Robust Programming for Large-Scale Sensor Networks
João Barros, Luis Lopes, DCC-FCUP

The ability to program large-scale networks of power constrained and computationally restricted sensing devices is widely recognized as one of the key challenges towards turning the sensor network vision into reality. In most of the current applications, the sensor nodes
are controlled by module-based operating systems such as TinyOS and are programmed using arguably somewhat ad-hoc languages e.g. nesC. Despite these undeniably useful practical developments, it is fair to state that current systems fall short of providing a fundamental formal model that captures the computational and communication properties of sensor networks.

The goal of this dissertation is to develop such a model use it as the basis for the development of robust and provably correct implementations of high-level programming languages for sensor hardware platforms.

The expected outcomes of this research are thus
(a) a process calculus,
(b) a programming language and
(c) implementation and testing of a complete programming framework.