A Framework for Opportunistic Context and Activity Recognition
Sprache des Titels:
9th International Conference on Pervasive Computing (Pervasive2011)
Opportunistic activity and context recognition systems draw from the characteristics to use sensing devices that just happen to be available rather than pre-defining a fixed sensor infrastructure at design time. Opportunistic sensing offers the possibility to obtain data from sensors that just happen to be available in the area surrounding the user. This enables users/or applications to state recognition goals, saying what has to be sensed for, at runtime to the system. The available sensing devices that can contribute to the recognition goal are configured to an ensemble, the best set of sensors for the recognition goal. This paper describes the OPPORTUNITY Framework and shows its functionality with respect to four features (goal querying and sensor configuration, sensor appears/disappears, sensor modifies data and sensor learns from other sensor) to show the dynamic nature of an opportunistic system as the available sensing infrastructure is not fixed and changes during runtime.