Distributed Synchronization

The goal of this research line is to analyze the possible options for network synchronization in dense sensor networks, and to design a power efficient and robust synchronization scheme. The considered scenario is characterized by a large set (up to thousands) of wireless nodes communicating by means of short-distance links. In order to reduce both the overall cost of deployment of the system and the size of the single device, the standard node is not equipped with a quartz crystal, and depends thus on an external reference for clock generation. Such external reference is provided by a small set of nodes, referred to as Reference Clocks, which provide the neighboring nodes with a time reference in the form of a periodic sequence of pulses.
Nodes that are within reception range from a reference clock lock themselves to the sequence of pulses by means of a Phase Lock Loop (PLL) circuit. An example of the scenario is presented in Figure 1.

In our research we compare two different classes of network synchronization algorithms:

The biology-inspired approach, in particular, is an appealing solution for synchronization in the considered application scenario, and is actually the main focus of the research.

References

[1]
R. E. Mirollo and S. H. Strogatz, "Synchronization of pulse-coupled biological oscillators," SIAM J. Appl. Math., Vol. 50, No. 6, pp. 1645-1662, December 1990.
[2]
F. Tong and Y. Akaiwa, "Theoretical analysis of interbase-station synchronization systems," IEEE Trans. Commun., Vol. 46, No. 5, pp. 590-594, May 1998.
[3]
M. Cremaschi, O. Simeone and U. Spagnolini, "Distributed timing synchronization for sensor networks with coupled discrete-time oscillators," International Workshop on Wireless Ad Hoc and Sensor Networks (IWWAN2006), New York, NY, June 28-30 2006.
Last updated: November 2, 2006 by Luca De Nardis