285 - Modelling and Simulation Tools for Research in Emerging Multiservice Telecommunications

Simulation for research and design accounts for a substantial fraction of engineering computation, especially in the context of today’s fully distributed and autonomously managed broadband networks. With valid and credible models, simulation can be made dramatically more cost-effective than real experiments, which may be expensive, dangerous, or, sometimes impossible to run.
The main objective of this Action was to enhance existing and develop new modelling and simulation tools for research in emerging multi-service telecommunications networks.
One of the main aspects addressed by the Action was model performance improvement.

The output of the corresponding activities is threefold. First, an overview of the shortcomings of commercially available simulation programs was provided and some concepts and tools to alleviate the shortcomings were identified. Second, one particular shortcoming was addressed, namely the lack of software to statistically assess simulation data built in into simulation programs. The COST Action 285 proposes a common data format for exchanging data such that simulation results can be imported in an off-the-shelf statistical analysis tool. Third, important progress has been made in developing models for multiplexing multimedia traffic, by taking into account that different protocols behave very differently at different time scales.

COST Action 285 performed as well a study on multilayer traffic models, focusing on TCP (Transmission Control Protocol) and developed methods to evaluate and validate new modeling tools. In particular, the Action defined and documented a framework for credibility assessment, tackling specific sub-problems, such as best performing pseudo random generators.

Furthermore, during the course of COST Action 285 an additional task on network emulation and tomography was introduced. In this context, some exploratory work was performed in network emulation, that is a step between simulation and implementation, and network tomography, namely in studying a network’s internal characteristics using information derived from end-point data.

Last, but not least, COST Action 285 made important contributions in terms of dissemination of the results, especially outreaching to the network and traffic simulator industry.