Generation of diagnostic data from interarrival times of VoIP packets

• US 8,885,485 B2
• Filed: 07/31/2006
• Issued: 11/11/2014
• Est. Priority Date: 08/18/2005
• Status: Active Grant

• Current Assignee: VOIPFUTURE GmbH (Jan Bastian Asset Management GmbH)
• Original Assignee: VOIPFUTURE GmbH (Jan Bastian Asset Management GmbH)
• Inventors: Zaencker, Olaf Carl
• Primary Examiner(s): Phung, Luat
• Application Number:
  US12/063,971
• Publication Number:
  US 20080186872A1
• Time in Patent Office: 3,025 Days
• Field of Search: 370/252, 370/241, 370/242, 370/352
• US Class Current: 370/244
• CPC Class Codes: H04L 1/205 jitter monitoring
• CPC Class Codes: H04L 1/24 Testing correct operation
• CPC Class Codes: H04L 43/045 for graphical visualisation of monitoring data
• CPC Class Codes: H04L 43/0823 Errors, e.g. transmission errors
• CPC Class Codes: H04M 2201/12 Counting circuits
• CPC Class Codes: H04M 7/0084 Network monitoring; Error detection; Error recovery; Network testing

Abstract

Disclosed is a method for analyzing errors of a data stream, particularly a real-time data stream, in a data network (100). According to said method, a time stamp and/or a sequence number is/are determined from or for each data packet transmitted between two communication terminals (10, 12) of the data network (100), a history (diagram) and/or a frequency distribution (histogram) of at least one data transmission parameter determined from the plurality of time stamps and/or sequence numbers of the data packets of the data stream is generated, and a pattern of the history and/or the frequency distribution of the at least one data transmission parameter is identified in order to determine the location and/or the cause of an error during the transmission of the data stream in the data network.

Claims

1. A method comprising using a computer to analyze a data stream of VoIP (Voice over IP) packets sent from a communication device, including using the computer to receive the VoIP packets, compute receive interarrival times of consecutive received VoIP packets, form a histogram or diagram of the receive interarrival times, and perform pattern recognition on the histogram or diagram to identify any sender problems in the communication device.

2. A method according to claim 1, wherein the communication device has an RTP (Realtime Transport Protocol) stack.

3. A method according to claim 1, wherein each receive interarrival time is computed as a difference in receipt times between two consecutive received packets having consecutive sequence numbers.

4. A method according to claim 3, wherein receipt time of each VoIP packet is indicated by a measurement timestamp.

5. A method according to claim 1, wherein the packets contain real voice data.

6. A method according to claim 1, wherein the histogram includes a frequency distribution of the interarrival times.

7. A method according to claim 1, wherein the diagram includes a temporal progression of the interarrival times.

8. A method according to claim 1, wherein the pattern recognition identifies configuration of the communication device as a sender problem.

9. The method according to claim 8, wherein a sender problem caused by the configuration is recognized as a complete shift of an undisturbed interarrival time pattern towards higher values.

10. A method according to claim 1, wherein the pattern recognition identifies a network problem instead of a sender problem.

11. The method according to claim 10, wherein the network problem of lower bandwidth is recognized as discrete interarrival time whole number values with no packet loss.

12. The method according to claim 10, wherein packet loss is recognized as harmonics of a normal state of the interarrival times.

13. A method according to claim 1, wherein the pattern recognition identifies an RTP (Realtime Transport Protocol) stack in the communication device as a sender problem.

14. The method according to claim 13, wherein a problem with an RTP stack of the communication device is recognized as NZ-IAT (near zero interarrival time) in the histogram.

15. A method according to claim 1, wherein the pattern recognition identifies an operating system of the communication device as a sender problem.

16. A method according to claim 1, wherein whether components involved with transmission of the data stream are able to process an RTP (Realtime Transport Protocol) data packet is ascertainable as a cause of a sender problem.

17. The method according to claim 15, wherein a problem with an operating system of the communication device is recognized as cyclic jitter in the histogram.

