Network-centric actuation control in sensor/actuator networks based on bio-inspired technologies

Self-organization mechanisms have been developed to efficiently operate networked embedded systems. Special focus was given to wireless sensor networks (WSN) and sensor/actuator networks (SANET). Looking at the most pressing issues in such networks, the limited resources and the huge amount of interoperating nodes, the proposed solutions primarily intend to solve the scalability problems by reducing the overhead in data communication. Well-known examples are data-centric routing approaches and probabilistic techniques. In this paper, we intend to go one step further. We are about to also move the operation and control for WSN and SANET into the network. Inspired by the operation of complex biological systems such as the cellular information exchange, we propose a network-centric approach. Our method is based on three concepts: data-centric operation, specific reaction on received data, and simple local behavior control using a policy-based state machine. In summary, these mechanisms lead to an emergent system behavior that allows to control the operation of even large-scale sensor/actuator networks.