Accuracy analysis of distributed simulation systems

Existed simulation works always emphasize on procedural verification, which put too much focus on the simulation models instead of simulation itself. As a result, researches on improving simulation accuracy are always limited in individual aspects. As accuracy is the key in simulation credibility assessment and fidelity study, it is important to give an all-round discussion of the accuracy of distributed simulation systems themselves. First, the major elements of distributed simulation systems are summarized, which can be used as the specific basis of definition, classification and description of accuracy of distributed simulation systems. In Part 2, the framework of accuracy of distributed simulation systems is presented in a comprehensive way, which makes it more sensible to analyze and assess the uncertainty of distributed simulation systems. The concept of accuracy of distributed simulation systems is divided into 4 other factors and analyzed respectively further more in Part 3. In Part 4, based on the formalized description of framework of accuracy analysis in distributed simulation systems, the practical approach are put forward, which can be applied to study unexpected or inaccurate simulation results. Following this, a real distributed simulation system based on HLA is taken as an example to verify the usefulness of the approach proposed. The results show that the method works well and is applicable in accuracy analysis of distributed simulation systems.

[1]  Kathleen V. Diegert,et al.  Error and uncertainty in modeling and simulation , 2002, Reliab. Eng. Syst. Saf..

[2]  Zhang Xiong,et al.  Semantic modeling in satellite network simulation , 2008, 2008 Asia Simulation Conference - 7th International Conference on System Simulation and Scientific Computing.

[3]  William L. Oberkampf,et al.  Estimation of Total Uncertainty in Modeling and Simulation , 2000 .

[4]  Gu Xue-mai Design of Multilayered Satellite Communication Networks and Geometrical Analysis of ISLs , 2005 .

[5]  Richard E. Nance,et al.  Perspectives on the Evolution of Simulation , 2002, Oper. Res..