Full duplex speakerphone design using acoustically compensated speaker distortion

• US 8,811,602 B2
• Filed: 06/30/2011
• Issued: 08/19/2014
• Est. Priority Date: 06/30/2011
• Status: Expired due to Fees

• Current Assignee: Avago Technologies International Sales Pte Limited (Broadcom, Inc.)
• Original Assignee: Broadcom Corporation (Broadcom, Inc.)
• Inventors: Khanduri, Prakash, Sollenberger, Nelson, Zeng, Huaiyu
• Primary Examiner(s): Tsang, Fan
• Assistant Examiner(s): LYTLE, JEFFREY P
• Application Number:
  US13/174,148
• Publication Number:
  US 20130003960A1
• Time in Patent Office: 1,146 Days
• Field of Search: None
• US Class Current: 379/406.06
• CPC Class Codes: H04M 9/082 using echo cancellers echo cancellers per se H04B3/23

Abstract

A telecommunication system including a fall duplex speakerphone, comprising a first microphone to generate a coupled signal including uplink information and non-linear distortion sensed by the first microphone in a speaker phone mode, a second microphone to generate a reference signal including downlink information and the non-linear distortion sensed by the second microphone in the speaker phone mode, and an acoustic echo canceller (AEC) to receive the coupled signal from the first microphone, to receive the reference signal from the second microphone, and to cancel out the non-linear distortion included in the coupled signal based on the non-linear distortion included in the reference signal.

Claims

1. A telecommunication system including a full duplex speakerphone, comprising:

a main signal path, including a main microphone, configured to sense an uplink signal and an echo signal and generate a coupled signal, the echo signal including non-linear distortion generated by a speaker of the speakerphone;

an echo signal path, including an echo reference microphone, configured to sense the echo signal and generate a reference signal, the reference signal including downlink information and the non-linear distortion; and

an acoustic echo canceller (AEC) configured to receive the coupled signal from the main signal path, to receive the reference signal from the echo signal path, and to cancel out the non-linear distortion included in the coupled signal based on the non-linear distortion included in the reference signal, wherein, to cancel out the non-linear distortion, the AEC is configured to introduce a delay in at least part of the reference signal, the delay being proportional to a path delay associated with the echo signal path.

2. The telecommunication system according to claim 1, wherein the non-linear distortion is sensed by the main microphone and the non-linear distortion is sensed by the echo reference microphone when the telecommunication system operates in a speaker phone mode.

3. The telecommunication system according to claim 1, wherein the echo signal path generates the reference signal based on an acoustic coupling of the echo reference microphone with the speaker.

4. The telecommunication system according to claim 3, wherein the echo reference microphone is acoustically coupled with the speaker by placing the echo reference microphone in an acoustical cavity of the speaker.

5. The telecommunication system according to claim 4, wherein the acoustical cavity of the speaker includes physical boundaries that encompass both the speaker and the echo reference microphone.

6. The telecommunication system according to claim 1, wherein the AEC cancels out the non-linear distortion by subtracting the non-linear distortion included in the reference signal from the non-linear distortion included in the coupled signal.

7. The telecommunication system according to claim 6, wherein the subtracting includes at least one of introducing the delay in at least a part of the reference signal, inverting a phase of the reference signal, and regulating an amplitude of the reference signal.

8. The telecommunication system according to claim 1, further comprising:

a processor configured to provide the downlink information to be broadcast by the speaker in a speaker phone mode, and wherein the echo signal path senses the echo signal when the processor is actively providing the downlink information.

9. The telecommunication system according to claim 1, further comprising:

a processor configured to provide the downlink information to be broadcast by the speaker in a speaker phone mode, wherein the AEC receives the reference signal when the processor is actively providing the downlink information, and cancels out the non-linear distortion included in the coupled signal when a sound level of the downlink information is at or above a predetermined threshold.

10. The telecommunication system according to claim 9, wherein the predetermined threshold is programmable.

11. The telecommunications system of claim 1, wherein the non-linear distortion is related to characteristics of the speaker, including signal drive level of the speaker.

12. A method for canceling out non-linear distortion in a telecommunication system, the method comprising:

generating, using a main microphone, a coupled signal including uplink information and non-linear distortion sensed by the main microphone, the non-linear distortion generated by a speaker of a speakerphone in the telecommunication system;

generating, using an echo reference microphone, a reference signal including downlink information and the non-linear distortion;

receiving, at an acoustic echo canceller (AEC), the coupled signal from the main microphone and the reference signal from the echo reference microphone; and

cancelling out, at the AEC, the non-linear distortion included in the coupled signal based on the non-linear distortion included in the reference signal, the canceling out including introducing a delay in at least part of the reference signal, the delay being proportional to a path delay associated with the echo signal path.

13. The method according to claim 12, wherein the generating the coupled signal and the generating the reference signal are performed in a speaker phone mode.

14. The method according to claim 12, wherein the generating the reference signal is performed based on an acoustic coupling of the echo reference microphone with the speaker.

15. The method according to claim 12, further comprising:

acoustically coupling the echo reference microphone with the speaker by placing the echo reference microphone in an acoustical cavity of the speaker.

16. The method according to claim 15, wherein the acoustical cavity of the speaker includes physical boundaries that encompass both the speaker and the echo reference microphone.

17. The method according to claim 12, wherein the cancelling out the non-linear distortion includes subtracting, in the AEC, the non-linear distortion included in the reference signal from the non-linear distortion included in the coupled signal.

18. The method according to claim 17, wherein the subtracting includes at least one of introducing the delay in at least a part of the reference signal, inverting a phase of the reference signal, and regulating an amplitude of the reference signal.

19. The method according to claim 12, further comprising:

providing, via a processor, downlink information to be broadcast by the speaker in a speaker phone mode, wherein the generating the reference signal includes generating the reference signal when the processor is actively providing the downlink information.

20. The method according to claim 12, further comprising:

providing, via a processor, downlink information to be broadcast by the speaker in a speaker phone mode;

receiving, at the AEC, the reference signal when the processor is actively providing the downlink information;

detecting, via a detector, a sound level of the downlink information; and

canceling out, at the AEC, the non-linear distortion included in the coupled signal when the sound level of the downlink information is at or above a predetermined threshold.

21. The method according to claim 20, wherein the predetermined threshold is programmable.