Re: Questions about NAB

Hello Luc, 
Thanks for your interest in NAB. 
As a first high-level answer, NAB is ongoing and in its early stages, ie 
the current release is '0.6' and we don't expect a real 1.0 release for 
about 4-6 more months.
So some of the features you ask about are not in the current release, but 
are either in the current unreleased cvs, or will be implemented soon.

Also i will likely be working on a metro-level simulations
in the next few months, in which case i will be adding quite a few 
features for citywide mobility models, use of fixed (wired) accesspoints, 
etc.

> Hugely interested into NAB, I have a few questions about it:
> 1. Is the mobility graph deduced from a model of the campus (which 
> consists of
> the buildings, the paths, roads, tramrails...) or has it been drawn by 
> hand and
> subsequently mapped on a picture?

the mob. model is derived from a set of geographic points, (buildings, 
paths, etc) between which nodes are moving. more info in 
http://nab.epfl.ch/gen/Mobs.epfl_waypoint.html

this is an area where substantial features are likely to be added during 
the fall.

> 2. Have you developed a radio wave propagation model (attenuation of the 
> signal
> because of the distance, because of obstacles like walls or people...)

currently there is no prop model. I do intend to add one in the next 
release, this should be a fairly small amount of work.

> 3. In the implementation of the 3 different MAC layers (null, CSMA and 
> IEEE802.11)
> your simulator supports, which features of the MAC do you consider? By 
> the way, do
> you explicitely model the different network stacks or do you only model the
> _impact_ of the MAC on the higher layers?


no, for the 802.11 MAC (which btw is not in the current release but mostly 
implemented in CVS) the MAC is fully modeled, ie RTS/CTS/ACK packets are 
actually sent (and cause interference with other nodes, may be subject 
to collisions, etc).

> 4. Does NAB deal with the power management?

It depends on what type of functionality you are referring to. IT is quite 
easy to do things like turn the radio on/off (ie if you are modeling MAC 
layers with sleep cycles), or count various quantities related to power 
consumption (ie # bytes/pkts transmitted).

It would require some work in the core to support multiple and changeable 
power levels (currently all nodes xmit at same power). 


> Congratulations for you great work!
> Read you soon. (in french or english)


Hope this helps, 

Henri


> 
> Regards,
> Luc.
> --
> Luc Hogie
> PhD student
> Luxembourg University
> 
> 
> 
> **********************************************************************
> This email and any files transmitted with it are confidential and
> intended solely for the use of the individual or entity to whom they
> are addressed. If you have received this email in error please notify
> the system manager.
> 
> This footnote also confirms that this email message has been checked for the presence of computer viruses.
> 
> **********************************************************************
> 
>