The Internet of Things (IoT) and Machine-to-Machine (M2M) revolutions are quietly coming, and with them an epochal turning point in the way people interact with the “things” surrounding them: appliances in a smart home, snow-level sensors in a smart ski resort, overﬂow sensors in a smart reﬁnery, etc.
Standardization bodies are playing a key role in this revolution. Different working groups are ﬁnalizing the protocols running at different levels of a communications stack we have developed, called OpenWSN for use as the de-facto protocol stack for tomorrow’s capillary networks.
This article gives an overview of the protocol stack, as well key integration details and the platforms and tools developed around it.
The pure C OpenWSN stack was ported to four off-the-shelf platforms representative of hardware currently used, from older 16-bit micro-controller to state-of-the-art 32-bit Cortex-M architectures.
The tools developed around the low-power mesh networks include visualization and debugging software, a simulator to mimic OpenWSN networks on a PC, and the environment needed to connect those networks to the Internet.
Experimental results presented in this article include a network where motes operate at an average radio duty cycle well below 0.1% and an average current draw of 68µA on off-the-shelf hardware.
These ultra-low power requirements enable arange of applications, with motes perpetually powered by micro-scavenging devices. OpenWSN is, to the best of our knowledge, the ﬁrst open-source implementation of the IEEE802.15.4e.
(**Coauthors of this paper were Xavier Vilajosana, Branko Kerkez, Fabien Chraim, Kevin Weekly, Qin Wang and Kris Pister, Linear Technology Dust Networks Division, and Steven Glaser, University of California, Berkeley
To read this external content in full, download the paper from the author archives at the University of California, Berkeley. http://robotics.eecs.berkeley.edu/~pister/publications/2013/openwsnETT