponding pair of audio signals, and a differential
amplifier having one input coupled to each of the
microphones for receiving the pair of audio signals. The
differential amplifier produces the input audio signal.
Therefore, acoustical sources near the pair of microphones
are distinguished from acoustical sources relatively distant
from the pair of microphones. The signals which are from
the distant acoustical source are attenuated.
In tne illustrated embodiment each of the
microphones is a directional microphone.
37
I The invention is still further characterized
i as an improvement in a method for hands-free two-way
¦ communication between a first and second station. The
improvement comprises the step of frequency shifting
the audio signal sent from the first station to the
second station by a predetermined amount relative to
the audio signal originating at the first station.
Thus, interfering feedback and hands-free communica-
tion at the first station is substantially eliminated.
BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1 is a diagrammatic depiction of
circuitry incorporating the invention whereby hands-
free operation in a noisy environment may be utilized.
Figure 2 is a diagrammatic depiction of an
overall system_in which the two-way hands free com-
munication of the invention is illustrated.
The invention and its various embodiments
may now be understood by turning to the following
detailed description.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
Hands free two-way communication is achieved,
without the necessity of selectively muting either the
sent or received signal and without the generation of
interfering acoustical feedback between the speaker
and microphone at the hands-free station, by frequency
offsetting the audio signal sent from the hands-free
station. The audio signal
I
-- 6 --
3 ~ 7
is frequency offset by first modulating the audio signal at
a first carrier frequency to produce modulated upper and
lower sideband outputs. The modulated upper sideband output
is filtered out, and the modulated lower sideband output is
coupled to the input of a demodulator. The modulated lower
sideband output is then demodulated at a second carrier
frequency offset from the first carrier frequency by a
predetermined amount. A demodulated upper sideband output
and lower sideband output is thus generated. The
demodulated upper sideband output is filtered out, and the
demodulated lower sideband output is provided for
transmission or broadcast in the two-way communication. The
demodulated lower sideband output is identical to the
original input audio signal with the exception that it has
been frequency offset by an amount equal to the
predetermined amount of frequency offset between the first
and second carrier frequencies. In the event of cross-talk
or any feedback within the two-way communication loop, each
transmission of the information is thus frequency offset,
and the establishment of interfering feedback loops is
thereby avoided.
A pair of directional microphones are used to pick
up a user's voice. The signal from each of the directional
microphones is amplified in a differential amplifier so that
distant signals are substantially cancelled while nearby
audio signals are amplified. The nearby audio signals, now
transformed into an electrical signal, is further amplified
~ 7
through an audio amplifier, and coupled to the input of a
double balanced modula.or. The audio signal is mixed with a
carrier oscillator, and appears in both upper and lower
sidebands. The output of the double balanced modulator,
containing the user's voice in upper and lower sidebands, is
then coupled to a low pass, three stage filter. The lower
sideband is transmitted to a double balanced demodulator.
The lower sideband is then demodulated at a slightly offset
carrier frequency to produce a demodulated lower and upper
sideband signals, but offset by the difference in frequency
of the carrier oscillators in the demodulator and the
modulator. The output of the demodulator is then coupled to
a low pass filter. The lower sideband, offset by the
difference in the carrier frequencies, is provided at the
output of the filter. This signal is then available for
transmission.
A speaker in the vicinity of the pick-up
microphones produces an audio signal which is received by
the microphones. However, the signal, which is ultimately
generated by the above-described circuitry, is frequency
offset from that produced by the speaker by predetermined
amount. Any amount of crosstalk or cross coupling within
the telephone communication line therefore fails to set up a
regenerative feedback since a signal, recirculating through
the communication loop, is frequency offset during each
cycle. Interfering feedback therefore is never established.
There need be no muting circuitry within the hands-free
~ 7
communication system. Therefore, the occurrence of a loud
extrinsic noise does not serve to interfere or cut off
transmission of signals through the microphone or the
speaker.
In the presently illustrated embodiment, the
hands-free operation is described a connection with a
roadside emergency callbox utilizing a hands-free telephone
module which communicates through a remotely powered
cellular radio telephone to conventional telephone land
lines, as described in greater detail in connection with
copending application entitled Apparatus and Method for a
Cellular Freeway Emergency Telephone Service, filed Nov. 25,
1985, Serial No. 801,410.
