Impact of mobility on the performance of context-aware applications using floating content

The growth of mobile computing and the evolution of smart user devices are progressively driving applications towards “contextawareness”, i.e., towards behaviors that change according to variations in context. Such applications use information that is restricted in space and time, making their communication requirements very different from those of conventional applications, so that opportunistic schemes are better suited to this case than more conventional communications. In this work we consider an opportunistic communication scheme called “Floating Content” (FC), which was specifically designed for server-less distributed context-aware applications, and we refine our previous investigation of the viability of FC for context-aware mobile applications, by considering the impact of different mobility models on the performance of FC. In particular, we consider four different mobility models, and, by using extensive simulation experiments, we investigate the performance of three different categories of context-aware applications that use FC. We also compare the simulation results to the performance predictions of our previously proposed simple analytical model. Results show that good performance can be achieved in content distribution by using FC under a variety of mobility models. They also show that a simple analytical model can provide useful performance predictions even for complex and realistic mobility models, although some application-specific characteristics might call for specialized models to improve the accuracy of performance estimates.