Multiplex communications – Data flow congestion prevention or control – Control of data admission to the network
Reexamination Certificate
1999-07-01
2003-07-22
Chin, Wellington (Department: 2664)
Multiplex communications
Data flow congestion prevention or control
Control of data admission to the network
C370S236000
Reexamination Certificate
active
06597660
ABSTRACT:
THE FIELD OF THE INVENTION
This invention relates in general to traffic analysis on packet networks. Packet networks are telecommunication networks in which the information is transmitted in small binary groups called packets. An advantage of packet networks is that it can handle different sources simultaneously by processing the packets sequentially. The packets travel through the network via a fast synchronous carrier; this can be viewed as a train of pulses which transport the packets. The speed of this carrier determines the maximum speed of the packet network. These networks can process only one packet at a time and thus the order in which the packets are processed depends on the priorities and quality of service required by the sources. Packets from a given source are mixed with packets from other sources. Each packet has a header that identifies its destination. Once the packets arrive at their destination, the headers are removed and the information is reassembled. Since these networks generally carry traffic from different types of sources which demand different service levels from the network, it is important from a network operation and management point of view to know the characteristics of the expected traffic. In the standards for packet networks that use the asynchronous transfer mode protocol (ATM), the traffic is described by first order statistics such as the peak cell rate, cell delay variation tolerance, sustainable cell rate, and maximum burst size. The size of the packets in ATM networks (that is networks which utilize the ATM protocol) is 53 bytes, of which 5 bytes compose the header and the remaining 48 bytes contain a section of the information being transmitted. The rate is defined as the number of packets that flow through the network in a given unit of time; this is a measure of the speed of the telecommunications network. The peak cell rate is defined as the inverse of the minimum time between successive packet arrivals to a switch. The cell delay variation tolerance is defined as the sensitivity of the information to changes in delay of the packets as they flow through the network. The sustainable cell rate is defined as the maximum average rate. The maximum burst size is defined as the maximum number of cells at the peak rate. Two of these parameters, the peak cell rate and the sustainable cell rate, have been defined as mandatory traffic parameters (or descriptors) in the ATM Forum UNI version 3.0 standards, as explained by McDysan and Spohn [11].
There are two types of networks which differ according to the way a connection is handled. The first type is connection-oriented. In these networks it is required to set up several parameters before any data transmission can take place. This is a process in which a source negotiates a level of service with the network. An end-to-end path with a quality of service is established and all the packets from the source will follow this path. The second type is connectionless in which it is not required to set up an end-to-end connection. The network handles the packets individually.
Quality of service (QoS) is a parameter meaningful from a source to a destination point of view, as well as at each link in the network. In connection-oriented networks, the negotiation for a QoS is carried out by agreeing on certain parameters. These parameters are based on first order statistical measures of the performance, such as the average delay, cell delay variation, error rates, and different levels of packet loss, as explained in the Bellcore requirements for broadband switching systems [2].
Packet networks have been designed to carry traffic from multimedia sources, among them different types of video and audio, voice, and data communications. Each traffic source presents the network with different requirements. The network must be able to handle all these traffic sources at their respective quality of service. The problem that arises is how to accurately characterize the different traffic sources for efficient network utilization. It might, in fact, be required to measure quality in different ways for different traffic sources. This makes the performance measurement problem very complex.
In our invention, a method is used to characterize the traffic in real-time. The method is used to calculate traffic descriptors considering properties of the traffic which have not been considered previously in commercial equipment. The descriptors are based on properties of the traffic that have been reported in the literature, but the techniques available are not suitable for real-time measurements. The algorithm presented is based on the simultaneous measurement of the traffic at different time scales. The data is represented in an appropriate form, processed and organized in an array of vectors. From this array, higher order statistical measures are derived. The traffic descriptors calculated in this way are used to characterize the traffic. The algorithm is implemented in real-time. The information is also used for traffic classification and performance prediction.
