Time Synchronization Services for Wireless Sensor Networks

Dissertation Proposal

Jeremy Elson
Department of Computer Science
University of California, Los Angeles
Los Angeles, CA, 90095
jelson@cs.ucla.edu

Date: April 10, 2001

Abstract:

Time synchronization is a critical piece of infrastructure for any distributed system. Distributed, wireless sensor networks make extensive use of synchronized time in many contexts--for example, in forming TDMA schedules, integrating a time-series of proximity detections into a velocity estimate, in detecting redundant detections of a phenomenon from multiple sensors, and in distributed beamforming systems. The variety of uses leads to highly varied and non-standard requirements in the scope, lifetime, and maximum error of the synchronization achieved, as well as the time and energy required to achieve it. Existing time synchronization methods were not created with wireless sensor networks in mind, and need to be extended or redesigned to meet the new requirements and constraints.

Our proposed work centers around the development of new time synchronization methods, their characterization according to metrics relevant to wireless sensor networks, and a proof-of-concept implementation of an application that makes use of our synchronization primitives. In this proposal, we first describe a number of metrics that we have found useful for the characterization of time synchronization services. Using these metrics, we describe existing services, and compare them to the synchronization requirements of future sensor networks. We also present an implementation of our own low-power synchronization scheme, post-facto synchronization, and describe an experiment that characterizes its performance for creating short-lived and localized but high-precision synchronization using very little energy. Finally, we describe our work plan.