Hi, I just want to add more:
The linkLayerModel use java rand.nextGaussian() to find gaussian value. This function generate large value as compared to Matlab function that add value in pathloss ( +6 db to -6db) or more (I don't remember exactly value). I have used alternatively (Math.exp(- Math.log(rx)*Math.log(rx) / (2 *inVar.sigma * inVar.sigma)) / Math.sqrt(2 * Math.PI) * inVar.sigma * rx); where rx = (rand.nextInt(inVar.numNodes)) /(20*inVar.sigma); Thanks, Poonam On Tue, Feb 24, 2009 at 3:43 PM, Alban Hessler <[email protected] > wrote: > To answer my own question: > > The problem is the noise generation of tossim, where I used the meyer > files to generate it. The noise level in those file are rather large. > I think they must be fine to simulate an office environment with some > AP interference. But, I don't want to have harsch conditions all the > time. (a -40db noise, in tossim, pure jamming). In average, the noise > -85 db, where in other simulations/papers, noise level is around -100 > db. So that was why I had a connectivity of a few meters using > LinkLayerModel.java. > > The Wiki is quite misleading, in my opinion. Stating that there are > spikes at -86 dbm, when they are in fact more dirac-like, topping at > -40. I just created a user account will change that.. Maybe the floor > is at -98, but there are a lot of planes at -80 dbm.. And 16 dbm is > quite a lot. > > Cheers, > > Alban > > > > > > > 2009/2/17 Alban Hessler <[email protected]>: > > Hi, > > > > The LinkLayerModel tool allows to create a grid from a topology file, > > stating that the unit is the meter. But from that statement, it seems > > that for default settings as proposed in the tutorial, the range of > > transmission is about from 5 to 10 meters only. In my outdoors > > settings with Micaz, I could have good connectivity with a range of 40 > > m. So why is there such a difference? Is the unit finally not meters? > > > > I tried to play around with the fading exponent to have something that > > match my experience, but not sure if it's a good approach. > > > > In the very end, what matters is the connectivity, but still I would > > like to have an answer on that. > > > > Thank you, > > > > Alban > > > _______________________________________________ > 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
