Re: [ns] Bug in NS2's 802.15.4 implementation?

2007-02-17 Thread Daniele Messina 

Hi Mukul, and hi all.

In section 7.1.3.1.3 of the 802.15.4 2003 standard it is said that:

If the MLME of an unassociated device successfully receives an
acknowledgment to its association request command, the MLME will wait for at
most aResponseWaitTime symbols for the association response command to
become available. If the MLME of the device does not extract an association
response command frame from the coordinator within this time, it will issue
the MLME-ASSOCIATE.confirm primitive with a status of NO_DATA.

This could be an explanation for this reduced transaction time at the
coordinator: since sending a response after that the aResponseWaitTime has
expired at the node which requested the association is useless. Hope this
helps.

Bye,

Daniele M.

[ns] Bug in NS2's 802.15.4 implementation?

2007-02-14 Thread mukul 



- Forwarded message from [EMAIL PROTECTED] -
Date: Wed, 14 Feb 2007 11:34:15 -0600
From: [EMAIL PROTECTED]
Reply-To: [EMAIL PROTECTED]
 Subject: Bug in NS2's 802.15.4 implementation?
  To: ns-users@isi.edu

Hi,

It looks like there may be a bug in Mac802_15_4::mlme_associate_response
function. The transaction corresponding to an association response is being
scheduled to expire after 2*aResponseWaitTime (or macResponseWaitTime)
duration. As per the standard, any transaction should expire only after
macTransactionPersistenceTime duration as is being done in
Mac802_15_4::mcps_data_request function.

I was wondering why the association response transaction is made to expire so
soon (2*aResponseWaitTime is way smaller than macTransactionPersistenceTime).
Is it a bug or there is some other explanation?

Thanks,
Mukul


- End forwarded message -