System and method for detecting unauthorized call activity

• US 7,079,637 B1
• Filed: 04/22/2003
• Issued: 07/18/2006
• Est. Priority Date: 04/22/2003
• Status: Active Grant

• Current Assignee: Securus Technologies Incorporated (Securus Special Investments LLC)
• Original Assignee: Evercom Systems Incorporated (Securus Special Investments LLC)
• Inventors: Jurcak, Tom, McNitt, George
• Primary Examiner(s): Hong, Harry S.
• Application Number:
  US10/420,585
• Time in Patent Office: 1,183 Days
• Field of Search: 379/188, 379/189, 379/196, 379/197, 379/198, 379/199, 379/377, 379/200, 379/207.01
• US Class Current: 379/189
• CPC Class Codes: H04M 1/67 by electronic means
• CPC Class Codes: H04M 3/38 Graded-service arrangements, i.e. some subscribers prevented from establishing certain connections queuing arrangements H04Q3/64

Abstract

Disclosed are systems and methods which detect unauthorized, impermissible, or otherwise undesired call activity through a silence detection technique employing multiple samples. One embodiment monitors a call in progress to detect an event such as might be indicative of unauthorized call activity. Thereafter, a sampling process may be invoked which analyzes multiple samples of the call to make a determination as to whether or not a particular unauthorized call activity is being attempted. A party to the call may be muted to thereby prevent that party from thwarting the unauthorized call activity determination. Embodiments implement techniques for defeating parties to calls learning the unauthorized call activity detection system in order to avoid detection, such as by implementing a plurality of detection techniques.

Claims

1. A method for detecting unauthorized call activity, said method comprising:

detecting a high signal level event;

implementing a plurality of sampling periods within a predetermined sample window following said high signal level event; and

analyzing samples from said sampling periods to make a determination with respect to said unauthorized call activity.

2. The method of claim 1, wherein said detecting said high signal level event comprises:

determining if signal energy associated with a call exceeds a predetermined threshold value.

3. The method of claim 2, wherein said predetermined threshold value is in the range of 55–

70% of a maximum signal amplitude.

4. The method of claim 1, wherein said implementing said plurality of sampling periods comprises:

muting a party to a call during said sampling periods.

5. The method of claim 1, wherein said sampling periods are within the range of from 100 msec to 500 msec in duration.

6. The method of claim 1, wherein said plurality of sampling periods comprises at least 3 sampling periods.

7. The method of claim 1, wherein said sample window is within the range of from 2 to 3 seconds in duration.

8. The method of claim 1, wherein said sampling periods are distributed evenly throughout said sample window.

9. The method of claim 1, wherein said sampling periods are distributed unevenly throughout said sample window.

10. The method of claim 1, wherein analyzing said samples comprises utilizing a predetermined silence threshold value.

11. The method of claim 10, wherein said predetermined silence threshold value is in the range of 2–

6% of a maximum signal amplitude.

12. The method of claim 10, wherein said predetermined silence threshold value is determined on a call by call basis.

13. The method of claim 10, wherein analyzing said samples comprises determining if any sample exceeds said predetermined silence threshold value.

14. The method of claim 10, wherein analyzing said samples comprises determining if an average derived from said samples exceeds said predetermined silence threshold value.

15. The method of claim 1, wherein analyzing said samples comprises discarding at least one of said samples.

16. The method of claim 1, further comprising:

implementing a quiescent time period between said detecting said high signal level event and implementing said plurality of sampling periods.

17. The method of claim 1, further comprising:

invoking quasi-randomization with respect to one or more parameters of said detecting a high signal level event, said implementing said plurality of sampling periods, and said analyzing said samples to thereby thwart learning a detecting technique through use thereof.

18. The method of claim 17, wherein said quasi-randomization with respect to said one or more parameters comprises varying a number of sampling periods.

19. The method of claim 17, wherein said quasi-randomization with respect to said one or more parameters comprises varying a duration of said sampling periods.

20. The method of claim 17, wherein said quasi-randomization with respect to said one or more parameters comprises varying a method of analysis used with respect to said samples.

