DTMF tone detection and suppression with application to computer telephony over packet switched networks

• US 6,226,303 B1
• Filed: 03/06/1998
• Issued: 05/01/2001
• Est. Priority Date: 03/06/1997
• Status: Expired due to Term

• Current Assignee: VOIP Dynamic LLC (General Patent Corporation)
• Original Assignee: Natural Microsystems Corporation (Live Microsystems, Inc.)
• Inventors: Levens, Bruce, Kerr, Paul R., Hoge, William S., Daly, Daniel F.
• Primary Examiner(s): Chin, Wellington
• Assistant Examiner(s): JONES, PRENELL P
• Application Number:
  US09/036,342
• Time in Patent Office: 1,152 Days
• Field of Search: 370/526, 708/312, 379/386, 379/282, 379/283, 455/412, 455/413, 455/418, 455/422, 455/563
• US Class Current: 370/526
• CPC Class Codes: H04L 65/1101 Session protocols
• CPC Class Codes: H04L 65/70 Media network packetisation
• CPC Class Codes: H04M 7/006 Networks other than PSTN/ISDN providing telephone service, e.g. Voice over Internet Protocol (VoIP) , including next generation networks with a packet-switched transport layer H04L65/00 takes precedence; aspects not specific to the type of network H04M3/00; special services in those networks H04M3/42
• CPC Class Codes: H04M 7/1295 Details of dual tone multiple frequency signalling

Abstract

A method of facilitating telephony communications over a packet-switched network includes detecting signaling tones in an audio stream and blocking them from the stream in response to the detecting. After the blocking, the audio stream is sent over an undependable channel in the packet-switched network. The detected signaling tones can be sent over a dependable channel instead of the undependable audio channel.

Claims

1. A method of facilitating telephony communication over a packet-switched network, comprising the steps of:

detecting signaling tones in an audio stream,

blocking the signaling tones detected in the step of detecting from the audio stream in response to the step of detecting, and

sending the audio stream over an undependable audio channel in the packet-switched network after the step of blocking the signaling tones.

2. The method of claim 1, further including the step of sending the signaling tones detected in the step of detecting over a dependable control channel instead of the undependable audio channel.

3. The method of claim 1, wherein the step of detecting includes a step of filtering.

4. The method of claim 1, wherein the step of detecting includes a step of low-pass filtering, a step of low band-pass filtering, and a step of high band-pass filtering.

6. The method of claim 1, wherein the step of detecting includes a step of low band-pass filtering and a step of high band-pass filtering.

7. The method of claim 6 wherein the step of detecting further includes a step of phase-shifting results of the steps of filtering.

8. The method of claim 6, wherein the step of detecting further includes steps of performing statistical ratio tests to determine if the steps of filtering are each applied to only one tone in their respective pass bands.

9. The method of claim 1, wherein the step of detecting includes steps of performing statistical ratio tests.

10. The method of claim 9, wherein the statistical ratio tests include a root of a sum of powers divided by a root of a sum of powers.

11. The method of claim 1, wherein the step of detecting detects DTMF tones as the signaling tone.

5. A method of facilitating telephony communication over a packet-switched network, comprising the steps of:

detecting signaling tones in an audio stream, including a step of low-pass filtering, a step of low band-pass filtering, a step of high band-pass filtering, determining if a result of the step of low-pass filtering is too large, determining if results of the steps of band-pass filtering are too small, and determining if results of the steps of band-pass filtering are too different,

blocking the signaling tones detected in the step of detecting from the audio stream in response to the step of detecting, and

sending the audio stream over an undependable audio channel in the packet-switched network after the step of blocking the signaling tones.

12. An audio detection and suppression system for use in connection with packet-switched networks, comprising:

means for detecting signaling tones in an audio stream, and

means for suppressing the signaling tones in the stream detected by the means for detecting before they are sent from an audio output over an undependable audio channel in the packet-switched network.

13. The system of claim 12, further including means for sending the signaling tones over a dependable control channel instead of the undependable audio channel.

14. The system of claim 12, wherein the means for detecting includes a low band-pass filter and a high band-pass filter.

15. The system of claim 14 wherein the filters are FIR filters.

16. The system of claim 14 wherein the means for detecting further includes means for phase-shifting outputs of the filters.

