[Ns-developers] Quick DIFS related question
Jens Mittag
jens.mittag at kit.edu
Wed Sep 1 01:46:43 PDT 2010
Hi,
I am currently writing some test cases in order to check whether the
CCA_BUSY mechanism is correctly implemented in our physical layer
simulator for OFDM-based 802.11 (which will finally be released next week).
While writing the tests, I discovered that (in AdhocWifiMac) after a
node transmits a message, it has to wait longer than DIFS to transmit
the next message in its queue. I don't know the parameters by heart, but
normally, a node has to wait a DIFS (34microseconds in a 20 MHz channel)
before being able to transmit its own message. This is what I observe if
the node sends its first message. For the next message, the waiting time
before being able to transmit to the channel is surprisingly
106microseconds. Is this how it should be? To me this value seems a
little akward. In the above scenario, the node is the only one who is
sending, so no interferers.
Is this because the node has to choose a backoff for the second frame?
Is there a difference between a new frame that arrives from upper layers
during the IDLE period of MAC and a frame that arrives during RX, TX or
CCA_BUSY?
Thanks for your help
Jens
--
Dipl.-Inform. Jens Mittag
Karlsruhe Institute of Technology (KIT)
Institute of Telematics
Decentralized Systems and Network Services Research Group
Zirkel 1, 76131 Karlsruhe, Germany
Building 20.20, Room 356.1
Phone: +49 (721) 608 5768
Fax: +49 (721) 608 6789
http://dsn.tm.uni-karlsruhe.de/
More information about the Ns-developers
mailing list