[Olsr-users] Anyone with verified 802.11n HT speeds on Ubiquiti M devices in adhoc/mesh mode?

[Ben West]

Hi Markus,

Indeed, I realize that anything below reported connection speed of 150Mbit/s
(regardless of actual throughput sustained by the IP stack) could indicate
only one chain is active.

My suspicion that perhaps both chains are active, tho not actually
connecting due to non-ideal SNR, comes from what a Nano M5
reported briefly after power up.  A single chain wouldn't report 240Mbit/s,
correct?

Nano M5 #1
(spam-protected):~# iw wlan01 station dump
Station 00:15:xx:xx:xx:f8 (on wlan01)
    inactive time:    70 ms
    rx bytes:    73232
    rx packets:    1119
    tx bytes:    2968
    tx packets:    30
    signal:      -14 dBm
    tx bitrate:    240.0 MBit/s MCS 13 40Mhz short GI

On Tue, Mar 22, 2011 at 4:27 PM, Markus Kittenberger <
(spam-protected)> wrote:

> hmm 45mbit is defintely above 802.11a throughput
> therefore great! to hear this,..
>
> but a single chain can easily do this,..
> (so i wouldn`t be too sure about having both chains working,..)
>
> MArkus
>
> On Tue, Mar 22, 2011 at 12:10 AM, Ben West <(spam-protected)> wrote:
>
>> For an update, I believe I was able to get HT modes working in IBSS (aka
>> adhoc) on two Nanostation M5s.  The Links reported 150Mbit/s connection on
>> both ends after settling, and throughput measured at 45Mbit/s.
>>
>> This required patches to iw and compat-wireless on OpenWRT r25206. More
>> info here:
>> https://forum.openwrt.org/viewtopic.php?pid=131299#p131299
>>
>> I would be interested to hear if anyone has been able to specify HT modes
>> in /etc/config/wireless on OpenWRT, since I was unable.
>>
>> Note that the link speeds did suggest I was getting multiple chains (i.e.
>> MIMO) very briefly.  My test environment (indoor room, 20feet separation)
>> could have been non-ideal.
>>
>> On Mon, Mar 14, 2011 at 2:48 AM, Ben West <(spam-protected)> wrote:
>>
>>> Sorry for not explaining this in the original post.  It is described in
>>> OpenWRT forum post; I had snipped my question to this list to focus on
>>> mesh-related question.
>>>
>>> Gateway: Rocket M5, flashed with pre-compiled
>>> openwrt-ar71xx-ubnt-rocket-m-squashfs-factory.bin for Backfire 10.03-rc4
>>>
>>> Repeaters (and a backup gwy): Nanostation M5, flashed with
>>> custom-compiled pre-compiled openwrt-ar71xx-ubnt-nano-m-squashfs-factory.bin
>>> from Backfire r24045, and with this patch to allow DHCP on 2nd LAN port:
>>> http://freifunk.net/pipermail/wlanware/2010-December/002439.html
>>>
>>> Basically, I'm curious to hear details (e.g. patches, what revision of
>>> OpenWRT/ath9k) where people have confirmed getting HT modes to work.
>>>
>>> Besides that, these nodes are actually working just fine, and their link
>>> speed generally outpaces the uplink I have for them anyway.  It's just that
>>> I would like to support ~40Mbit link speed in the intermediate future,
>>> should a fast enough uplink appear.
>>>
>>> On Mon, Mar 14, 2011 at 2:37 AM, Daniel A. Nagy <
>>> (spam-protected)> wrote:
>>>
>>>> On 03/14/2011 06:10 AM, Ben West wrote:
>>>> > I just set up 3 Ubiquiti M5 devices as mesh nodes running OpenWRT
>>>>
>>>> Since Ubiquiti M5 devices are not as identical as the pre-M series,
>>>> could you be more specific?
>>>>
>>>> These three categories are quite distinct, require different firmware
>>>> and have their own, distinct quirks:
>>>> - PicoStation / Bullet M5
>>>> - NanoStation M5
>>>> - NanoStation M5 Loco
>>>>
>>>> Cheers,
>>>>
>>>> --
>>>> Daniel
>>>>
>>>>
>>>> --
>>>> Olsr-users mailing list
>>>> (spam-protected)
>>>> http://lists.olsr.org/mailman/listinfo/olsr-users
>>>>
>>>
>>>
>>
>>
>> --
>> Ben West
>> (spam-protected)
>>
>> --
>> Olsr-users mailing list
>> (spam-protected)
>> http://lists.olsr.org/mailman/listinfo/olsr-users
>>
>
>


-- 
Ben West
(spam-protected)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.olsr.org/pipermail/olsr-users/attachments/20110322/5e829e8b/attachment.html>