Chinese Remainder Theorem Implementation

[Atul Chavan]

Does anyone have ever implemented Chinese remainder theorem for packet splitting in NS3 or NS2 ?

looking for exact code to learn its implementation in deep.

thank you !

[Konstantinos]

Hi,

I am not aware of any such implementation but a google search returned this:

http://rosettacode.org/wiki/Chinese_remainder_theorem

You can see "the exact code to learn its implementation in deep" in many languages. You will use C/C++ in NS3. 

Not, the packet splitting will depend on where you want to implement this mechanism. For example there is IP fragmentation.

[Tommaso Pecorella]

Hi,

there are a couple of papers about CRT and packet splitting, but I haven't checked them yet. They may be interesting to implement (or not). However, mind that more than often when you re-implement something described in a paper you find that things are not so shiny as they seems.

Cheers,

T.

[Atul Chavan]

hey Konstantinos

i went through that link...

thing is i am not getting how exactly CRT i.e chinese remainder theorem can be used for packet splitting..

i thought its exact implementation in ns3 or ns2 would help me to understand it clearly !

me too have implementation of CRT algorithm in c++ but how it is turned into helpful for packet spliting thats the real challenge to me :(

[Atul Chavan]

hey tommaso

i agreed..

i am having one transaction paper over the same  f

as its a transaction paper being honest i am sharing this hardly few things i could understand...

the relation between CRT & packet splitting is the big question for me...

I am attaching here that transaction paper..

-PFA

Thank you !

[Tommaso Pecorella]

Hi,

from a fast check to the paper, I strongly doubt it could be replicated in a realistic network simulator like ns-3.

The point is that the authors "forgot" to consider some important network elements (headers, collisions and so on).

I don't know why you're trying to fully understand the paper tho...

Cheers,

T.

[Atul Chavan]

Dear Tommaso ,

As u asked me why do i need to understand  that paper here is my answer...

I am doing my master of engineering in computer. WSN is my domain & i am working over a project whose major focus is on Reliability & energy saving.
Before forwarding packets i need to split them to save energy & later over those small tiny packets i will be applying my algorithm so they can reach to the base station with less packet loss ratio.

So its very much necessary for me to understand the paper in order to implement it. I cant start my project without it as its my very first module i.e packet splitting !

after reading all this what would you suggest Tommaso ? anyhow i need to implement it.. what way you can help me in this ? so accordingly i can ask you my difficulties !

Thank you !

[Tommaso Pecorella]

Hi,

I don't know if I can help, that is a very mathematical paper. Plus I know one of the authors, and this... well, it doesn't help.

As a general thing, I'd say that the idea of splitting the packets to save energy is not really a good idea. It would be if the overhead wasn't there, but each split packet will have a new MAC header, and the MAC header is a huge penalty. It's worth considering that in Wi-Fi the energy saving is made through packet *aggregation*... enough said.

The reliability is a different thing, and packet splitting may help there.

Overall... well, if I could forecast a research topic outcome in one post I'd be a wizard (and I'm not).

Anyway, you're asking for suggestions. I'd start by defining the exact packet structures and headers needed to split and rebuild the packets. Possibly in a way that can be integrated with some standard.

the actual algorithm for splitting up a packet... I'd help but it's not my field, sorry.

Cheers,

T.