Telephonic communications – Echo cancellation or suppression
Reexamination Certificate
1998-08-25
2001-01-09
Hudspeth, David R. (Department: 2741)
Telephonic communications
Echo cancellation or suppression
C379S413020, C381S104000, C381S106000, C381S107000
Reexamination Certificate
active
06173056
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to the field of communications, and more particularly to telephone communications methods and systems.
BACKGROUND OF THE INVENTION
Hands-free loudspeaker radiotelephones have been developed for use in automobiles so that the driver can talk on the radiotelephone while keeping hands free for driving. Accordingly, the driver can operate the automobile with greater concentration while talking on the radiotelephone. The driver does not need to hold a handset to the head because a loudspeaker broadcasts speech transmitted from a remote communications device being used by a remote party throughout the passenger compartment. A microphone is used to pick up the speech in the passenger compartment for transmission to the remote communications device.
An acoustic feedback path from the loudspeaker to the microphone, however, may cause undesired echoes to be transmitted to the remote communications device. Accordingly, echo suppressors have been developed for loudspeaker radiotelephones. In particular, an echo suppressor monitors an input audio signal to be applied to the loudspeaker and models an echo path of this signal from the speaker to the microphone to generate an estimate of the undesired echo portion of the output audio signal generated by the microphone. The echo estimate is then combined with the output audio signal to suppress (or reduce) the echo portion of the signal transmitted to the remote party. Echo suppression is discussed, for example, in U.S. Pat. No. 5,600,718 to Dent et al. entitled Apparatus and Method for Adaptively Precompensating for Loudspeaker Distortions, and in U.S. Pat. No. 5,680,450 to Dent et al. entitled Apparatus and Method for Canceling Acoustic Echoes Including Non-Linear Distortions in Loudspeaker Telephones. The disclosures of these patents are hereby incorporated herein in their entirety by reference.
Non-linear distortions of the input audio signal generated down stream from the point monitored by the echo suppressor, however, may not be modeled by the echo suppressor so that these distortions may result in unwanted echo and/or distortion being transmitted to the remote party. In particular, the echo suppressor may be implemented as a portion of a digital signal processor so that the input audio signal is processed in a digital form and then converted from digital form to analog form before being applied to the loudspeaker. Moreover, a power amplifier can be used to amplify the analog signal from the D-to-A converter before applying the analog signal to the loudspeaker. Accordingly, non-linear distortions caused by the power amplifier may not be accurately modeled by the echo suppressor.
In particular, the maximum output of the power amplifier generally is limited by the level of the electrical voltage supplied thereto. The electrical power supply for a loudspeaker radiotelephone in an automobile, for example, may be an automotive battery with a specified voltage output level of 1 0.8V to 15.6V. As a loss of 0.8V can be expected from the battery to the loudspeaker radiotelephone, the radiotelephone can be expected to operate from a power supply level as low as of 10V. A conventional radiotelephone may thus be designed so that the maximum allowed input signal to the power amplifier does not cause the power amplifier to generate signals having peak-to-peak amplitudes which would be clipped when the power supply is at 10 V, for example.
Situations may occur, however, where an automobile battery voltage is lower than the specified range. This may occur, for example, when the car is not running; when the battery is weak; when the alternator is not operating properly; when the engine is started; when the battery is cold; when the headlight highbeams are on; and/or when many electrical accessories are on. Accordingly, non-linear distortions may be generated by the power amplifier, and these non-linear distortions may cause an increase in echo and/or distortion transmitted to the remote party because the echo suppressor is unable to model these distortions. Moreover, these distortions may be generated at a time when reliable communications are most desired, such as when the user is having car problems.
SUMMARY OF THE INVENTION
It is, therefore, an object of the present invention to provide improved telephone methods and systems.
It is another object of the present invention to provide improved methods of generating sound from electronic devices.
It is still another object of the present invention to provide improved echo suppression for telephones.
These and other objects are provided according to the present invention by adjusting an audio input signal responsive to changes in a power supply level, and then amplifying the adjusted audio input signal. The adjusted and amplified audio input signal is then used to generate sound. Accordingly, clipping during amplification can be reduced or even eliminated. By reducing clipping during amplification, the quality of sound generated by a loudspeaker telephone can be improved. Furthermore, the adjustments prior to amplification can be accounted for during echo suppression.
A method according to the present invention includes the steps of monitoring a level of an electrical power supply for an electrical communications device, and adjusting an audio input signal responsive to changes of the level of the electrical power supply for the electronic communications device. After the adjusting step, the adjusted audio input signal is amplified, and an output sound is generated responsive to the adjusted and amplified audio input signal.
In addition, an audio output signal can be generated responsive to received sound including an echo of the output sound, and an estimate of the echo of the output sound can be generated using the adjusted audio input signal and a model of an echo path of the output sound. This estimate can then be used to suppress an echo portion of the audio output signal. Because the adjustments to the audio input signal are included in the estimate, the echo suppression is improved.
The adjusting step can include adjusting a gain of the audio input signal so that the adjusted audio input signal does not exceed an amplitude that would result in clipping during the amplifying step at the monitored level of the electrical power supply. The quality of the output sound can thus be improved. Alternately, the adjusting step can include clipping portions of the audio input signal which exceed an amplitude that would result in clipping during the amplifying step at the monitored level of the electrical power supply. Accordingly, portions of the audio input signal which would not result in clipping during amplification are not adjusted.
The monitoring step can include detecting a reduction in the level of the electrical power supply, and the adjusting step can include reducing a peak-to-peak swing of the audio input signal responsive to the detected reduction in the level of the electrical power supply. In addition, the monitoring step can include detecting an increase in the level of the electrical power supply, and the adjusting step can include increasing a peak-to-peak swing of the audio input signal responsive to the detected increase in the level of the electrical power supply. Accordingly, the peak-to-peak swings of the audio input signal can track both reductions and increases in the detected power supply level.
According to an alternate aspect of the present invention, a level of an electrical power supply for the electrical communications device is monitored, and a digital audio input signal is adjusted responsive to changes of the level of the electrical power supply for the electronic communications device. The adjusted digital audio input signal is converted to an adjusted analog audio input signal, and an output sound is generated responsive to the adjusted analog audio input signal. Accordingly, a digital signal processor can be used to adjust the audio input signal. Furthermore, the same digital signal processor can be used to provide echo
Lilja Patrik
Romesburg Eric
Ericsson Inc.
Hudspeth David R.
Myers Bigel & Sibley & Sajovec
Wieland Susan
LandOfFree
Methods for adjusting audio signals responsive to changes in... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Methods for adjusting audio signals responsive to changes in..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Methods for adjusting audio signals responsive to changes in... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2468741