Dynamic and autonomous channel adaptation in wireless sensor networks

Wireless Sensor Networks (WSNs) are often deployed in places where they have to share the radio frequency spectrum with other wireless devices. Functioning and performance of WSNs can be harmfully influenced by such near-by devices. Consequently, there is a strong need for WSNs to provide capabilities for autonomous adaptation to possible interference in the wireless medium. In this paper we propose two general-purpose distributed protocols that can dynamically adapt the transmission channel of a WSN according to the state of the RF spectrum. They exploit scanning abilities of radios embedded on individual nodes to detect local interference and use beacons to notify other nodes in order to build a global view of the spectrum state enabling a local decision for the best channel to use. The algorithms do not require any form of time synchronization. We evaluate the proposed protocols through simulations and show that they significantly outperform existing solutions by providing better reliability and convergence time. We also implement them on real sensor nodes and justify our simulation results with real-life experiments.