| Abstract: |
Mobility models have become essential tools for networking performances evaluation. Yet, evaluations are still done with unrealistic models. To circumvent this flaw, we propose Ghost, a
meta-modeling approach (and associated tool) to federate the design of more realistic mobility models. Ghost allows defining (i) simple mobility scenario using a high-level language (ii) or more
complex mobility models (iii) that can be calibrated and validated with empirical measurements. We illustrate the use of Ghost with models of indoor and outdoor mobility. Next, we present SIMPS,
a mobility model of human crowd motion based on social premises. This model is one of the first to tackle the roots governing mobility. Eventually, we present our latest findings on mobility
measurements and mobility modeling. We conclude with the presentation of FLmm, a Framework for Large-scale mobility measurements. |
