On Fri, Apr 8, 2011 at 2:03 AM, Sergio Valcarcel <[email protected]>wrote:

> Hi,
>
> Interesting links...
>
> For such a kind of signal processing (involving optimization and so
> on). I would embed many C functions instead of coding everything in
> nesC.
>

Doesnt' make any difference.  nesC is a superset of C so the bulk really
is C.  Any linkages that nesC creates also tend to get optimized out by
the toolchain.


>
> In my case I feel more comfortable. What is your opinion, would you
> rather code everything in tasks and modules?
>

Yes but because it is the predominant pardigm for how to hook things
together.  It very clearly delineates how modules communicate with
each other.


>
> Cheers!
> Sergio
>
> On Fri, Apr 8, 2011 at 10:44 AM, Fernando Pianegiani
> <[email protected]> wrote:
> > Eric, the papers that I posted were about the efficient implementation of
> > the SVM, in particular for a couple of low power micro-controllers (the
> > Atmel ATmega128L and the Microchip PIC16F877). The first of the tow was
> > programmed by NesC for the Mica2, Mica2Dot, MicaZ, etc.
> >
> > We used nu-SVM, which is for implementations on platforms with
> > reduced capabilities [3], [4].
> > Unfortunately, IEEE papers have to be purchased.
> >
> > Free definitions of SVM with some references can be found at
> > http://en.wikipedia.org/wiki/Support_vector_machine
> >
> > or at many other websites.
> >
> > Some important references are:
> >
> > [1] V. Vapnik, Statistical Learning Theory. Hoboken, NJ: Wiley, 1998.
> >
> > [2] B. Scholkopf and A. Smola, Learning With Kernels. Cambridge, MA:
> >
> > MIT Press, 2002.
> >
> > [3] B. Schölkopf, A. Smola, R. Williamson, and P. L. Bartlett. New
> support
> > vector algorithms. Neural Computation, 12, 2000, 1207-1245.
> >
> > [4]
> >
> http://www.google.it/url?sa=t&source=web&cd=1&ved=0CBgQFjAA&url=http%3A%2F%2Fciteseerx.ist.psu.edu%2Fviewdoc%2Fdownload%3Fdoi%3D10.1.1.6.9729%26rep%3Drep1%26type%3Dpdf&rct=j&q=tutorial%20nu-svm&ei=RsaeTeq8NMeLswaS0oz9AQ&usg=AFQjCNEJdh4s5wUKz1C7bQhL9h2LnrP5nA&sig2=0RV7GfzPjbw5o8-UfQXw0A&cad=rja
> >
> > [5] LIBSVM, SVM Tool. [Online]. Available: http://www.csie.ntu.edu.tw/
> >
> > cjlin/libsvm/
> >
> > Fernando
> >
> > On Fri, Apr 8, 2011 at 9:36 AM, Eric Decker <[email protected]> wrote:
> >>
> >> Neither of those abstracts define what support vector machines are.
> They
> >> reference the term.
> >> And one would need to by the papers to see if it is defined inside the
> >> paper.
> >>
> >> doesn't help.  no problem.
> >>
> >> From wikipedia: http://en.wikipedia.org/wiki/Support_vector_machine.
> Is
> >> that what the original
> >> poster is referring to?
> >>
> >> If so, the first paper talks about an implementation on the mica2.  Not
> >> sure how to find
> >> the source code but the paper maybe helpful.
> >>
> >> On Thu, Apr 7, 2011 at 11:49 PM, Fernando Pianegiani
> >> <[email protected]> wrote:
> >>>
> >>> Hi,
> >>>
> >>> perhaps these two works can be of help:
> >>>
> >>>
> >>>
> http://ieeexplore.ieee.org/search/freesrchabstract.jsp?tp=&arnumber=4404133&queryText%3Dpianegiani%26openedRefinements%3D*%26searchField%3DSearch+All
> >>>
> >>>
> >>>
> http://ieeexplore.ieee.org/search/freesrchabstract.jsp?tp=&arnumber=4061072&queryText%3Dpianegiani%26openedRefinements%3D*%26searchField%3DSearch+All
> >>>
> >>> Fernando
> >>>
> >>> On Fri, Apr 8, 2011 at 2:16 AM, Eric Decker <[email protected]> wrote:
> >>>>
> >>>> Hi Sid,
> >>>>
> >>>> what pray tell is a support vector machine?
> >>>>
> >>>> On Thu, Apr 7, 2011 at 3:09 PM, sid sid <[email protected]> wrote:
> >>>>>
> >>>>>  Hi all.
> >>>>>
> >>>>>
> >>>>> i want to develop a support vector machine (SVM) on nesc
> >>>>>
> >>>>> Does anybody have an experience about this topic?
> >>>>>
> >>>>> Thanks in advance
> >>>>>
> >>>>> sid
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>> _______________________________________________
> >>>>> Tinyos-help mailing list
> >>>>> [email protected]
> >>>>>
> >>>>>
> https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
> >>>>
> >>>>
> >>>>
> >>>> --
> >>>> Eric B. Decker
> >>>> Senior (over 50 :-) Researcher
> >>>>
> >>>>
> >>>>
> >>>> _______________________________________________
> >>>> Tinyos-help mailing list
> >>>> [email protected]
> >>>>
> https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
> >>>
> >>>
> >>
> >>
> >>
> >> --
> >> Eric B. Decker
> >> Senior (over 50 :-) Researcher
> >>
> >>
> >
> > _______________________________________________
> > Tinyos-help mailing list
> > [email protected]
> > https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
> >
>
> _______________________________________________
> Tinyos-help mailing list
> [email protected]
> https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
>



-- 
Eric B. Decker
Senior (over 50 :-) Researcher
_______________________________________________
Tinyos-help mailing list
[email protected]
https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help

Reply via email to