A Programming Platform for Future Wireless Sensor Networks
ProFuN was a research project funded by SSF, the Swedish Foundation for Strategic Research, 2009 - June 2015.
Our goal has been to provide a programming environment where sensors can be programmed as an ensemble rather than individually. In this environment programmers focus on the applications and on the services provided by the collection of sensors rather than on which particular sensor to program or on how communication takes place. Energy levels and scheduling to some extent are controlled by the programmers, e.g., to give priority to specific application tasks. The approach we chose is to employ macro programming of networks of sensors rather than micro programming of individual sensors. Our research focus has been on wireless sensor networks, but the general concept of ensemble programming is applicable to other multiprocessor areas such as grid computing, multicore, and server farms, like Google, etc. Through the ProFuN TG (Task Graph) tool we have demonstrated that sensor networks can be programmed as an ensemble under severe constraints, including being devices with very limited capabilities, frequently failing sensors, mobility, energy constraints and harsh security threats.