Skip to Main Content
Autonomic smart routing has previously been used in wired cognitive packet networks in order to achieve and maintain quality of service goals by continuously monitoring QoS metrics and adapting routes accordingly. In this paper we employ aspects of this routing protocol for use in wireless sensor networks with many devices of limited energy, memory, and processing capabilities. Through simulations we examine the use of smart routing in wireless sensor networks for static and mobile networks consisting of hundreds of devices, and show that it is an effective means to minimise the transmission power use. The locality of motes is also used to aggregate sensor readings for further transmission power reductions. These concepts are implemented as the tinyCPN routing protocol for the popular tinyOS sensor network operating system, and experimental results from a real sensor network testbed are presented.