MODEL-DRIVEN PROTOTYPING SUPPORT FOR PERVASIVE HEALTH CARE APPLICATIONS

Pervasive health care systems help to improve elderly and needy persons’ habitability by assisting them in living autonomously,and letting them participate in social communities and family life. The data gained from many wireless sensors running on different sensor platforms is usually further processed and interpreted by machine learning and pattern recognition components. The complexity of these systems stems from different types of environmental and vital parameters, different sampling rates, heterogeneous sensor platforms, unreliable network connections, as well as different programming languages that must be tailored to the use-case and the application environment. Therefore, the development of such applications often requires a lot of prototyping work, because significant data must be gained from individuals and from the environment through experiments and machine learning components must be trained for specific situations. In this paper we present a model driven prototyping approach for the development of pervasive health care applications. We provide tools like graphical editors and generators to simplify the development of pervasive health application prototypes that typically span multiple platforms.