Premium
Analysing packetized voice and video traffic in an ATM multiplexer
Author(s) -
Rai Suresh,
Oh Young C.
Publication year - 1998
Publication title -
international journal of communication systems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.344
H-Index - 49
eISSN - 1099-1131
pISSN - 1074-5351
DOI - 10.1002/(sici)1099-1131(199807/08)11:4<225::aid-dac364>3.0.co;2-n
Subject(s) - computer science , queueing theory , multiplexer , poisson distribution , markov process , real time computing , poisson process , computer network , queue , superposition principle , markov chain , algorithm , asynchronous transfer mode , multiplexing , telecommunications , statistics , mathematics , mathematical analysis , machine learning
This paper considers packetized voice and video traffic incoming to an ATM multiplexer. Each voice and video source is approximated by a two‐state Markov–modulated Poisson process (MMPP(2)). We propose a method based on the asymptotic results of a renewal process to obtain the equivalent MMPP(2) parameters. The integrated voice and video sources are modelled by the superposition of two MMPP(2) processes, denoted MMPP(4). The analysis of the superposed traffic in terms of buffer occupancy, cell loss probability and delay is performed with an MMPP(4)/Er( k )/1/ B queuing model considering different buffer sizes and numbers of sources. Results of the proposed method are also compared with those of Heffes and Lucantoni ( IEEE J. Select. Areas Commun ., SAC‐4 , 856–868 (1986)) and with a Poisson approximation. © 1998 John Wiley & Sons, Ltd.