#1008: wrong (values do not make sense) signal level and link quality
----------------------------------------+-----------------------------------
Reporter: [EMAIL PROTECTED] | Owner: scottr
Type: defect | Status: assigned
Priority: major | Milestone: version 0.9.3
Component: madwifi: 802.11 stack | Version:
Resolution: | Keywords:
Patch_attached: 1 |
----------------------------------------+-----------------------------------
Comment (by [EMAIL PROTECTED]):
I've checked the patch proposed by Scott.
It looks like it solves the problem.
Still I'm not completely happy because of two reasons:
1. The rssi is passed as int to ieee80211_input. It is used in several
places besides set_quality, where it is also expected to be unsigned. For
example: uint8_t ni_rssi in struct ieee80211_node.
So negative values could result in overflow in other locations as well. I
propose to clip the rssi value upon entry in ieee80211_input().
2. I have seen fluctuations in the received noise level from -75..-96:
almost 20 dB. These fluctuations are much larger than in the rssi. So
using the actual noise in the calculation of the signal level may make it
less reliable.
For this reason I decided to revert to the 0.9.2.1 version, which uses a
hardcoded noise level, in combination with my patch. That works OK. I have
attached my patch for the 0.9.2.1 version.
--
Ticket URL: <http://madwifi.org/ticket/1008>
MadWifi <http://madwifi.org/>
Multiband Atheros Driver for Wireless Fidelity
-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Madwifi-tickets mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/madwifi-tickets
