System and Method for Estimating an Effective Bandwidth

• US 20140169192A1
• Filed: 06/18/2013
• Published: 06/19/2014
• Est. Priority Date: 12/13/2012
• Status: Active Grant

• Current Assignee: Huawei Technologies Co., Ltd. (Huawei Investment & Holding Co., Ltd.)
• Original Assignee: Futurewei Technologies Incorporated (Huawei Investment & Holding Co., Ltd.)
• Inventors: Zhang, Hang, Cheng, Ho-Ting
• Granted Patent: US 10,091,675 B2
• Time in Patent Office: Days
• US Class Current: 370/252
• CPC Class Codes: H04W 24/08 Testing, supervising or monitoring using real traffic

Abstract

In one embodiment, a method for a method of estimating an effective bandwidth of a traffic source includes obtaining a first traffic pattern from a first traffic source. Also, the method includes setting a first effective bandwidth between a mean data rate of the first traffic source and a peak data rate of the first traffic source. Additionally, the method includes determining a first outage rate of the first traffic source in accordance with the first traffic pattern and the first effective bandwidth.

Claims

1. A method of estimating an effective bandwidth of a traffic source, the method comprising:

obtaining a first traffic pattern from a first traffic source;

setting a first effective bandwidth between a mean data rate of the first traffic source and a peak data rate of the first traffic source; and

determining a first outage rate of the first traffic source in accordance with the first traffic pattern and the first effective bandwidth.

2. The method of claim 1, further comprising adjusting the first effective bandwidth by a backoff margin.

3. The method of claim 2, further comprising calculating the backoff margin in accordance with real-time data feedback.

4. The method of claim 3, wherein the real-time data feedback comprises instantaneous traffic arrivals.

5. The method of claim 1, further comprising:

recording a trace of the first traffic pattern; and

determining whether the first traffic pattern is a match to a previously recorded traffic pattern.

6. The method of claim 5, wherein determining whether the first traffic pattern is a match is in accordance with traffic statistics.

7. The method of claim 6, wherein the traffic statistics comprise a mean data rate, a peak data rate, and an auto correlation.

8. The method of claim 5, wherein determining whether the first traffic pattern is a match to a previously recorded traffic pattern comprises determining whether an application tag of the first traffic pattern matches an application tag of a previously recorded traffic pattern.

9. The method of claim 5, wherein determining whether the first traffic pattern is a match to a previously recorded traffic pattern comprises identifying features in the first traffic pattern that are similar to features in a previously recorded traffic pattern.

10. The method of claim 5, wherein determining whether the first traffic pattern is a match to a previously recorded traffic pattern comprises performing deep packet inspection on the first traffic pattern.

11. The method of claim 5, wherein determining whether the first traffic pattern is a match to a previously recorded traffic pattern comprises:

assigning a traffic ID to the first traffic pattern; and

comparing the traffic ID of the first traffic pattern to a traffic ID of the previously recorded traffic pattern.

12. The method of claim 5, wherein determining whether the first traffic pattern is a match to a previously recorded traffic pattern comprises comparing statistics of the first traffic pattern to statistics of the previously recorded traffic pattern.

13. The method of claim 5, wherein recording the trace of the first traffic pattern comprises recording a type of a packet, a number of bits of the packet, and a timestamp for the packet for packets in the first traffic pattern.

14. The method of claim 5, wherein recording the trace of the first traffic pattern comprises recording an average bit rate of the first traffic pattern.

15. The method of claim 5, wherein recording the trace of the first traffic pattern comprises recording a weighted average bit rate of the first traffic pattern.

16. The method of claim 1, wherein determining an outage performance of the first traffic source comprises:

receiving a first traffic stream from the first traffic source at a data rate of the first effective bandwidth;

measuring the first outage rate for the first traffic stream; and

recording the first outage rate.

17. The method of claim 16, further comprising:

adjusting the first effective bandwidth in accordance with the outage performance;

generating a second traffic stream by the first traffic source;

measuring a second outage rate for the second traffic stream; and

recording the adjusted effective bandwidth and the second outage rate.

18. The method of claim 17, further comprising fitting a curve to the first outage rate, the second outage rate, and the first effective bandwidth.

19. The method of claim 1, further comprising:

selecting a playback rate; and

determining a size of a playback buffer.

20. The method of claim 1, further comprising:

obtaining a plurality of traffic patterns from a plurality of traffic sources;

determining a plurality of outage rates of the plurality of traffic sources in accordance with the plurality of traffic patterns and the first effective bandwidth; and

storing the plurality of outage rates.

21. The method of claim 1, further comprising:

estimating a second effective bandwidth of a second traffic source;

estimating a third effective bandwidth of a traffic mix, wherein the traffic mix comprises traffic from the first traffic source and traffic from the second traffic source; and

determining a multiplexing gain in accordance with the first effective bandwidth, the second effective bandwidth, and the third effective bandwidth.

22. The method of claim 21, further comprising estimating a fourth effective bandwidth of a third traffic source, wherein the traffic mix further comprises traffic from the third traffic source.

23. The method of claim 22, further comprising estimating a plurality of effective bandwidths of a plurality of traffic sources, wherein the traffic mix further comprises traffic from the plurality of traffic sources.

24. The method of claim 23, wherein estimating the third effective bandwidth comprises scheduling packets of the traffic mix.

25. The method of claim 23, wherein estimating the third effective bandwidth of a traffic mix comprises:

obtaining an effective bandwidth of the traffic mix, wherein the effective bandwidth of the traffic mix is greater than a sum of a mean data rate of the first traffic source and a mean data rate of the second traffic source, and wherein the effective bandwidth is less than a sum of a peak data rate of the first traffic source and a peak data rate of the second traffic source;

obtaining the first traffic pattern of the first traffic source;

obtaining a second traffic pattern of the second traffic source;

determining the first outage rate of the traffic mix in accordance with the first traffic pattern of the first traffic source and the second traffic pattern of the second traffic source;

recording the effective bandwidth of the traffic mix; and

recording the first outage rate of the traffic mix.

26. The method of claim 25, further comprising:

adjusting the effective bandwidth of the traffic mix in accordance with the first outage rate of the traffic mix;

obtaining a third traffic pattern of the first traffic source;

obtaining a fourth traffic pattern of the second traffic source;

determining a second outage rate of the traffic mix in accordance with the third traffic pattern of the first traffic source and the fourth traffic pattern of the second traffic source;

recording the adjusted effective bandwidth of the traffic mix; and

recording the second outage rate of the traffic mix.

27. A gateway comprising:

a processor; and

a computer readable storage medium storing programming for execution by the processor, the programming including instructions to

obtain a traffic pattern from a traffic source,

set an effective bandwidth between a mean data rate of the traffic source and a peak data rate of the traffic source, and

determine an outage rate of the traffic source in accordance with the traffic pattern and the effective bandwidth.