[Ns-developers] TxVector Bug 917

Ghada Badawy gbadawy at gmail.com
Wed Mar 21 08:07:14 PDT 2012

Hi Ruben,
Right now I am just trying to add the different 802.11n time related
parameters to WifiPhy.
As you know in 11n a HT STA can use either long GI or short GI. These
intervals affect the symbol duration which in turn affect the payload
duration. So to account for both GIs I need to pass to WifiPhy which
interval to use. Since WifiPhy CalculateTxDuration only takes size, payload
and preamble. I was thinking of adding a new vector to the inputs that
carries the new parameters.

When we get to adding MIMO I will also have to include Nss and Nes since
they both affect the number of training symbols included in the PLCP header
which in turn affect the transmission duration.

I think a TXVECTOR (as defined in the standard) would be a good way to go
since it will allow future expansion when new standards come along. Just
add the new parameters to the vector. Since the WifiStationManager will be
responsible of dynamically controlling these parameters (using which GI
depends on capabilities of both sender and receiver) I find that the patch
proposed to solve bug 917 is a good option

What do you think?

On Wed, Mar 21, 2012 at 10:36 AM, Ruben Merz
<ruben at net.t-labs.tu-berlin.de>wrote:

> On 3/21/12 3:34 PM, Ghada Badawy wrote:
>> Hi Ruben,
>> Thanks for the quick reply. I know developing 11n is far from a
>> straightforward task. I am taking it one step at a time :) I will have a
>> look at the ns-devel mailing list.
>> So do you think I can apply the patch provided with the bug and use it
>> for now? or how do you think I should proceed?
> If it applies :-), yes, you can use it. But, in general, are you sure you
> really need the patch. What kind of design do you have for 802.11n ?
> Again, come discuss this on ns-devel
> Ruben
>  Thanks,
>> Ghada
>> On Wed, Mar 21, 2012 at 10:21 AM, Ruben Merz
>> <ruben at net.t-labs.tu-berlin.de <mailto:ruben at net.t-labs.tu-**berlin.de<ruben at net.t-labs.tu-berlin.de>>>
>> wrote:
>>    On 3/21/12 2:25 PM, Ghada Badawy wrote:
>>        Hi,
>>        I am wondering why the TxVector patch bug 917 was not added to the
>>        latest ns3 release. I want to implement 802.11n and was thinking of
>>        adding the TxVector to pass some parameters to the WifiPhy (as
>>        use short
>>        GI or long GI, the number of streams (Nss),...). These
>>        parameters will
>>        affect the CalculateTXDuration function.
>>        Your reply will be highly appreciated.
>>        Regards,
>>        Ghada
>>    Dear Ghada, the TxVector patch is now part of another patchset, for
>>    Click monitor mode support. It is likely to be integrated soon.
>>    Now, for your original 802.11n proposal, you should show up on the
>>    ns-devel mailing list. Implementing 802.11n is far from a
>>    straightforward tasks.
>>    Ruben

More information about the Ns-developers mailing list