Tao Liu, Alberto E. Cerpa
Link quality estimation is a fundamental component of the low power wireless network protocols and is essential for routing protocols in Wireless Sensor Networks (WSNs). However, accurate link quality estimation remains a challenging task due to the notoriously dynamic and unpredictable wireless environment. In this paper, we argue that in addition to the estimation of current link quality, prediction of the future link quality is more important for the routing protocol to establish low cost delivery paths. We pro- pose to apply machine learning methods to predict the link quality in the near future to facilitate the utilization of intermediate links with frequent quality changes. Moreover, we show that by using on-line learning methods, our adaptive link estimator (TALENT) adapts to network dynamics better than statically trained models without the need of a priori data collection for training the model before deployment. We implemented TALENT in TinyOS with Low-Power Listening (LPL) and conducted extensive experiments in three testbeds. Our experimental results show that the addition of TALENT increases the delivery efficiency 1.95 times on average compared with 4B, the state of the art link quality estimator, as well as improve the end-to-end delivery rate when tested on three different wireless testbeds.
Tao Liu, Alberto E. Cerpa, "TALENT: Temporal Adaptive Link Estimator with No Training," Proceedings of the Tenth ACM conference on Embedded Network Sensor Systems (SenSys 2012), pp. 253--266, Toronto, Ontario, Canada, November, 2012.
@Conference{Liu12a,
author = "Tao Liu and Alberto E. Cerpa",
booktitle = "Proceedings of the Tenth ACM conference on Embedded
Network Sensor Systems (SenSys 2012)",
title = "{TALENT}: Temporal Adaptive Link Estimator with No
Training",
pages = "253--266",
address = "Toronto, Ontario, Canada",
year = "2012",
month = nov,
URL = "http://www.andes.ucmerced.edu/papers/Liu12a.pdf",
accept = "18",
cited = "4",
}