Ultrasonically-calibrated fast-start echo canceller for...

Telephonic communications – Echo cancellation or suppression – Using digital signal processing

Reexamination Certificate

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

C379S406010

Reexamination Certificate

active

06510225

ABSTRACT:

TECHNICAL FIELD
The present specification generally relates to communication devices. More particularly, the present specification describes ultrasonically-calibrated fast-start adaptive echo canceller system.
BACKGROUND
Echoes are present in most conversations conducted indoors. The actual echoes within a room or car are typically on the order of about a microsecond. The so-called “cyclo-acoustic effect” governs how this echo is perceived. If the delay between the speech and the echo is less than a few tens of milliseconds, the echo is not very noticeable but perceived as a form of spectral distortion or reverberation. For longer delays, on the order of about 100 milliseconds, the echo becomes more noticeable.
Echoes may be experienced on a telephone circuit. Every telephone in a given area is connected to a central office by a channel that serves the need for communications in both directions. Accordingly, there has to be provision for connecting the two-wire circuit to the four-wire circuit. This connection is accomplished by means of a hybrid transformer.
When a speech signal encounters an impedance mismatch at any point on a telephone circuit, such as at a hybrid transformer, a portion of that signal may reflect as an echo. In mobile communication systems, vocoder delays and convolution coding algorithms may introduce additional delay. The round trip signal delays can be on the order of 100 milliseconds, for example.
Another type of echo is an acoustic echo. The signal transmitted from a speaker is reflected from nearby objects and is picked up by a microphone. This type of echo commonly occurs when a caller is using a speakerphone in a vehicle where the speaker and microphone are both in a relatively small and enclosed space.
The echoes can be canceled using adaptive echo cancellers. The basic principle is to synthesize a replica of the echo and subtract it from the returned signal. For the adaptive echo cancellation circuit to operate satisfactorily, the impulse response of the adaptive filter should have a length greater than the longest echo delay that needs to be accommodated.
Effective electronic echo cancellers require a training interval at the start of each call to determine delay and attenuation coefficients corresponding to the time variant acoustic echo configuration. During this training interval the echo cancellation is not effective.
With sampling rate for a speech signal conservatively chosen as 8 kHz, the length of the adaptive filter would have to be 512 or greater to accommodate echo delays between 50 and 100 milliseconds. Therefore, it can take a considerable amount of time to train and adapt such a large number of delay and attenuation coefficients for an adaptive echo cancelling synthesizer. This means that a caller could wait as much as two to three seconds after a call is initiated before the optimum set of coefficients is adapted and applied.
SUMMARY
The present disclosure describes an ultrasonically-calibrated fast-start adaptive echo canceller system. The system has at least one cellular transceiver; at least one speaker for converting the input signal to audible sound; and at least one microphone for picking up a desired voice signal. The microphone also picks up an echo of the input signal resulting in a combined desired and undesired signal. The ultrasonic portion includes at least one ultrasonic transceiver for transmitting and receiving an ultrasound signal.
The echo canceller system also includes an adaptive echo cancelling synthesizer that has, for example, “stages” of adjustable time delays, gain amplifiers and adders configured to synthesize a replica of the echo of the input signal. In a preferred embodiment, the echo cancelling synthesizer is a transversal filter. The replica of the echo is synthesized by taking the input signal and successively time-delaying, amplifying and adding to the previously-calculated value.
The output of the adaptive echo cancelling synthesizer is then subtracted from the combined signal picked up by the microphone. An error signal is generated by this subtraction.
The echo canceller system also includes a selection element that receives inputs from the error signal and the input signal. The selection element has a selection controller and a memory.
An echo pulse counter analyzes recorded time delays of the ultrasound echo signal. The time delays are recorded when the amplitude of the ultrasound echo signal exceeds certain pre-specified thresholds. The ultrasonic echo pulse counter operates prior to call initiation and produces the set of ultrasonic coefficients based on the recorded time delays. This set is compared with previous sets of ultrasonic coefficients to find the closest match. The matched ultrasonic set is used to correlate an initial audio-frequency set of coefficients using a pre-determined conversion table.
The selection controller selects and applies an initial set of audio-frequency coefficients for a particular echo configuration from the audio-frequency sets in the memory.
In a further embodiment, the echo canceller system includes a system controller that monitors call initiations and terminations and generates commands. The system controller commands a nonvolatile memory to load and store sets of coefficients corresponding to previously-obtained ultrasonic echoes.
The present disclosure further describes a method for quickly starting an echo canceller system. The method includes steps of recording time delays of an ultrasonic echo signal when the amplitude of the echo exceeds certain pre-specified thresholds; generating set of ultrasonic coefficients based on the recorded time delays; and comparing the current set of ultrasonic coefficients with previous sets to find the closest match.
Once the sets of ultrasonic coefficients are loaded into the memory, the selection controller correlates the matched set of ultrasonic coefficients with set of audio-frequency coefficients using a predetermined ultrasonic to audio-frequency conversion table. Then, the selection controller applies an initial set of audio-frequency coefficients from the audio-frequency sets in the memory. Finally, the selection element controller starts a normal adaption process of iteratively synthesizing an echo of an audio input signal and subtracting the synthesized echo from a combined signal of desired voice and audio input signal based on a previous difference.
The details of one or more embodiments are set forth in the accompanying drawings and the description below. Other embodiments and advantages will become apparent from the following description and drawings, and from the claims.


REFERENCES:
patent: 5022082 (1991-06-01), Eriksson et al.
patent: 5050160 (1991-09-01), Fuda
patent: 5668794 (1997-09-01), McCaslin et al.

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Ultrasonically-calibrated fast-start echo canceller for... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Ultrasonically-calibrated fast-start echo canceller for..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Ultrasonically-calibrated fast-start echo canceller for... will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-3029902

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.