All of lore.kernel.org
 help / color / mirror / Atom feed
From: Marek Lindner <mareklindner@neomailbox.ch>
To: b.a.t.m.a.n@lists.open-mesh.org
Cc: laurent <laurent@wifibot.com>
Subject: Re: [B.A.T.M.A.N.] BATMAN-V for robotics
Date: Thu, 25 Feb 2016 00:12:23 +0800	[thread overview]
Message-ID: <1623151.0bLTxGPOFY@voltaire> (raw)
In-Reply-To: <56CAE7C9.1070006@wifibot.com>

[-- Attachment #1: Type: text/plain, Size: 2639 bytes --]

Hi,

> After some failure last year to us BATMAN-ADV on picostation plugged in
> a mobile robot. I have re flashed last week the last open-wrt with the
> last stable batman-adv (2006.0).
> I still have some problem making a single hop fast.  I have 3 nodes A B
> C, I put the robot (nodeB) very close to node C and A is fare from B and
> C.  B is still connecting directly to A. sometime If I wait for a very
> long time
> the route became A - C - B , what I exactly want. I have this behavior
> since 2010 and on each version I test on picostation.
>
> To resume, when the robot along 2 nodes and it takes a very long time
> before he got a new route to increase his range.

don't know what your definition of 'fast' is but did you play with the 
originator interval ? Check our doc:
https://www.open-mesh.org/projects/batman-adv/wiki/Tweaking#originator-interval

The default settings are tailored for wireless community networks with rather 
static setups. 


> I understand but I could be wrong, that the metric calculation in
> BATMAN-IV is not good for mobile application because of some radio that
> keeps the link until the last minute before beginning to increase their
> packet errors ?

That assessment is not entirely correct. If you can live with the overhead of 
faster originator intervals (read: increasing the protocol exchange rate) you 
can run batman inside cars driving around.

This problem is unrelated to the metric.


> BATMAN-V seems to be more promising because of the new metric
> calculation based on link throughput we get from the driver, and also
> because we can override the data throughput value to force a route (we
> are imagining using GPS for that).

I don't see how you would force a route with BATMAN IV or BATMAN V. The latter 
allows specifying a speed over an interface, not per neighbor. This interface 
speed applies to everything routed over that interface.

The reason why these overrides are missing is simple: It defeats the purpose 
of an automatic protocol. If you want static routes there is no need for 
batman.

Furthermore, meshing based on location / proximity with the help of GPS or 
coordinates only works as long as you don't have any obstacles anywhere. 
Meaning: Not in real world setups.


> I really want to test BATMAN-V,  I tried to switch to BATMAN-V but it
> was not on the available routing algorithm list. Do I need to compile a
> devel version on open-wrt ?

Yes, you would need to compile the devel version hosted on git.open-mesh.org 
(https://git.open-mesh.org/openwrt-feed-batman-adv.git) or wait for the next 
stable release. Feedback is welcome!

Cheers,
Marek

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 473 bytes --]

      reply	other threads:[~2016-02-24 16:12 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-02-22 10:49 [B.A.T.M.A.N.] BATMAN-V for robotics laurent
2016-02-24 16:12 ` Marek Lindner [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1623151.0bLTxGPOFY@voltaire \
    --to=mareklindner@neomailbox.ch \
    --cc=b.a.t.m.a.n@lists.open-mesh.org \
    --cc=laurent@wifibot.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.