Link capacity allocation by input power spectrum

Consider a multimedia input traffic stream, described by a stationary random process. Our previous study indicates that second-order input statistics, which are described by a power spectrum in the frequency domain, play a dominant role in queueing performance as compared with higher-order input statistics. In this paper, we show the significant effect of an input power spectrum on link capacity allocation in a finite-buffer system to achieve a desired performance criterion. The study also explores a fundamental limit of buffer sizing to improve link capacity when traffic contains significant low-frequency power.

[1]  D. Mitra,et al.  Stochastic theory of a data-handling system with multiple sources , 1982, The Bell System Technical Journal.

[2]  Maurizio Decina,et al.  Bandwidth allocation and selective discarding for variable bit rate video and bursty data calls in ATM networks , 1991, IEEE INFCOM '91. The conference on Computer Communications. Tenth Annual Joint Comference of the IEEE Computer and Communications Societies Proceedings.

[3]  Hamid Ahmadi,et al.  Equivalent Capacity and Its Application to Bandwidth Allocation in High-Speed Networks , 1991, IEEE J. Sel. Areas Commun..

[4]  Walter Willinger,et al.  On the self-similar nature of Ethernet traffic , 1993, SIGCOMM '93.

[5]  Luigi Fratta,et al.  ATM: bandwidth assignment and bandwidth enforcement policies , 1989, IEEE Global Telecommunications Conference, 1989, and Exhibition. 'Communications Technology for the 1990s and Beyond.

[6]  Murad S. Taqqu,et al.  On the Self-Similar Nature of Ethernet Traffic , 1993, SIGCOMM.

[7]  San-qi Li,et al.  Fundamental limits of input rate control in high speed network , 1993, IEEE INFOCOM '93 The Conference on Computer Communications, Proceedings.

[8]  Debasis Mitra,et al.  Effective bandwidth of general Markovian traffic sources and admission control of high speed networks , 1993, TNET.

[9]  B. Melamed,et al.  The transition and autocorrelation structure of tes processes , 1992 .

[10]  San-qi Li,et al.  Folding algorithm: a computational method for finite QBD processes with level-dependent transitions , 1994, IEEE Trans. Commun..

[11]  San-qi Li,et al.  Second order effect of binary sources on characteristics of queue and loss rate , 1994, IEEE Trans. Commun..

[12]  San-qi Li,et al.  Queue response to input correlation functions: continuous spectral analysis , 1993, TNET.

[13]  San-qi Li,et al.  On Input State Space Reduction and Buuer Noneeective Region , 1994 .

[14]  San-qi Li,et al.  Queue response to input correlation functions: discrete spectral analysis , 1993, TNET.

[15]  San-qi Li,et al.  Analysis of multi-media traffic queues with finite buffer and overload control. I. Algorithm , 1991, IEEE INFCOM '91. The conference on Computer Communications. Tenth Annual Joint Comference of the IEEE Computer and Communications Societies Proceedings.

[16]  San-qi Li,et al.  Analysis of multimedia traffic queues with finite buffer and overload control II. Applications , 1992, [Proceedings] IEEE INFOCOM '92: The Conference on Computer Communications.

[17]  San-qi Li,et al.  On input state space reduction and buffer noneffective region , 1994, Proceedings of INFOCOM '94 Conference on Computer Communications.