WizSync: Exploiting Wi-Fi Infrastructure for Clock Synchronization in Wireless Sensor Networks
Authors:
- 1 June 2014
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 13 (6) , 1379-1392
Abstract
Time synchronization is a fundamental service for wireless sensor networks (WSNs). Although a number of message passing protocols can achieve satisfactory synchronization accuracy, they suffer poor scalability and high transmission overhead. An alternative approach is to utilize the global time references such as those induced by GPS and timekeeping radios. However, they require the hardware receiver to decode the out of band clock signal, which introduces extra cost and design complexity. This paper proposes a novel WSN time synchronization approach by exploiting the existing Wi-Fi infrastructure. Our approach leverages the fact that 802.15.4 sensors and Wi-Fi nodes often occupy the same or overlapping radio frequency bands in the 2.4 GHz unlicensed spectrum. As a result, a 802.15.4 node can detect and synchronize to the periodic beacons broadcasted by Wi-Fi access points (APs). A key advantage of our approach is that, due to the long communication range of Wi-Fi, a large number of 802.15.4 sensors can synchronize clock rates to the same beacons without any message exchange. This paper makes several key contributions. First, we experimentally characterize the spatial and temporal characteristics of Wi-Fi beacons in an enterprise Wi-Fi network consisting of over 50 APs deployed in a 300,000 square foot office building. Motivated by our measurement results, we design a novel synchronization protocol called WizSync. WizSync employs digital signal processing (DSP) techniques to detect periodic Wi-Fi beacons and use them to calibrate the frequency of native clocks. WizSync can intelligently predict the clock skew and adaptively schedules nodes to sleep to conserve energy. We implement WizSync in TinyOS 2.1.1 and conduct extensive evaluation on a testbed consisting of 19 TelosB motes. Our results show that WizSync can achieve an average synchronization error of 0.12 milliseconds over a period of 10 days with radio power consumption of 50.9 microwatts/node.Keywords
This publication has 16 references indexed in Scilit:
- Exploiting FM radio data system for adaptive clock calibration in sensor networks2011
- ZiFi2010
- High-resolution, low-power time synchronization an oxymoron no more2010
- Low-power clock synchronization using electromagnetic energy radiating from AC power lines2009
- Optimal clock synchronization in networks2009
- A tale of two synchronizing clocks2009
- RT-Link: A global time-synchronized link protocol for sensor networksAd Hoc Networks, 2008
- Dynamic spectrum access in WLAN channels2006
- Telos: enabling ultra-low power wireless research2005
- The flooding time synchronization protocol2004