A social aggregation based combination movement model for opportunistic networks

Simulations with trace dataset and synthetic movement models are the two main methods for measuring routing protocols and QoS schemes in Opportunistic Networks. But the trace dataset cannot meet the needs of all situations and it is notoriously difficult to be collected. Traditional synthetic movement models are easy to be used, but they do not capture the detailed mobile characteristics of the human. We propose a social aggregation based combination movement model for Opportunistic Networks simulations. In our model, we use several adjustable random distribution generators to generate time-variant mobility parameters and build multiple sub-models to describe more detailed human behavior. By combining those sub-models, which based on a reasonable combination mechanism, one can reproduce complex human activities. So, routing protocols and applications can be tested on our model. We compare our model with different trace datasets and traditional movement models in the Opportunistic Networks Environment (ONE) simulator. Simulation results demonstrate that the model has higher accuracy than traditional movement models and its mobility pattern approximates to real trace datasets.