Re: [manet] Hiden- & exposed terminal problem in MANET

[Sumeeth Nagaraj <snagaraj@cc.usu.edu>]

Hi Manoj,


> > Each node has a unique PN code in the network and transmits using its
own PN
> > code. The transmitter senses the channel for the known PN codes, if the
> > receiver is idle(data not for this node) then the node switches to the
> > transmitter and transmits. The ACK can be for cumulatively sent for a
set of
> > packets or for individual packet. Since the PN-codes are unique in the
> > network, the hidden node problem is addressed, the node need not be idle
> > till the completion of data transfer of neighboring nodes and this
addresses
> > the exposed node problem. Am I missing something here?
>
> A better way is to use a set of codes.  A sender includes in its
> RTS (sent over a common code) a preferred code (unused at
> that time) and receiver sends CTS if the code is acceptable.
> Then DATA-ACK takes over the selected code.


The idea of using multiple codes was to remove the RTS-CTS over head.
RTS-CTS can cause congestion problems for a netowork with high density of
nodes.

You can give a look at the channel access scheme proposed by Dr. Schlegel:
Here they have a common  header spread with a identical  code that uses the
basic aloha scheme and the data is spread by a random spreading code, that
is sent to the destinatioin in the header.
http://www2.elen.utah.edu/~schlegel/publications/KotSch01.pdf

and also, the MAC protocol proposed by Dr. Rabaey for the pico radio
project(BWRC).

Best Regards,
Sumeeth

_______________________________________________
manet mailing list
manet@ietf.org
https://www1.ietf.org/mailman/listinfo/manet