21. The method of claim 1, further comprising:

implementing a plurality of techniques for detecting unauthorized call activity, wherein said detecting said high signal level event, said implementing said plurality of sampling periods, and said analyzing said samples comprises a first technique of said plurality of techniques.

22. The method of claim 21, wherein a second technique of said plurality of techniques comprises monitoring for signal energy continuously below a predetermined silence threshold value for at least a predetermined time to thereby provide silence period detection.

23. The method of claim 21, wherein different techniques of said plurality of techniques are implemented with respect to a particular call.

24. The method of claim 21, wherein different techniques of said plurality of techniques are implemented on a call by call basis.

25. The method of claim 1, wherein sampling of said call is accomplished with respect to a linear representation of a call signal.

26. A system for detecting unauthorized call activity, said system comprising:

an event detector indicating when signal energy associated with a communication link between said calling party and said called party exceeds a predetermined event threshold;

a signal sampler to produce a plurality of samples from said communication link following event detection by said event detector; and

a sample analyzer providing analysis with respect to said samples and providing a determination with respect to unauthorized call activity.

27. The system of claim 26, wherein said event detector, said signal sampler, and said sample analyzer are part of a call processing system.

28. The system of claim 27, wherein said unauthorized call activity comprises a party disposed remote to said call processor signaling telecommunications equipment to invoke an enhanced calling service.

29. The system of claim 27, wherein said call processing system is disposed within a controlled environment facility.

30. The system of claim 29, wherein said controlled environment facility comprises a prison facility.

31. The system of claim 26, said event detector indicates signal energy exceeding said predetermined event threshold without regard to signal frequency.

32. The system of claim 26, further comprising:

a database of parameters for adjusting operation of at least one of said event detector, said signal sampler, and said sample analyzer.

33. The system of claim 32, wherein a subset of said parameters are provided to a call processing system hosting said event detector, said signal sampler, and said sample analyzer upon installation.

34. The system of claim 32, wherein a subset of said parameters of said database are accessed by a call processing system hosting said event detector, said signal sampler, and said sample analyzer during real-time processing of a call.

35. A method for detecting a three-way call attempt, said method comprising:

detecting a high signal level event in a call;

implementing a plurality of sampling periods within a predetermined sample window following said high signal level event, wherein a party to said call is muted during each of said sampling periods and unmuted between said sampling periods; and

analyzing samples from said sampling periods to determine if silence below an ambient noise level exists in said sample window.

36. The method of claim 35, wherein said detecting said high signal level event comprises:

determining if signal energy associated with said call exceeds a predetermined threshold value.

37. The method of claim 35, wherein said plurality of sampling periods comprises at least 3 sampling periods.

38. The method of claim 35, wherein said sampling periods are distributed evenly throughout said sample window.

39. The method of claim 35, wherein said sampling periods are distributed unevenly throughout said sample window.

40. The method of claim 35, wherein analyzing said samples comprises determining if any sample exceeds a predetermined silence threshold value.

41. The method of claim 35, wherein analyzing said samples comprises determining if an average derived from said samples exceeds a predetermined silence threshold value.

42. The method of claim 35, further comprising:

implementing a quiescent time period between said detecting said high signal level event and implementing said plurality of sampling periods.

43. The method of claim 35, further comprising:

varying one or more parameters of said detecting a high signal level event, said implementing said plurality of sampling periods, and said analyzing said samples.

44. The method of claim 43, wherein said varying said one or more parameters comprises varying a number of sampling periods.

45. The method of claim 43, wherein said varying said one or more parameters comprises varying a duration of said sampling periods.

46. The method of claim 43, wherein said varying said one or more parameters comprises varying a method of analysis used with respect to said samples.

47. The method of claim 35, further comprising:

implementing a plurality of techniques for detecting unauthorized call activity, wherein said detecting said high signal level event, said implementing said plurality of sampling periods, and said analyzing said samples comprises a first technique of said plurality of techniques.

48. The method of claim 47, wherein different techniques of said plurality of techniques are implemented with respect to a particular call.

49. The method of claim 47, wherein different techniques of said plurality of techniques are implemented on a call by call basis.

50. The method of claim 35, wherein sampling of said call is accomplished with respect to a linear representation of a call signal.