Enabling Intelligence on a Wireless Sensor Network Platform

Conventional Wireless Sensor Networks (WSNs) usually adopt a centralised approach to data processing and interpretation primarily due to the limited computation and energy resources available on sensor nodes. These constraints limits the potential of intelligent techniques to data analy- sis and such activities on the centralised host. In contrast, Intelligent WSNs (iWSNs) will be significantly more powerful thus enabling the harnessing of intelligent techniques for diverse purposes. One such purpose is the practical realisation of smart environments, and facilitating mobility and interaction with the inhabitants of such environments. As a step in this direction, this paper presents the design of an iWSN sensor node platform that enables the hosting of lightweight Artificial Intelligence (AI) frameworks whilst enabling the ubiquitous energy constraints be quantified, mitigated and managed.