[ns] RE: ALSO! Does Release 2.27 support 802.11 Infrastructure mode?
Mike.Moreton at Synad.com
Thu Jan 22 00:10:32 PST 2004
In this case the frames are from the same BSS.
In 802.11 (ignoring the wilder TGe proposals for the moment) a STA can't
send a group addressed frame directly - it has to unicast it to the AP,
which forwards it out as a group addressed frame.
This is the bit that wasn't happening in the code I ported. The frames
were actually being broadcast by the originating STA rather than going
via the AP.
Both bugs are required for broadcast to work at all - they effectively
cancel each other out. So I would expect that 2.27 either has both
bugs, or none of them, but I very much doubt it has only one of them!
Wireless LAN Business Unit
From: aditi kapoor [mailto:akapoor at usc.edu]
Sent: 21 January 2004 22:26
To: Padmaparna Haldar
Cc: Mike Moreton; Lalit Kotecha; ns-users at ISI.EDU
Subject: ALSO! Does Release 2.27 support 802.11 Infrastructure mode?
However, Since i introduced BSS's on different channel, I dont think the
first bug should still exist.
Since each BSS is on a different channel, no packets from an AP outside
this BSS (an hence on a diff channel) will be accepted by any STA on
The 2nd bug would still exist.
----- Original Message -----
From: Padmaparna Haldar <haldar at ISI.EDU>
Date: Wednesday, January 21, 2004 11:24 am
Subject: RE: [ns] Does Release 2.27 support 802.11 Infrastructure mode?
> On Wed, 21 Jan 2004, Mike Moreton wrote:
> > Just to add something - all I did was port part of the Atheros
> > Since when I discovered two bugs, that had the effect of cancelling
> > other out. Firstly, STAs should ignore broadcasts that don't come
> > their AP, but they don't. Secondly, STAs should send broadcasts via
> > AP, but actually they send them directly.
> > So the simulation appears to work, but broadcast/multicast
> > about twice what it should be.
> > I don't know if this "feature" has been fixed in 2.27.
> As her report (see below) states, I don't think these bugs were fixed
> aditi. I've cc'ed her incase she wants to add her comments to this.
> Mike, thanks for clarifying the bugs.
> > Mike Moreton
> > Wireless LAN Business Unit
> > STMicroelectronics
> > -----Original Message-----
> > From: Padmaparna Haldar [haldar at ISI.EDU]
> > Sent: 21 January 2004 18:36
> > To: Lalit Kotecha
> > Cc: ns-users at ISI.EDU
> > Subject: Re: [ns] Does Release 2.27 support 802.11 Infrastructure
> > Hi Lalit,
> > The 2.27 release supports the infrastructure mode of 802.11. It
> > implemented partly by Mike morton and Aditi Kapoor (as part of her
> > work
> > here). Please read her DR report (see below) for details of the
> > implementation. Note that scan/join functionality was not
> > I shall add a section for this addition to 802.11 in the ns manual.
> > Hope this helps,
> > --Padma
> > -------------------------------------------------------------
> > FALL 2003 Directed Research Report under Prof. Hiedemann
> > ABSTRACT
> > This directed research was to bring certain concepts of 802.11 and
> > implement them in the network simulator NS-2. I implemented the
> > that introduces a Basic Service Set (BSS), which is a set of
> > that
> > form
> > an adhoc network, or are controlled by a centralized entity called
> > Access
> > Point (AP), into the network simulator NS-2. The code that
> > in
> > NS-2 , was implemented by Mike Moreton of Synad Technologies, and I
> > ported it in NS-2, understood, tested and validated that for too for
> > this
> > DR.
> > In 802.11 each BSS, is on a different channel. The last portion of
> > DR
> > aimed at introducing each BSS on a different channel in NS-2 and
> > sure
> > that nodes are allowed to change their BSS's and hence their AP's
> > meaning
> > they even switch channels, and yet follow the pattern of 802.11.
> > tried
> > to make sure that since two BSS's are on different channels, and can
> > hence
> > not communicate through the wireless channel amongst each other,
> > still
> > can communicate through the wired link to do so. I tested that
> > once
> > reached the BSS through another BSS via the wired link, still follow
> > 802.11 pattern inside a particular BSS. The concept of data transfer
> > between
> > two BSS's through the wired channel in NS-2 hence, is in place
> > irrespective
> > of whether the nodes are moving or stationary.
> > 1. INTRODUCTION
> > The directed research was started with the aim of introducing the
> > join
> > functionality and hence bring the concept of a basic service set. A
> > is
> > a
> > collection of stations sharing a channel. Though the aims were big
> > initially,
> > here is the list that we finally decided to do, which was different
> > the
> > initial proposal.
> > * Introducing the concept of a BSS with each BSS having an AP.
> > this
> > code and adding test suites for the same. Testing the code that
> > introduces
> > End of Transmission Frames and why they are introduced.
> > and porting Mike's code to do this.
> > * Introducing more than one BSS, testing that it works as expected
> > making sure that packets from one BSS don't reach another BSS.
> > Adding test suites for the same.
> > * Bringing the concept of each BSS on a different channel and each
> > communicating only on one channel.
> > Packets from diff channels are dropped.
> > * Communication made possible between wireless suites through wired
> > network. Test Suites to do the same.
> > On Wed, 21 Jan 2004, Lalit Kotecha wrote:
> > I am interested in simulating characteristics of 802.11 MAC (DCF
> > infrastructure). Does recent release
> > 2.27 support infrastructure mode? What are configurations
> > for
> > setting it to work in
> > infrastructure mode?
> > >
> > >
> Be true to your work, your word, and your friend.
> Padmaparna Haldar
More information about the Ns-users