18. A method according to claim 1, wherein the pattern recognition identifies at least one of an operating system of the communication device, configuration of the communication device, and an RTP (Realtime Transport Protocol) stack of the communication device as a sender problem.

19. A method according to claim 1, wherein the data stream is also analyzed at more than one place of measurement.

20. The method of claim 1, wherein the pattern recognition reveals sender and receiver problems that impact VoIP speech quality.

21. The method of claim 1, wherein the histogram or diagram is formed from the interarrival times of the received packets without using corresponding transmit times of the communication device.

22. The method of claim 1, wherein the computer provides a single measurement point for the interarrival times.

23. An article comprising non-transitory computer-readable memory encoded with computer-readable code that causes a computer to monitor a data network for VoIP packets from a sending communication device to a receiving communication device, compute interarrival times of consecutive VoIP (Voice over IP) packets, form a histogram or diagram of the interarrival times, and perform pattern recognition on the histogram or diagram to determine whether the sending device was a root cause of a transmission error, each interarrival time computed as a difference in receipt times between two consecutive packets.

24. The article according to claim 23, wherein the pattern recognition identifies insufficient bandwidth and/or finite size of a buffer store of active network components as the root cause instead of the sending device.

25. The article according to claim 23, wherein the pattern recognition is performed on the diagram, which includes a temporal progression of the interarrival times.

26. The article according to claim 23, wherein the pattern recognition is performed on the histogram, which includes a frequency distribution of the interarrival times.

27. The article according to claim 26, wherein at least one of operating system of the sending communication device, configuration of the sending communication device, and a stack of the sending communication device are ascertainable from the pattern recognition.

28. The article according to claim 26, wherein a problem with an RTP stack of the sending device is ascertainable from the pattern recognition.

29. The article according to claim 23, wherein a problem with an operating system of the sending device is ascertainable from the pattern recognition.

30. The article according to claim 23, wherein a problem with configuration of the sending device is ascertainable from the pattern recognition.

31. A communication system comprising a supervisory computer programmed to monitor a data network for VoIP (Voice over IP) packets sent from a sending communication device to a receiving communication device, compute receive interarrival times of consecutive packets, form at least one of a histogram and a diagram of the receive interarrival times, and perform pattern recognition on the diagram or histogram to identify any sender problems in the sending device.

32. The system according to claim 31, wherein a receipt time is indicated by a measurement timestamp.

33. The system according to claim 31, wherein each receive interarrival time is computed as a difference in receipt times between two consecutive received packets having consecutive sequence numbers.

34. The system according to claim 31, wherein the diagram is a temporal progression of the interarrival times and wherein the histogram is a frequency distribution of the interarrival times.

35. The system according to claim 34, wherein the pattern recognition is performed on the histogram.

36. The system according to claim 34, wherein the pattern recognition is performed on the diagram.

37. The system according to claim 31, wherein the pattern recognition identifies configuration of the sending device as a sender problem.

38. The system according to claim 31, wherein the pattern recognition identifies an operating system of the sending device as a sender problem.

39. The system according to claim 31, wherein at least one of operating system of the sending communication device, configuration of the sending communication device, and a stack of the sending communication device are ascertainable as the sender problem.

40. The system according to claim 31, wherein the pattern recognition identifies an RTP (Realtime Transport Protocol) stack in the sending device as a sender problem.

41. The system according to claim 31, wherein the pattern recognition identifies insufficient bandwidth and/or finite size of a buffer store of active network components.

42. The system according to claim 31, wherein the pattern recognition reveals errors that caused impact on VoIP speech quality.

43. A method comprising using a computer to:

receive VoIP (Voice over IP) packets;

compute receive interarrival times of consecutive received VoIP packets, each receive interarrival time computed as a difference in receipt times between two consecutive received packets having consecutive sequence numbers;

form a histogram or diagram of the receive interarrival times; and

perform pattern recognition on the histogram or diagram to identify a cause of an error in a sending communication device;

wherein an operating system of the sending communication device, configuration of the sending communication device, and a stack of the sending communication device are ascertainable as the cause of the error.