Follow this innovation
[ via 
] [?]
Computation Of Burstiness Curve Of Multimedia Traffic
University of California System: University of California, Santa Cruz
posted on 01/17/2012
The rate variability of video sources has introduced the need for characterizing traffic so that the amount of resources to be allocated by a network (such as bandwidth, buffer space, etc.) can be estimated during the call admission control (CAC) process. Characterization of the traffic stream is also necessary for efficient policing of the traffic. One means of characterizing a traffic source is with its burstiness curve. Each point in a burstiness curve corresponds to the maximum queue size encountered (or the amount of buffering needed) when a traffic source is fed into a server with a deterministic service rate. The burstiness curve is useful in the optimal allocation of resources to satisfy a desired quality of service for a video stream in a packet network.
Suggested Uses
The high efficiency of the new algorithms makes them attractive for Quality of Service (QoS) provisioning in both off-line environments such as Video-on-Demand (VoD) servers and real-time applications such as live TV distribution systems. The algorithms are also useful for QoS provisioning in both IP and ATM networks. In addition, the algorithms can be used to characterize any bursty ON-OFF source, including voice and data.
Advantages
The algorithms are optimal in the number of points needed for an exact computation of the burstiness curve. They exhibit low time and space complexity compared to traditional simulation-based approaches.
See also UC Case 2000-138, Computation of Loss Curve of Multimedia Traffic.
Detailed Description
Scientists at the University of California have developed deterministic algorithms for exact computation of the burstiness curve for both elementary video streams and MPEG-2 transport streams. The algorithms exploit the piecewise linearity of the burstiness curve and compute only the points at which the slope of the burstiness curve changes. Approximate versions of these algorithms have also been developed, which save computational effort by considering only a small number of candidate points at which the slope of the burstiness curve may change.
File Number: 10169
| Patent Number(s): | 6795397 |
|---|---|
| Copyright: | ©2009-2011, The Regents of the University of California |
This innovation currently is not available for online licensing. Please contact University of California, Santa Cruz Office for Management of Intellectual Property at University of California System: University of California, Santa Cruz for more information.
request more info
Case Manager:
 |
University of California, Santa Cruz Office for Management of Intellectual Property
Innovations (35) |
|---|
ProfileFind more innovations
