openWNS WiFiMAC

TXOP changes

Asked by Mohammad Siddique on 2009-08-18

Hello Sebastian/All,

When I was going to merge 'locally in Bremen' the 802.11e code from rahmatullah to newer version of openwns-wifimac (r61), I have found as following.

If i understand correctly, previously TXOP FU has some friends like phyUserName, nextFrameHolderName which are missing in new TXOP friend list and new friends are added like protocolCalculatorName, txopWindowName which might be for the same purposes.

As rahamatullah follow the previous TXOP code for his TXOP11e, he is using some methods from "PhyUser.cpp" as follows

wns::simulator::Time PhyUser::getPreambleDuration
wns::simulator::Time PhyUser::getPSDUDuration

so i think there are two options for me during merging,

1. add the backward compatibility in new version by means of coping "NextFrameGetter" FU and merge back the function mentioned above in PhyUser.cpp etc.

2. adapt 11e (TXOP11e) code according to new TXOP code

In my view, first one might be faster, however, in the long run might go for the second.

Can you please comment on this and if possible give some more information.

Kind regards,
Mohammad Siddique

Question information

Language:
English Edit question
Status:
Answered
For:
openWNS WiFiMAC Edit question
Assignee:
No assignee Edit question
Last query:
2009-08-18
Last reply:
2009-08-19
Harald Radke (rat-comnets) said : #1

Hello Mohammad....uhm, I knew that day would come... (:

Actually the "new" TXOP implementation became "necessary" for the draftN FUN, since with aggregation, determining possible follow-up transmissions became a little bit more complicated...therefore TXOP now simply asks the TXOPWindow friend for the duration of a possible next transmission, relying on that FU to correctly calculate sizes and duration. (Which unfortunatly isn't the case right now in the current draftN FUN where the buffer plays the role of the TXOPWindow...which however is not a TXOP but a buffer problem (-: ). I general the current TXOP still might need some cleanup, but BASICALLY should work

so much for the FU history (: regarding your question I would like to leave it tho the WiFiMAC grandfathers to decide which way to go since I am pretty sure they have a better overview and see the big picture

Greez,

Harald Radke

Maciej Muehleisen (mue-comnets) said : #2

I would go for number two. The functions used in TXOP11e are pretty much all the same as in TXOP. Back then we just had a look how TXOP does it and then reused the same code in TXOP11e. So again everything should be present in current TXOP FU code.

Can you help with this problem?

Provide an answer of your own, or ask Mohammad Siddique for more information if necessary.

To post a message you must log in.