[Ns-developers] Fwd: Regarding Project

Tommaso Pecorella tpecorella at mac.com
Thu Mar 29 10:45:12 PDT 2012


Hi Mukund,

I checked your project and I have to say it's pretty ambitious !

Now, down to the actual proposal:
Point 5 (Header Compression) is already done (well, in progress), so I'd strip it out.
About the other points, my doubt is if you'll be able to complete them in time. They do span a lot of modules (from modifications to lr-wpan to completely new modules to write), and allocating just one week to each one might be quite optimistic at best.

I'd go for a lower profile plan, with the other points to be developed after the GSOC end, so to make sure you'll maximize the success probability.

Mind that there are a few things that are to be done during the development period, and they'll consume time:
1) documentation
2) unit test development (and use)
3) examples

As you noticed, the topic 1+2 and 3+4 are strictly related, so I'd choose either 1+2 or 3+4 as "official" targets. I'll leave the decision about which of the 2 to choose to you. The only suggestion is: 1+2 means to use the already developed lr-wpan module. 3+4 is a new module.
To me they are all very interesting topics, so I'll not give you constraints.

This said, check Vedran's comments on one of the other's proposals. He was pointing out what's best to include in the proposal so to make it better.

Basically, an overview of the classes you plan to implement, your programming approach, what ns-3 classes you have to "mess up" with and so on. Simply indicating "week 3 : implement bootstrap" is a bit too short.

Last thing: I'd avoid to modify or refine the proposal in week 1, your project should be already well defined by when you'll start.

Cheers,

T.



On 29 Mar 2012, at 02:31, mukund tripathi wrote:

> Hi Tommaso ,
> 
> I am finally done with making the proposal !!
> Kindly review the proposal and suggest changes to be made ?
> 
> I have divided the design into five modules , and that is the way i plan to
> work upon it !
> Kindly also mention if there is a change required to that .
> ( although bootstrap and address autoconfiguration are inter - related ,
> neighbor discovery and optimizations are inter - related and header
> compression is separate .This makes three !! )
> 
> Also , i would like to know which of the features are the least important
> and that can be done away with , since i feel that 10 weeks is a
> comparatively less time slot to implement all the features ( not impossible
> , but I dont want to compromise with quality ) .
> 
> The probable informal plan which i propose is :
> 
> week 1 : updating proposal and further research and modifications
> 
> week 2 : creating design for all the five modules( structuring ,activity
> analysis and data flow )
> 
> week 3 : implement bootstrap
> 
> week 4 : implement auto configuration
> 
> week 5 : implement ND
> 
> week 6 : optimize RS/RA
> 
> week 7 : implement header compression
> 
> week 8 : unit testing
> 
> week 9 : integration testing
> 
> week 10 : Merge code .
> 
> Any changes required in the above time slots would be welcome .
> 
> -Mukund
> <ns-3.docx>

--------------------------------------------------------------

Thinking evolution:
 "To be is to do" - Socrates
 "To do is to be" - Sartre
 "Do Be Do Be Do" - Sinatra
 "Scooby Dooby Do" - Scooby Do
 "Yaba Daba Doo!" - Fred Flintstone 

--------------------------------------------------------------

Tommaso Pecorella - Ph.D.

Assistant professor
Dpt. Elettronica e Telecomunicazioni
Università di Firenze

CNIT - Università di Firenze Unit

via di S. Marta 3
50139, Firenze
ITALY

email: tommaso.pecorella at unifi.it
       tommaso.pecorella at cnit.it

phone : +39-055-4796412
mobile: +39-320-4379803
fax   : +39-055-494569







More information about the Ns-developers mailing list