Hi all,

I'm quite effectively using the LogStorage to store sample data in a
deployment. As of today, whenever I have to extract the data from the
storage, I simply seek to the beginning and then read each entry, printing
it out in a CSV-like format and collecting the printouts from a terminal.
The problem here is that, depending on the hardware device I'm using, I
need to add a certain delay between reads in TinyOS (between 25 and 50 ms),
otherwise the output from the serial line becomes garbled (basically,
printfs overlap each other). This means that it takes about 20 to 40
minutes to download a month of data (sampling once per minute).

Because of this, I think this is one of the less efficient ways I can do
this: do you know which would be a more efficient (i.e., faster) way? One
idea could be to use the SerialAM interface and a corresponding program (in
Java, for example) on the computer side to interpret each message, but I
wonder if the same problem of overlapping printfs would appear also in this
case, thus rendering the messages corrupted.

Any idea?

-- 
Sivieri Alessandro
[email protected]
http://sivieri.wordpress.com/
_______________________________________________
Tinyos-help mailing list
[email protected]
https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help

Reply via email to