Hi all,
This question is more a request for advice and/or opinion. Some research
articles which deal with synchronization (i.e., aimed at designing some kind of
TDMA scheme) in wireless sensor networks, state that nodes can maintain
synchronization by exchanging clock skew/drift information with neighboring
nodes in coverage area. In other words, maintain any sort of local
synchronization. In this sense, in the specialized literature and/or
datasheets, it can be found information about the clock skew associated to many
devices. For example, in [1], authors indicate that, on average, TelosB
introduce 30 us per second of drift; also MicaZ have 40 us per second on
average of clock drift [2].
Taking into account the aforementioned, had no global synchronization (a unique
node, e.g., the coordinator of the network, conducts a periodic synchronization
proccess with all nodes of the network), would be any simple way to make this
local synchronization in TinyOS? I mean, is it wise to consider in your
synchronization protocol either these values found in the literature (e.g., 30
us in case you worked with TelosB); or exchange timestamps between two nodes
and calculate the difference of time; etc.?
Thanks in advance,
David
[1] W. Pak, K.-T. Cho, J. Lee, and S. Bahk, “W-MAC: Supporting Ultra Low Duty
Cycle in Wireless Sensor Networks,” in Proceedings of the 2008 IEEE Global
Telecommunications Conference (IEEE GLOBECOM 2008), 2008, pp. 1–5.[2] Y. Wu,
J.A. Stankovic, T. He, S. Lin, "Realistic and Efficient Multi-Channel
Communications in Wireless Sensor Networks", in: 27th IEEE International
Conference on Computer Communications, IEEE, Phoenix, Arizona, 2008, pp.
1193–1201.
_______________________________________________
Tinyos-help mailing list
Tinyos-help@millennium.berkeley.edu
https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
