중요샘플링 시뮬레이션에 의한 ATM 노드의 성능평가

Performance Evaluation of ATM Node by Importance Sampling Simulation

  • 국광호 (서울산업대학교 산업공학과) ;
  • 이창호 (강남대학교 산업공학과) ;
  • 강성열 (한국전자통신연구원 교환기술연구단) ;
  • 오창환 (한국전자통신연구원 교환기술연구단)
  • 발행 : 1997.12.01

초록

The cell loss probability recommended in the B-ISDN is in the range $10^{-8}~10^{-12}$. When a simulation technique is used to analyze the performance of the ATM node, an enormous amount of computer processing time is required. In this study, we derive an importance sampling simulation technique that can be used to evaluate the performance of the ATM node very quickly, that is, the probability that the queue size at the ATM node reaches some large value N. The simulation results show that the backlog probability obtianed by the importance saimpling simulation is very close to that obtained by the ordinary simulation and the computer time can be reduced drastically by the importance sampling simulation.

키워드

참고문헌

  1. IEEE Trans. on Auto. Cont. v.AC-28 no.9 Large Deviations and Rare Events in the Study of Stochastic Algorithms M. Cottrel;J. C. Fort;G. Malgouyres
  2. ITC-13 Copenhagen, Proceedings Queueing Performance and Control in ATM Extrapolating ATM Simulation Results using Extreme Value Theory V. Dijk;E. Aanen;H. van den Berg;J. M. van Noortwijk
  3. Proc. IEEE Global Telecom. Conf., GLOBECOM '93 Stochastic Gradient Techniques for the Efficient Simulation of High Speed Networks using Important Sampling M. Devetsikiotis;W. A. Al-qaq;J. A. Freebersyser;J. K. Townsend
  4. IEEE Trans. on Auto. Cont. v.36 no.12 Optimally Efficient Estimation of the Statistics of Rare Events in Queueing Networks M. R. Frater;T. M. Lennon;B. D. O. Anderson
  5. Management Science v.31 no.11 Important Sampling for Stochstic Simulations P. W. Glynn;D. L. Iglehart
  6. IEEE JSAC v.SAC-2 no.1 Techniques for Estimating the Bit Error Rate in the Sinulation of Digital Communication Sustems M. C. Jeruchim
  7. Artech House, Inc. Asynchronous Transfer Mode Techniques R. O. Onvural
  8. IEEE Tra. on Auto. Con. v.34 no.1 Quick Simulation Method for Excessive Backlogs in Networks and Queues S. Parekh;J. Walland
  9. ITC 14 Quick Simulation of Stationary Tail Probabilities at a Packet Switch S. P. Parekh
  10. IEEE Tran. on Auto. Con. v.36 no.12 Large Deviations Theory and Efficient Simulation of Excessive Backlogs in a GI/GI/m Queue J. S. Sadowski
  11. ITC 13 RESTART : A Method for Accerating Rare Events Simulation M. J. Villen-Altamirano
  12. ETRI 위탁연구보고서 속산시뮬레이션에 의한 ATM 텔레트래픽 연구 국광호;이창호