-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 11/04/2013 07:23 PM, james dee wrote:
tried SERVAL mesh (which is currently available).
The need to
root a phone in order to enable ad hoc networking limits the
utility of these
Not all builds of Android can put their wireless chipsets into ad-hoc
mode. On some of the drivers we'd reversed, the code which actually
tickles those bits isn't even there - it seems to have been commented
out prior to compilation.
I would also strongly recommend that everyone read the following
ticket in the Android bug tracker:
https://code.google.com/p/android/issues/detail?id=82
Long story short, our contacts at Google stopped talking to us and
they locked comments on the ticket ("Only users with Commit permission
may comment.")
opportunities to speculate. Old phones can be cheap,
especially
if they can’t be used as phones… think of that blacklisted CDMA
phone that you couldn’t manage to change the IMEI on… what’s its
market value? If
They are certainly useful as PDA's, assuming that wireless is
accessible in that location.
you were to tie it to a $20 solar panel and throw it
up on
someone’s roof you’d have a disposable node that any anarchist
would love. If
Something to keep in mind is that the older a phone gets, the more
fragile the silicon gets. Running a mobile in wi-fi mode 24x7 is
going to put a hurt on it. That's how I killed mine last summer. You
could probably get by with a slightly cheaper solar panel (like the
cellphone charger kits sold at HOPE last summer in the hacker village).
you put one at the focal point of a satellite dish how
far down
the block could you get? Better yet, could you use the phone
itself as the
A good question. I know of an antenna which would be a good place to
start. However, if you are going to go the dish route, it will by
definition narrow the beam to trade off for distance. It would work
for a point-to-point link, but it will make communication with clients
problematic because they will have to be positioned within the beam to
pick up the transmission reliably.
driven element of a homebrew yagi? A city-wide mesh
would
require several phones per block and probably wouldn’t be very
practical but
An optimal configuration is one per building, we found. That seems to
give decent enough coverage in environments where there are obstacles
that attenuate the signal and cause blank spots.
If you want to link nodes that are far apart and have second
interfaces for clients, dishes or yagis work. If you want clients to
connect to your node, you want either an omni or something to reflect
signal away from the big flat surface the node is near toward the
clients. Pick the antenna for the use case, and accept trade-offs to
get the job done.
imagine trying to cover a smaller area… Hmmm
For small areas, you do not want lots of nodes. That is what killed
the OLPC as a viable mesh network platform - every OLPC was both a
client and a mesh node, and the network overhead crushed the wireless
network. You want one, maybe two decent mesh nodes at most with clean
signals. Also, pick a channel that is lightly trafficked, because
neighboring wireless access points will interfere with a mesh that
uses the same wi-fi frequency everybody else and their backup is using.
- --
The Doctor [412/724/301/703] [ZS]
Developer, Project Byzantium:
http://project-byzantium.org/
PGP: 0x807B17C1 / 7960 1CDC 85C9 0B63 8D9F DD89 3BD8 FF2B 807B 17C1
WWW:
https://drwho.virtadpt.net/
"Where there is love, there is life, and where there is life there is
hope." --The Cruxshadows
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.20 (GNU/Linux)
Comment: Using GnuPG with Thunderbird -
http://www.enigmail.net/
iEYEARECAAYFAlJ5Tw0ACgkQO9j/K4B7F8FH6wCfQr9Jo0UPywpjkn0UspXGkfkp
kvgAoIYLRbHwu+BCnMEeWOtbFwOS/Z1I
=dOKf
-----END PGP SIGNATURE-----