From: Ghassan AKRAM on
Hi

I want to create a model of MC-CDMA system build usin matlab and then I want to test the performance of variuos types of TCP over this model.

I think that is complete project but I hope to get assistance to do.

Ghassan
From: Walter Roberson on
Ghassan AKRAM wrote:

> I want to create a model of MC-CDMA system build usin matlab and then I
> want to test the performance of variuos types of TCP over this model.
>
> I think that is complete project but I hope to get assistance to do.

Various kinds of TCP ? You mean like TCPv4 and TCPv6 ? Those are the
only official versions of TCP still extent. Some formalized alternatives
to TCP are listed at
http://en.wikipedia.org/wiki/Transmission_Control_Protocol#Alternatives

If you are using your own TCP-*like* protocol then you are strongly
encouraged to use a different IP protocol number (rather than TCP's 6 or
UDP's 17) such as the 253 and 254 that are reserved for experimenting.
http://en.wikipedia.org/wiki/List_of_IP_protocol_numbers
From: Ghassan AKRAM on
Walter Roberson <roberson(a)hushmail.com> wrote in message <do_0o.18054$0A5.4784(a)newsfe22.iad>...
> Ghassan AKRAM wrote:
>
> > I want to create a model of MC-CDMA system build usin matlab and then I
> > want to test the performance of variuos types of TCP over this model.
> >
> > I think that is complete project but I hope to get assistance to do.
>
> Various kinds of TCP ? You mean like TCPv4 and TCPv6 ? Those are the
> only official versions of TCP still extent. Some formalized alternatives
> to TCP are listed at
> http://en.wikipedia.org/wiki/Transmission_Control_Protocol#Alternatives
>
> If you are using your own TCP-*like* protocol then you are strongly
> encouraged to use a different IP protocol number (rather than TCP's 6 or
> UDP's 17) such as the 253 and 254 that are reserved for experimenting.
> http://en.wikipedia.org/wiki/List_of_IP_protocol_numbers


Thanks for your answer, But I mean that after I Get efficient MC-CDMA model , I will try to test the performance of ordinary TCP, snoope, freeze, I-TCP, Mobile-TCP, Reno, Vegas, and others.
regards
From: Walter Roberson on
Ghassan AKRAM wrote:
> Walter Roberson <roberson(a)hushmail.com> wrote in message
> <do_0o.18054$0A5.4784(a)newsfe22.iad>...
>> Ghassan AKRAM wrote:
>>
>> > I want to create a model of MC-CDMA system build usin matlab and
>> then I > want to test the performance of variuos types of TCP over
>> this model.
>> > > I think that is complete project but I hope to get assistance to do.
>>
>> Various kinds of TCP ? You mean like TCPv4 and TCPv6 ? Those are the
>> only official versions of TCP still extent. Some formalized
>> alternatives to TCP are listed at
>> http://en.wikipedia.org/wiki/Transmission_Control_Protocol#Alternatives
>>
>> If you are using your own TCP-*like* protocol then you are strongly
>> encouraged to use a different IP protocol number (rather than TCP's 6
>> or UDP's 17) such as the 253 and 254 that are reserved for experimenting.
>> http://en.wikipedia.org/wiki/List_of_IP_protocol_numbers
>
>
> Thanks for your answer, But I mean that after I Get efficient MC-CDMA
> model , I will try to test the performance of ordinary TCP, snoope,
> freeze, I-TCP, Mobile-TCP, Reno, Vegas, and others.

Are there existing simulation blocks for each of those? TCP itself is
notoriously hard to code _correctly_ and some of the other possibilities
such as SCTP are much more complex. I would go as far as to say that
your only hope of implementing those protocols *correctly* would be by
automatic code (or model) generation from very complete state transition
tables.

 | 
Pages: 1
Prev: textscan()
Next: Issues with AVIREAD?