This is beginning to feel like being in the software industry again: Work to some set of requirements, then they change the requirments... MS
Chao Sun wrote: > Michael Schippling wrote: >> very cool, nice paper. >> But did you really run the data collector for a month? >> And nothing crashed? >> heh >> >> MS >> >> >> Aurélien Francillon wrote: >> >>> Michael Schippling a écrit : >>> >>> ... >>> >>> >>>> To get a more "random" sequence you need to use some kind >>>> of natural input, like an open ADC or maybe the RSSI with no >>>> active transmission. Even then one can make an argument that >>>> the data is not Truly Random since everything carries some >>>> notion of state... >>>> >>> you may be interested in TinyRNG which is doing similar stuff than >>> above + is designed for security (key generation etc.... ) : >>> >>> http://planete.inrialpes.fr/~francill/Papers/TinyRNG.pdf >>> >>> http://planete.inrialpes.fr/~francill/stuff/TinyRNG-0.1.tar.gz >>> >>> >>> cheers >>> Aurélien >>> >> _______________________________________________ >> Tinyos-help mailing list >> [email protected] >> https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help >> >> > Hi, > > Now the problem is that I need to generate a sequence of random numbers > for all the sensor nodes, that is, after initiation and generation of > the numbers, every node knows exactly what these numbers are. > If I call the random number generator in the StdControl.init(), then > each node may generate different numbers, and thus they can not achieve > a consensus. Thus, do I have to generate these numbers via some outer > programs and explicitly declare them in the node initiation? Is their > any better methods? > > Thanks, > Chao -- Platform: WinXP/Cygwin TinyOS version: 1.x, Boomerang Programmer: MIB510 Device(s): Mica2, MicaZ, Tmote Sensor board: homebrew _______________________________________________ Tinyos-help mailing list [email protected] https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