Turn now to the diagrammatic depiction of the
circuitry as set forth in Figure 1. Directional microphones
10 and 12 are conventional, and are mounted within the face
of a roadside speaker box (not shown) at or near mouth
level. Directional microphones 10 and 12 are placed
immediately next to each other, or at least in close
proximity to each other. Therefore, acoustical wavefronts
impinging upon the speakers from nearly objects are likely
to have dramatically different acoustical phases inasmuch as
the distance of separation between microphones 10 and 12 is
a large fraction of the total distance between the average
position of the microphones and the source of the acoustical
wavefront. On the other hand, acoustical wavefronts
originating at distant sources are substantially more planar
~ 7
when arriving at microphones 10 and 12, and have very nearly
the same phases at each of the microphones, since the
distance of separation between microphones 10 and 12 is then
an insignificant fraction of the total distance between the
microphones and the distant source. The outputs of
microphones 10 and 12 are coupled, throu~h appropriate
buffer circuitry, which is diagrammatically depicted by
variable resistor 14, to the inputs of a differential
amplifier 16. Since the signals from distant sources will
have nearly the same phase, they will be cancelled within
differential amplifier 16. On the other hand, signals
received from relatively nearby sources will have
dramatically different phases, and thus produce an amplified
audio output at output 18 of differential amplifier 16.
Output 18 of amplifier 16 is coupled to a low pass filter
20. Low pass filter 20 substantially filters out all
frequencies above a first~predetermined frequency. The
first predetermined frequency, for reasons described below,
is set slightly below a first carrier oscillator frequency,
also more completely described below.
For the purposes of illustration and clarity,
assume that the first predetermined frequency is 3.5 kHz.
Therefore, low pass filter 20 filters out all frequencies
above 3.5 kHz and freely passes frequencies below 3.5 kHz.
The first carrier oscillator frequency, in this illustrative
example, is 4.0 kHz.
~ 7
The output of low pass filter 20 is coupled to the
input of an audio amplifier 22. Audio amplifier 22 is a
mutable amplifier having a DC input 24, which can be
selectively activated by the remote receiver for purposes of
communication control. The muting capability of amplifier
22 does not relate to or affect the hands-free operation.
The output of amplifier 22 is coupled in turn to
the input of a double balanced modulator 26 which, in the
illustrative embodiment, is based upon switched capacitor
filters, and in particular is a balanced modulator model Mo.
MC1596, manufactured by Motorola Corporation. Balanced
modulator 26 is coupled to a first carrier oscillator 28,
which operates in the illustrative embodiment at 4.0 kHz,
The audio signals input to balanced modulator 26 are thus
mixed with the carrier frequency, and upper and lower
sidebands are created at the output of balanced modulator
26.
The upper and lower sidebands, representing the
filtered audio signal from microphones 10 and 12, are then
coupled to the input of a three stage, low pass filter 30
which sharply and completely filters out all the upper
sideband frequencies. For example, in the illustration, low
pass filter 30 will have a sharp cutoff at 3.5 kHz of
approximately 24-26 db per octave.
The audio signals are now presented at the output
of low pass filter 30 as lower sideband signals only carried
on the 4.0 kHz first carrier frequency of oscillator 28.
~ 7
These lower sideband signals are then coupled to the input
of a double balanced demodulator 32. In the illustrative
embodiment, demodulator 32 is a demodulator model No.
MC1536, manufactured by Motorola Corporation. Demodulator
32 is coupled to a second carrier oscillator 34, which is
operating at a second carrier frequency slightly offset from
the first carrier frequency. Again, in the illustrative
embodiment, the carrier frequency of oscillator 34 is
assumed to be 4.1 kHz. The output of demodulator 32 thus
represents the demodulated conversion of the modulated
output signals from modulator 26. However, since the
carrier frequency of demodulator 32 is slightly offset from
the carrier frequency of modulator 26, the demodulated
signals form upper and lower sidebands are offset by a
similar frequency offset.
The output of demodulator 32 is coupled to a low
pass filter 36 which filters out the upper sideband of the
demodulated signals. The lower sideband represents the
demodulated offset signals which correspond, subject only to
the frequency offset of the carriers, to the frequencies of
the original audio signals received by microphones 10 and
12. The output of low pass filter 36 is coupled to gate 38
which is controlled by a mute signal input 40. The output
of gate 38 is a logically controlled offset audio signal
available for transmission or broadcast according to
conventional means. As in the case with audio amplifier 22,
gate 38 is controlled by remote commands received according
12
~ 7
to considerations which are irrelevant to hands-free
operation.
The circuit architecture having been described in
connection with the diagrammatic depiction of Figure 1 can
better be understood by now considering the operation of the
circuitry of Figure 1 in the context of a simple example.
