On the trade-off of power consumption and time synchronization quality in Wireless Sensor Networks

Abstract

In this work we find a lower bound on the energy required for synchronizing a Wireless Sensor Network (WSN) based on clock estimation techniques. The energy required for synchronizing a WSN is specified by both the transmit power and the required number of messages in order to achieve a desired estimation error level. This paper brings two main contributions: 1. A general relationship between the energy spent in the transmission and the clock offset estimation error by means of the number of transmitted messages, for both large-scale and small-scale wireless channel fading. 2. An optimal lower bound on the required energy and the number of messages by minimizing their product.

A Primer on Energy-Efficient Synchronization of WSN Nodes over Correlated Rayleigh Fading Channels

“…For small values of S and slow moving nodes, m increases due to the effect of high channel correlation, which is in line with the results presented in [5]. As the velocity grows, the channel correlation decreases and the problem tends to the uncorrelated channel case studied in [4]. Fig.…”

“…For example, the Time Flooding Synchronization Protocol (TFSP) [3] message exchange mechanism to produces a linear regression estimation of the transmitter's clock offset. Clock estimation accuracy depends on the number of received messages and the degree of accuracy of the retrieval of time stamps [4]. In [5] the authors minimize the outage probability as a function of the transmit power, which does not imply energy optimization; moreover, the authors did not consider the relative velocity of the sensing nodes in their model.…”

On Energy-Efficient Time Synchronization for Wireless Sensors under Large-Scale and Small-Scale Fading Effects

“…Moreover, the trade-off "transmit power-clock synchronization quality" is a critical issue for wireless embedded systems that requires to find an optimal solution. This paper extends the work we presented in [13] and provides a concrete answer to the following question: "What is the minimum energy required to synchronize a WSN within a target clock accuracy? "…”

A model for the classification and survey of clock synchronization protocols in WSNs

“…Since the energy consumption depends on how long and how frequently each sensor node executes the clock synchronization process, the energy efficiency of synchronous clock synchronization protocols can be improved further by reducing the duration of synchronization phase and increasing the synchronization interval as far as possible. Besides, energy efficiency is also improved by minimizing the number of messages sent and received at each sensor node during each synchronization phase [30,31].…”