17. The system of claim 14 wherein the means for detecting further includes means for performing statistical ratio tests to determine if the filters each contain only one tone in their respective pass bands.

18. The system of claim 12, wherein the means for detecting further includes means for performing statistical ratio tests.

19. The system of claim 18, wherein the means for performing statistical ratio tests includes means for performing a root of a sum of powers of absolute values divided by a root of a sum of powers of absolute values.

20. The system of claim 12, wherein the means for detecting includes a low-pass filter, a low band-pass filter, and a high band-pass filter.

22. The system of claim 12, wherein the means for detecting includes means for detecting DTMF tones as the signaling tone.

21. An audio detection and suppression system for use in connection with packet-switched networks, comprising:

means for detecting signaling tones in an audio stream, including a low-pass filter, a low band-pass filter, a high band-pass filter, means for determining if an output of the low-pass filter is too large, means for determining if outputs of the band-pass filters are too small, and means for determining if outputs of the band-pass filters are too different, and

means for suppressing the signaling tones in the stream detected by the means for detecting before they are sent from an audio output over an undependable audio channel in the packet-switched network.

23. A method of facilitating telephony communication over a packet-switched network, the method comprising the steps of:

passing an audio stream through a system having at least one filter prior to an audio output,

detecting a selected audio signal in the audio stream utilizing the at least one filter, and

blocking the selected audio signal from passage through the audio output in response to the step of detecting.

24. The method of claim 23, wherein the selected audio signal is a DTMF tone.

25. The method of detecting a selected signal in a digital signal, comprising:

decimating the digital signal to obtain a decimated digital signal,

applying a statistical test to the decimated digital signal, and

determining whether the selected signal is present in the digital signal based on the result of the step of applying.

26. The method of claim 25, further including the step of issuing a signal over a control channel in response to a result of the step of determining.

27. The method of claim 25, wherein the step of applying a statistical function applies a statistical ratio function to the decimated digital signal.

28. The method of claim 25, wherein the step of applying a statistical function applies a statistical ratio function to the decimated digital signal, and wherein the statistical ratio function includes a root of a sum of powers of absolute values divided by a root of a sum of powers of absolute values.

29. The method of claim 25, wherein the selected signal is a DTMF tone.

30. A system for receiving first audio signals and transmitting second audio signals to a packet-based network, the system comprising:

an input for receiving the first audio signals;

a band-pass filter arrangement coupled to the input and configured to pass a filtered signal having a frequency associated with a dual-tone, multi-frequency pair; and

a processor coupled to the filter arrangement;

wherein at least one of the filter arrangement and the processor is configured to decimate the filtered signal to produce a filtered, decimated signal; and

wherein the processor is configured to compare characteristics of the filtered, decimated signal to determine whether the filtered, decimated signal is representative of a portion of a dual-tone, multi-frequency signal.

31. The system of claim 30 wherein the processor is configured to determine whether the filtered, decimated signal is representative of one tone or more than one tone.

32. The system of claim 31 wherein the processor is configured to compute a ratio of characteristics of the filtered, decimated signal.

33. The system of claim 32 wherein the characteristics are peak and root mean square (RMS), fourth root of average of 4^th power and RMS, RMS and average of absolute value, or an n^th root of a sum of n^th powers and an m^th root of a sum of m^th powers.

34. A method for receiving first audio signals and transmitting second audio signals to a packet-based network, the method comprising:

receiving the first audio signals;

filtering and decimating the first audio signals to pass a filtered, decimated signal having a frequency associated with a dual-tone, multi-frequency pair; and

comparing characteristics of the filtered, decimated signal to determine whether the filtered, decimated signal is representative of a portion of a dual-tone, multi-frequency signal.

35. The method of claim 34 wherein the comparing determines whether the filtered, decimated signal is representative of one tone or more than one tone.

36. The method of claim 35 wherein the comparing includes computing a ratio of characteristics of the filtered, decimated signal.

37. The method of claim 36 wherein the characteristics are peak and root mean square (RMS), fourth root of average of 4^th power and RMS, RMS and average of absolute value, or an n^th root of a sum of n^th powers and an m^th root of a sum of m^th powers.