A Rule-based System for Programming Self-Organized Sensor and Actor Networks

We describe a programming scheme for massively distributed systems that are assumed to self-organize according to a given set of simple rules. The focus of this investigation is operation and control in Sensor and Actor Networks (SANETs). The main issues addressed by self-organization techniques are scalability, network lifetime, and real-time support. In the literature, biological principles are often cited as inspirations for technical solutions, especially in the domain of self-organization. We developed a system named Rule-based Sensor Network (RSN) according to the observed communication and control behavior in cellular communication. Cellular signaling cascades allow the event-specific reaction initiated by individual cells in collaboration with their direct neighbors. Information between cells are transmitted via proteins and result in the cascade of protein-protein or protein-DNA interactions to produce a specific cellular answer, e.g. the activation of cells or the transmission of mediators. These processes are programmed in every individual cell and lead to a coordinated reaction on a higher organization platform. We transferred these mechanisms to operation and control in SANETs. In particular, a rule-based processing scheme relying on the main concepts of cellular signaling cascades has been developed. It relies on simple local rules and provides problem specific reaction such as local actuation control and data manipulation. We describe this RSN technology and demonstrate comparative simulation results that show the feasibility of our approach.