An Agent-Based Data-Generation Tool for Situation-Aware Systems

This paper describes the design, implementation and evaluation of a simulator that generates plausible data for the debugging and testing of situation-aware monitoring systems. The use of simulated data is necessary because there is very little sensor-level behavioural-data available in the literature and, moreover, these data are very specific of the set-up used to collect them, i.e. the kind of sensors, the people behavior and age, the environment characteristics, and so on. Moreover, data collection requires a long time since the phenomena we are interested in last months or years. The use of a tool like Repast has substantially sped up the design and implementation phase while still achieving the necessary time performance and quality of the data generated.