Assume that audio frequencies between the range of 1,000 and
2,000 Hz are generated by a user at the roadside callbox at
microphones 10 and 12. These near signals are thus
amplified by differential amplifier 16, and pass through low
pass filter 20. ~hey are amplified by audio amplifier 22
presented as inputs to balanced modulator 26. The output of
balanced modulator will produce both the sum and difference
between the input and carrier signals at its output. For
example, a 2 kHz and 1 kHz signal to input of balanced
modulator 26 will produce the sum frequencies 6 kHz and 5
kHz at the output, as well as the difference frequencies, 2
kHz and 3 kHz. The upper sideband is represented by 5 and 6
kHz signals, &~ the lower sideband is represented by 2 and 3
kHz signals. If a signal greater than the carrier
frequency, such as a 6 kHz signal, were presented to the
input of balanced modulator 26, the sum and difference
signals would be a 2 kHz and 10 kHz signal. In this case
there would be no way in which to distinguish the 2 kHZ
signal produced from the difference between the 6 kHz signal
and the carrier, and the 2 kHz signal produced between the
difference between the 2 kHz signal and the carrier. This
~ 7
frequency mixing would lead to information garbling, and is
thus avoided by filtering out all signals near, at or above
the carrier frequency. In the illustrated embodiment, a
margin of safety and clear separation between the upper and
sidebands are created by filtering out all frequencies above
3.5 kHz.
Low pass filter 30 is thus used to filter out the
5 and 6 kHz upper sideband signals and to pass the 2 and 3
kHz lower sideband signals. The lower sideband signals are
thus presented to the input of demodulator 32. However, in
demodulator 32, the 2 and 3 kHz sideband signals are mixed
with a 4.1 kHz carrier signal. Thus, the output of
demodulator 32 is comprised of the sum frequencies 7.1 kHz
and 6.1 kHz and the difference frequencies, 2.1 and 1.1 kHz.
The 6.1 and 7.1 kHz signals thus represent the upper
sideband while the 2.1 and 1.1 kHz signals represent the
lower sideband.
Again, the upper sideband signals are filtered out
by low pass filter 36. The resulting audio signals coupled
through gate 38 are signals at 1.1 and 2.1 kHz. These are
in fact the initial audio signals we have assumed were
received by microphones 10 and 12, but offset by the amount
of difference in the carrier frequencies between carrier
oscillator*28 and 34.
Thus, the listener or receiver at the other end of
microphones 10 and 12 will hear the speaker's voice offset
by small frequency differential. However, the frequency
~ 7
offset is not sufficient to lose any significant
intelligibility, and in the case of an emergency roadside
callbox, there is no need or desirability to reproduce the
speaker's voice with high fidelity. On the other hand, when
the receiver transmits to the speaker at the callbox, his
voice is transmitted through the callbox at a nearby speaker
at a volume sufficient to be heard at roadside noise levels,
and therefore at a volume of the same order of magnitude as
the caller himself.
The remote receiver's voice is thus picked up upon
microphones 10 and 12. The signal is amplified by the
circuitry of Figure 1 with the same degree as any other
nearby audio signal. However, the receiver's voice is
offset by the difference between the carrier frequencies of
oscillators 28 and 34. Therefore, any crosstalk which
occurs at any point within the telephone communication loop
is rapidly escalated in frequency as it ls multiply
transmitted through the communication loop with the result
that no interfering feedback is established.
Particular note should be made of the fact that at
no point within the circuitry of Figure 1 is it necessary to
mute the operation of microphones 10 and 12 or any nearby
speaker in order to utilize the hands-free operation. AS
the caller is using the roadside box in hands-free
operation, the passage of a loud, nearby truck or other
noise will, firstly, tend to be ignored by the directional
nature of microphones 10 and 12, and, secondly, tend to be
J ~ 7
cancelled, at least when relatively distant, by the
differential amplification of the output of microphones 10
and 12. In any case, the passing of a large, nearby noise
in direct line with microphones 10 and 12 will not serve to
cut or mute the operation of any nearby speaker, and the
caller will be freely to communicate with the remote
receiver without interference. At the same time, the remote
receiver will be able to clearly communicate with the caller
in a hands-free mode without the creation of interfering
feedback loops.
Many modifications and alterations may be made by
those having ordinary skill in the art without departing
from the spirit and scope of the invention. For example, in
the illustrated embodiment above, for simplicity and clarity
it was assumed that the first and second carrier oscillator
rrequencies were at 4.0 and 4.1 kHz. However, any carrier
frequencies cou!d have been selected as may b~ desired, as
well as their relative offset. In the presently preferred
embodiment, the first carrier frequency is 4.0000 kHz, while
the second carrier frequency is set at 4.01024 kHz. It has
generally been determined according to the invention that
carrier frequency offsets in the range of 0~ to 20 Hz do not
substantially degrade the intelligibility of the caller's
voice, while such an offset at the same time provides
sufficient inhibition against the establishment of
interfering feedback loops. The closer the carrier
frequencies are to each other, the greater is the fidelity
16
~ 7
of the caller's voice signal, and the greater the tendency
to create interfering feedback loops. Similarly, the
further apart the carrier frequencies, the less is the
fidelity to the caller's voice, and less tendency there is
to the generation of interfering feedback loops. In
practice, a frequency separation between the carrier
oscillators of approximately 5 to 20 Hz provides a practical
optimum.
Therefore, the illustrative embodiment must be
understood only as an example, and not as a limitation of
the invention, which is defined in the following claims:
