Method and apparatus for performing quality-of-service calculations on packet-based networks
View Patent ↗A method and apparatus for performing Quality-of-Service (QoS) calculations on packet-based communications networks using a QoS measure which is based on data included in non-lost packets, as well as on data included in lost packets, when the proper interpretation of the data in non-lost packets depends upon data in one of the lost packets. Two new QoS measures that address the limitations inherent in the prior art PLR (Packet Loss Rate) measure are introduced. The Packet Loss Distortion Rate (PLDR) measure determines both packets which are lost, as well as packets whose proper interpretation depends on one or more packets which have been lost. The Media Distortion Rate (MDR) measures the actual quantity of media material that is lost, regardless of how the material is grouped into individual packets.
1. A method for calculating a Quality-of-Service measure in a packet-based communications network under packet loss conditions, each packet of said packet-based communications network comprising packet header information and data, the method comprising the steps of:
identifying, over a given time interval, one or more lost packets;
identifying, over said given time interval, one or more received packets which comprise data whose proper interpretation is based on data comprised in one or more of said identified lost packets; and
calculating said Quality-of-Service measure for said given time interval based on both said identified lost packets and on said identified received packets which comprise data whose proper interpretation is based on data comprised in one or more of said identified lost packets.
2. The method of claim 1 wherein said Quality-of-Service measure for said given time interval is calculated based on a number of said identified lost packets over said given time interval and on a number of said identified received packets which comprise data whose proper interpretation is based on data comprised in one or more of said lost packets over said given time interval.
3. The method of claim 2 wherein said Quality-of-Service measure for said given time interval is calculated by computing a ratio of
(i) a summation of said number of said identified lost packets over said given time interval and said number of said identified received packets which comprise data whose proper interpretation is based on data comprised in one or more of said lost packets over said given time interval, to
(ii) a summation of said number of said number of said identified lost packets over said given time interval and a total number of received packets over said given time interval.
4. The method of claim 1 wherein said Quality-of-Service measure for said given time interval is calculated based on a number of data samples comprised in said identified lost packets over said given time interval and on a number of data samples which are comprised in said identified received packets and for which proper interpretation of said data samples is based on said data samples comprised in one or more of said lost packets over said given time interval.
5. The method of claim 4 wherein said Quality-of-Service measure for said given time interval is calculated by computing a ratio of
(i) a summation of said number of data samples comprised in said identified lost packets over said given time interval and said number of data samples which are comprised in said identified received packets and for which proper interpretation of said data samples is based on said data samples comprised in one or more of said lost packets over said given time interval, to
(ii) a summation of said number of data samples comprised in said identified lost packets over said given time interval and a total number of data samples comprised in packets received over said given time interval.
6. The method of claim 4 wherein said data samples comprise speech samples.
7. The method of claim 4 wherein said data samples comprise video pixels.
8. The method of claim 4 wherein said Quality-of-Service measure for said given time interval is calculated by computing a ratio of
(i) a weighted summation of said number of data samples comprised in said identified lost packets over said given time interval and said number of data samples which are comprised in said identified received packets and for which proper interpretation of said data samples is based on said data samples comprised in one or more of said lost packets over said given time interval, to
(ii) a weighted summation of said number of data samples comprised in said identified lost packets over said given time interval and a total number of data samples comprised in packets received over said given time interval,
wherein each of said data samples has associated therewith a corresponding weighting factor with which said data sample is multiplied thereby prior to performing said weighting summations.
9. The method of claim 1 wherein said Quality-of-Service measure for said given time interval is calculated as a burst ratio, wherein the burst ratio is based on a number of said identified lost packets over said given time interval and on a number of said identified received packets which comprise data whose proper interpretation is based on data comprised in one or more of said lost packets over said given time interval.
10. The method of claim 9 wherein the burst ratio comprises a ratio of
(i) an average length of observed bursts over said given time interval, each of said observed bursts comprising a contiguous sequence of packets each of which is one of said identified lost packets over said given time interval or one of said identified received packets which comprise data whose proper interpretation is based on data comprised in one or more of said lost packets over said given time interval, to
(ii) an average length of observed bursts expected over said given time interval in accordance with a random loss model for said packet-based communications network.
11. A computer readable medium having computer executable instructions for calculating a Quality-of-Service measure in a packet-based communications network under packet loss conditions, each packet of said packet-based communications network comprising packet header information and data, said data not including said packet header information, the computer executable instructions comprising steps to:
identify, over a given time interval, one or more lost packets;
identify, over said given time interval, one or more received packets which comprise data whose proper interpretation is based on data comprised in one or more of said identified lost packets; and
calculate said Quality-of-Service measure for said given time interval based on both said identified lost packets and on said identified received packets which comprise data whose proper interpretation is based on data comprised in one or more of said identified lost packets.
12. The computer executable instructions of claim 11 wherein said Quality-of-Service measure for said given time interval is calculated based on a number of said identified lost packets over said given time interval and on a number of said identified received packets which comprise data whose proper interpretation is based on data comprised in one or more of said lost packets over said given time interval.
13. The computer executable instructions of claim 12 wherein said Quality-of-Service measure for said given time interval is calculated by computing a ratio of
(i) a summation of said number of said identified lost packets over said given time interval and said number of said identified received packets which comprise data whose proper interpretation is based on data comprised in one or more of said lost packets over said given time interval, to
(ii) a summation of said number of said number of said identified lost packets over said given time interval and a total number of received packets over said given time interval.
14. The computer executable instructions of claim 11 wherein said Quality-of-Service measure for said given time interval is calculated based on a number of data samples comprised in said identified lost packets over said given time interval and on a number of data samples which are comprised in said identified received packets and for which proper interpretation of said data samples is based on said data samples comprised in one or more of said lost packets over said given time interval.
15. The computer executable instructions of claim 14 wherein said Quality-of-Service measure for said given time interval is calculated by computing a ratio of
(i) a summation of said number of data samples comprised in said identified lost packets over said given time interval and said number of data samples which are comprised in said identified received packets and for which proper interpretation of said data samples is based on said data samples comprised in one or more of said lost packets over said given time interval, to
(ii) a summation of said number of data samples comprised in said identified lost packets over said given time interval and a total number of data samples comprised in packets received over said given time interval.
16. The computer executable instructions of claim 14 wherein said data samples comprise speech samples.
17. The computer executable instructions of claim 14 wherein said data samples comprise video pixels.
18. The computer executable instructions of claim 14 wherein said Quality-of-Service measure for said given time interval is calculated by computing a ratio of
(i) a weighted summation of said number of data samples comprised in said identified lost packets over said given time interval and said number of data samples which are comprised in said identified received packets and for which proper interpretation of said data samples is based on said data samples comprised in one or more of said lost packets over said given time interval, to
(ii) a weighted summation of said number of data samples comprised in said identified lost packets over said given time interval and a total number of data samples comprised in packets received over said given time interval,
wherein each of said data samples has associated therewith a corresponding weighting factor with which said data sample is multiplied thereby prior to performing said weighting summations.
19. The computer executable instructions of claim 11 wherein said Quality-of-Service measure for said given time interval is calculated as a burst ratio, wherein the burst ratio is based on a number of said identified lost packets over said given time interval and on a number of said identified received packets which comprise data whose proper interpretation is based on data comprised in one or more of said lost packets over said given time interval.
20. The computer executable instructions of claim 19 wherein the burst ratio comprises a ratio of
(i) an average length of observed bursts over said given time interval, each of said observed bursts comprising a contiguous sequence of packets each of which is one of said identified lost packets over said given time interval or one of said identified received packets which comprise data whose proper interpretation is based on data comprised in one or more of said lost packets over said given time interval, to
(ii) an average length of observed bursts expected over said given time interval in accordance with a random loss model for said packet-based communications network.