THE DESCRIPTION OF RELATED ART
The following references have been identified in a search in this field, some of which are relevant to the present invention:
Publications
[1] R. Addie, M. Zukerman, and T. Neame, “Fractal Traffic: Measurements, Modelling and Performance Evaluation”, in Proc. IEEE Infocom, pp. 977-984, 1995.
[2] Bellcore, “Broadband Switching System Generic Requirements”, GR-1110-CORE, Revision 3, April 1996.
[3] J. Beran, R. Sherman, M. Taqqu, and W. Willinger, “Long-Range Dependence In Variable-Bit-Rate Video Traffic”, in IEEE Trans. on Communications, vol. 43, no. 4, pp. 1566-1579, April 1995.
[4] Y. Chen, Z. Deng, and C. Williamson, “A Model for Self-Similar Ethernet LAN Traffic: Design, Implementation, and Performance Implications”, internal report, University of Saskatchewan, Canada, 1995.
[5] M. Devetsikiotis, I. Lambadaris, R. Kaye, “Traffic Modeling and Design Methodologies for Broadband Networks”,
Canadian Journal on Electrical and Computer Engineering
”, vol. 20, no. 3, 1995.
[6] M. Garrett and W. Willinger, “Analysis, Modeling and Generation of Self-Similar VBR Video Traffic”, in Proc. ACM Sigcom, London, UK, pp. 269-280, 1994.
[7] R. Guerin, H. Ahmadi and M. Naghshineh, “Equivalent Capacity and its to Bandwidth Allocation in High Speed Networks”,
IEEE JSAC
, vol. 9, no. 7, 1991.
[8] C. Huang, M. Devetsikiotis, I. Lambadaris, and A. Kaye, “Modeling and Simulation of Self-Similar Variable Bit Rate Compressed Video: A Unified Approach”, in ACM Sigcom, Cambridge 1995.
[9] W. Lau, A. Erramilli, J. Wang, and W. Willinger, “Self-Similar Traffic Generation: The Random Midpoint Displacement Algorithm and its Properties”, in Proc. IEEE Int Conf. Commun., 1995.
[10] W. Leland, M. Taqqu, W. Willinger, and D. Wilson, “On The Self-Similar Nature of Ethernet Traffic (extended version)”,
IEEE/ACM Trans. Networking
, vol. 2, no. 1, pp. 1-15, February 1994.
[1] B. Mandelbrot, “Self-Similar Error Clusters in Communication Systems and the Concept of Conditional Stationarity”, in IEEE Trans. on Communication Technology, pp. 71-90, 1965.
[12] D. McDysan and D. Spohn, “ATM Theory and Application”. Toronto: McGraw-Hill, 1995.
[13] D. McLaren and D. Nguyen, “A Fractal-Based Source Model for ATM Packet Video”, in Int. Conf. on Digital Processing of Signals in Communications, Univ. of Loughbovough, September 1991.
[14] V. Paxson, “Fast Approximation of Self-Similar Network Traffic”, report LBL-36750, Univ. of California at Berkeley, Lawrence Berkeley Laboratory, 1995.
[15] A. Rueda and W. Kinsner, “A Survey of Traffic Characterization Techniques in Telecommunication Networks”,
Proc. IEEE Canadian Conference on Electrical and Computer Engineering
, pp. 830-833, May 1996.
U.S. Patents
U.S. Pat. No. 5,050,161 Congestion management based on multiple framing strategy
U.S. Pat. No. 5,274,625 Traffic me
Bourlas Yair
Diamond Jeffrey E.
McLeod Robert D.
Rueda Jose A.
Battison Adrian D.
Chin Wellington
Dupuis Ryan W.
Schultz William
Telecommunications Research Laboratory
LandOfFree
Method for real-time traffic analysis on packet networks does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for real-time traffic analysis on packet networks, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for real-time traffic analysis on packet networks will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3027044