On Sun, Jan 10, 2016 at 1:32 AM, Marc Juul <juul(a)labitat.dk> wrote:
On Sat, Jan 9, 2016 at 2:26 PM, Valent Turkovic <valent(a)otvorenamreza.org>
wrote:
Hi all,
I'm one of people who started
www.meshpoint.me - idea is to build open
source wifi router that can be easily deployed in any type of crisis
situations.
I have been setting up wifi networks in crisis situations and I know
that there are no devices that are usable in crisis situations.
Belive it or not but first responders usually aren't network admins :)
So that is why we started developing MeshPoint, wifi device that
anybody can use in crisis situation.
First iteration will use off the shelf components and later we plan
develop our own wifi router board using atheros hardware.
Our current design (as you can see on the web site) uses three TP-LINK
WR842ND (AR9341) boards in 3D printed case.
Interesting. No 5 ghz though? If you're worried about L2 retransmissions
that seems like the first thing to solve.
Any ideas how to test and how to solve if there L2 retransmissions are
an issue? Would shielding help?
The TP-Link dual-band routers are
cheap as well and you could probably easily swap out your current boards for
dual-band models.
We will probably have devices with dual band radios also, but right
now I choose to use three 2.4Ghz radios because this is way we get
capacity for more people and these wifi routers are much cheaper than
dual band.
Most client devices are still single band, only premium devices have
dual band chips, and my rough estimate that that is around 10% od
devices... so currently it makes little sense to for dual band, better
to have two or three radios and have more capacity.
In refugee situation as one we were involved in there are lots of
people, we worked in refugee camps which had capacity of 2000 people
and on most days they were full. So having dual band radio who would
have nice but we were on a almost no budget and providing and for same
amount of money you can get more 2.4 routers than dual band routers...
and it is a question how many people would benefit with 5.8 radios...
And TP-LINK WDR3600 and WDR4300 boards are quite bit larger that WR841
boards and I'll have to see if that even makes it possible to 3D print
whole case...
This first version we are developing will have only 2.4Ghz radios and
next one will have dual band radios as well...
Our concern is
that this could generate too much interference and
cause too much L2 retransmissions because of co-channel interference.
We are using latest OpenWrt 15.04 Chaos Chalmer and ath9k driver and
would like to know if it is possible to extract metrics that shows
number or L2 retransmissions.
Do mac80211 or ath9k tracks L2 retransmission statistics?
I'm no expert on this but try looking at:
/sys/kernel/debug/ieee80211/phy0/ath9k/xmit
Some info can be gleaned from:
http://lxr.free-electrons.com/source/drivers/net/wireless/ath/ath9k/debug.h
Awesome, thanks!
Is there
anyway to get AR9341 datasheet?
Don't have that one, is AR9331 close enough?
https://www.openhacks.com/uploadsproductos/ar9331_datasheet.pdf
AR9331 datahseet is easily found online but AR9341 or QCA9533 or any
other datasheet is not, strange...