[Tinyos-help] Outdoor Range with Tmote sky

Michael Schippling schip at santafe.edu
Wed Feb 21 18:51:51 PST 2007


That ref you give looks like a pretty good analysis. Although
I can't attest to the accuracy of any of it, it's got lots of
cool equations. I assume you are referring to the graph on pg 17,
where the no-problem-with-rain statement is made...That particular
page is a really fabulous example of what I call the PowerPoint
Information Removal Feature, which is installed by default.
Aside from not labeling the axes, the traces are annotated
in urgo-finnish or something. Perhaps there were speaker's notes
attached (which, as part of another PIRF default, you just can't
print without reading the help file) but as it stands I don't
see any proof of anything, other than, of course, the PIRF default
usage, on that slide...

At least your observation agrees with my RoT. And being a RoT I
don't have to provide concentration/absorption curves...although
I suppose I could do it with PowerPlonk...heh

MS

Aleksandr N. Sadkov wrote:
> In theory Rain doesn't affect on Path Loss at 2.4GHz, for example
> (http://www.radionet.com/_FileRoot/318040.pdf)
> 
> But I have measured great drop in RSSI during the rain, I guess it's due to
> antenna effects.
> 
> Regards, Sadkov Aleksandr.
> 
>> A few rules of thumb...
>>     The lower the frequency, the more likely the signal will "curve"
>>       or diffract around obstacles.
>>     The bigger and more conductive the obstruction, the more it will
>> absorb.
>>     Microwaves at 2.4 GHz are used to heat food because water absorbs
>>       at that frequency, thus 802.11b,g and CC2420's don't work so well
>>       when it's raining below sea level...
>>
>> MS
>>
>> David Gay wrote:
>>> On 2/20/07, Philip Levis <pal at cs.stanford.edu> wrote:
>>>> On Feb 20, 2007, at 10:42 AM, Jacob Sorber wrote:
>>>>
>>>>> You are asking for a simple answer where none exists.  Wireless
>>>>> channels are tricky.  While line-of-sight is straightforward, "not
>>>>> in line of sight" could mean a lot of things.  What is obstructing
>>>>> the signal?  A building?  A person?  A hill of dirt and rock?  Is
>>>>> it raining?  What is the humidity?  I have yet to find a radio that
>>>>> is immune to obstructions.  If you find one, I would love to see
>>>>> it.  My experience has been that the 2.5Ghz radios are usually more
>>>>> robust to obstructions than the CC1000 radios on the Mica2/Mica2Dot
>>>>> motes, but not always.  When one mote is a meter under water the
>>>>> CC1000 seems to do much better than the CC2420, though neither work
>>>>> very well in that situation.  Also a rain storm can reduce the
>>>>> range some.  The best thing to do is to program two motes and take
>>>>> them outside and see.  It will only take a few minutes.  If you
>>>>> need a more technical explanation for why node A can't hear node B,
>>>>> then you need to do some reading in the wireless comm literature.
>>>>> This might be a good place to start ( www.eecs.berkeley.edu/~dtse/
>>>>> cu_day1.ppt).
>>>>>
>>>> My understanding is that the waves which are really resistant to
>>>> obstructions are extremely low frequency (ELF) ones: think < 100Hz.
>>>> That's what submarines use to communicate when submerged, for example
>>>> [1].
>>> Lower frequencies propagate better through water (which is why a
>>> CC1000 at 400 or 900MHz is better than a CC2420 at 2.4GHz). And
>>> propagation through fresh water is significantly better than through
>>> salt water (if I remember the numbers correctly, salt water ==
>>> essentially no range at the frequencies under discussion, while fresh
>>> water might get a few meters).
>>>
>>> Lots more detail on all this is available if you look up electrical
>>> engineering source materials rather than sensor network ones... (i.e.,
>>> tinyos-help is probably not the best place to find out about this, try
>>> finding a friendly professor with RF knowledge).
>>>
>>> David Gay
>>> _______________________________________________
>>> Tinyos-help mailing list
>>> Tinyos-help at Millennium.Berkeley.EDU
>>> https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-
>> help
>> _______________________________________________
> 


More information about the Tinyos-help mailing list