L. Bölöni, D. Turgut, G. Wang, and D.C. Marinescu

Challenges and Benefits of Time-Parallel Simulation of Wireless Ad Hoc Networks


Cite as:

L. Bölöni, D. Turgut, G. Wang, and D.C. Marinescu. Challenges and Benefits of Time-Parallel Simulation of Wireless Ad Hoc Networks. In Proceedings of First International Conference on Performance Evaluation Methodologies and Tools (Valuetools-2006), October 2006.

Download:

Download 

Abstract:

In this paper we discuss time-parallel simulation of wireless ad hoc networks. Such systems are rarely amenable to analytical modelling due to the complexity of the models involved. Thus, to obtain a quantitative evaluation of such models we often resort to simulation. Yet, current simulation techniques do not allow us to study systems consisting of thousands of nodes for extended periods of time. We introduce the notion of perturbation of measurements and present a layer-by-layer analysis of the impact of perturbations on the functioning of the wireless network. This model allows us to predict the accuracy of the simulation for several measures of performance through an analysis of the protocols involved at the various layers. We also present an implementation of time-parallel simulation based on iterative extension of the warmup period. In a series of experiments, we find good concordance between the predicted findings and the results of the simulations.

BibTeX:

@inproceedings{Boloni-2006-VALUETOOLS,
   title = "Challenges and Benefits of Time-Parallel Simulation of Wireless Ad Hoc Networks",
   author = "L. B{\"o}l{\"o}ni and D. Turgut and G. Wang and D.C. Marinescu",
   booktitle = "Proceedings of First International Conference on Performance Evaluation Methodologies and Tools (Valuetools-2006)",
   year = "2006",
   month = "October",
   location = "Pisa, Italy",
      abstract = {
      In this paper we discuss time-parallel simulation of wireless ad hoc
      networks. Such systems are rarely amenable to analytical modelling due to
      the complexity of the models involved. Thus, to obtain a quantitative
      evaluation of such models we often resort to simulation. Yet, current
      simulation techniques do not allow us to study systems consisting of
      thousands of nodes for extended periods of time.
      We introduce the notion of perturbation of measurements and present a
      layer-by-layer analysis of the impact of perturbations on the functioning
      of the wireless network. This model allows us to predict the accuracy of
      the simulation for several measures of performance through an analysis of
      the protocols involved at the various layers.
      We also present an implementation of time-parallel simulation based on
      iterative extension of the warmup period. In a series of experiments, we
      find good concordance between the predicted findings and the results of
      the simulations.
   },
}

Generated by bib2html.pl (written by Patrick Riley, Lotzi Boloni ) on Fri Oct 06, 2017 18:15:24