A Wireless Actuator-Sensor Neural Network for Evacuation Routing

A wireless sensor-actuator network is formed by nodes capable of sensing and acting upon its environment. Typical challenges in designing such networks include distributed signal processing, synchronisation and communication, as well as deployment of network nodes and scalable architectures for these networks. In this paper, we look at the application of neural networks to individual nodes in a wireless network, which result in a wireless sensor-actuator neural network model. We explain how the combination of sensor network and neural network technology can describe a system which maintains desired characteristics such as scalability and adaptability. We show further that such model can be successfully applied to an evacuation routing scenario.