Re: [Madwifi-devel] Translate Madwifi Project Page to Spanish (Isma De Andres)

[sandra <ma...@eh...>]

Hello Isma and all,

I am also interested in that project. I'm from Spain and my native 
language is Spanish, but I can understand perfectly written english and 
I'm working with MadWiFi now for my PHD related with wireles 
technologies for developing countries.

If we are interested, please, let me know how I can participate.

    -- Sandra.

mad...@li... wrote:

>Send Madwifi-devel mailing list submissions to
>	mad...@li...
>
>To subscribe or unsubscribe via the World Wide Web, visit
>	https://lists.sourceforge.net/lists/listinfo/madwifi-devel
>or, via email, send a message with subject or body 'help' to
>	mad...@li...
>
>You can reach the person managing the list at
>	mad...@li...
>
>When replying, please edit your Subject line so it is more specific
>than "Re: Contents of Madwifi-devel digest..."
>
>
>Today's Topics:
>
>   1. Re: How to test the maximum speed of the packet transmitting
>      in the driver? (Derek Smithies)
>   2. Re: disabling the carrier sensing (Michael Renzmann)
>   3. Re: [RFC] Commit policy (Michael Renzmann)
>   4. Re: [RFC] Commit policy (Michael Renzmann)
>   5. Translate Madwifi Project Page to Spanish (Isma De Andres)
>   6. Re: [RFC] Removing no longer used branches	in	repository
>      (Michael Renzmann)
>   7. #1109 -- ad hoc nets invisible to Centrino on	Windows
>      (Dennis Edwards)
>   8. Re: How to test the maximum speed of the packet	transmitting
>      in the driver? (Lrj)
>
>
>----------------------------------------------------------------------
>
>Message: 1
>Date: Tue, 15 Apr 2008 09:23:13 +1200 (NZST)
>From: Derek Smithies <de...@in...>
>Subject: Re: [Madwifi-devel] How to test the maximum speed of the
>	packet transmitting in the driver?
>To: Benoit PAPILLAULT <ben...@fr...>
>Cc: Lrj <ren...@gm...>,	Madwifi-devel Maillist
>	<mad...@li...>
>Message-ID:
>	<Pin...@ka...>
>Content-Type: text/plain; charset="utf-8"
>
>On Mon, 14 Apr 2008, Benoit PAPILLAULT wrote:
>Lrj a ?crit :
>| Hi,
>|       I'm trying to test the maximum speed of sending packets with the
>| AR5212. So I tried to send packets in the driver. And my solution is to
>| initialize the pakcet as a struct sk_buff, call the
>| ieee80211_monitor_encap(), and use the function dev_queue_xmit() to send
>| it out, just like the ordinary sending process under the monitor mode in
>| the driver.
>|       And now two problems meet:
>|
>|    1. When sending a packet with just 1 byte size, machine A sent one,
>|       and B got one; However, when sending a 1500 bytes packet, A sent
>|       one, B got 10(sometimes got 11) packets! While with the counter I
>|       added in the driver, A just sent one packet out. How could this
>|       happen?
>|    2. Now, under the rate of 54Mbps, A sent packets continuously, B
>|       received and counted. According to the result, the maximum
>|       speed(num * size / time *8) is 40Mbps. But I don't think that this
>|       is the finally ceiling the NIC could get. So does anyone have
>|       better idea to fill the channel's bandwidth and to get the real
>|       maximum sending speed?
>|
>
>  
>
>>Now, if B got 11 packets, it's probably the automatic retry feature.
>>Check the retry bit in your received packets. This has probably some
>>consequence in your later speed computation. You'd better check what
>>happens on the air with a third card in monitor mode.
>>    
>>
>
>Benoit is correct. The packet is put on the air and sent. At the time of 
>sending, the packet is put out at speed of 54Mbps. Then the radio goes 
>silent, and the sending node waits for an 802.11 ack to come back from the 
>far end. This 802.11 ack is a radio level ack, and not to be confused with 
>the tcp ack.
>
>To get more of the channel's bandwidth used? Well send the packets out a 
>broadcast queue, which is not 802.11 acked. However, in this case, how do 
>you know if the packets actually get through?
>
>However, the radio hardware enforces a delay between sequential packets. 
>Thus, it is not going to be possible to fill the air with your packets, 
>and get 54mbps data through.
>
>Further, from looking at the results of a card in monitor mode, you will 
>see that a radio packet is huge. There are literally hundreds of bytes of 
>header. These will drop the measured througput of your application data.
>
>Derek.
>
>  
>

-- 
Estamos tan gastados como la misma tierra, pero por intentar no pisarla.
-----------

Sandra Salmerón Ntutumu    <ma...@eh...>
+34 91 488 8703 / Móvil: + 34 653 574 298
Fundación EHAS: Enlace Hispanoamericano de Salud - www.ehas.org
Telemedicina rural para zonas aisladas de países en desarrollo
-----------
Sandra Salmerón Ntutumu    <ma...@eh...>
+34 91 488 8703 / Mobile: + 34 653 574 298
EHAS Foundation: Spanish-American Health Link - www.ehas.org
Telemedicine for rural isolated zones in developing countries