Linux rtl8192se driver (Fedora)
Some problems with wifi module on fedora (may be only with rtl8192se driver).
Here is the discription of the problem (link)
I have a laptop (Zoostorm Kangaroo VME50) running Fedora 14. I try to keep up-to-date, but usually only get around to running updates every week or 2.
I’m using rtl8192se wireless drivers, downloaded from a link someone posted on here. When I got the last kernel update, I just rebuilt the drivers rather than checking if there’s a new version.
Today I was trying to connect to a VPN, so did `yum install kde-plasma-networkmanagement-openvpn (assuming that normal dependency resolution would pull in all the other relvant packages … I’m not sure if that’s sane or not). Among the new packages, NetworkManager was ‘updated for dependencies’.
After a reboot, the VPN tab in NetworkManager appeared, but I was immediately prompted for my WEP key. (Not my house, not my network. The hex key is written down on a piece of card, so that’s quite a hassle)
After entering the key, the networkmanagement plasma widget kept on showing the status as “Waiting for authorization” … on reboot, or trying a different network, it does exactly the same.
When I plugged in the wired network cable, the KDE crash handler pops up with a segfault from NetworkManager. Unfortunately, the window vanished before I could take in any more than the fact that most of the stack trace visible was full of references to Qt* stuff.
The plasma widget still shows the icon for wireless, with the little ‘key’ that normally appears when its asking for a passphrase or key; the wireless network (“FAMILY”) disappeared from the list of available networks, but the interfaces list still shows “WLAN Interface: Waiting for authorization”. However, it appears the wired network is connected, so hopefully someone can tell me what’s the issue here, or at least help to track it down.
I don’t have a clue where to look.
The best way to solve this problem till now is to make downgrade
yum downgrade NetworkManager NetworkManager-glib
than restart Network manager service
MLE, MAP, Bayes
Estimation: ML, Bayes, MAP
Principles and calculus at http://www-users.cselabs.umn.edu/classes/Fall-2008/csci5525/notes/estimate.pdf
Estimation
Simple explanation here:
09.11.10Given the evidence X, MLE considers the parameter vector Θ to be a constant and seeks out that value for the constant that provides maximum support for the evidence. ML does NOT allow us to inject our prior beliefs about the likely values for Θ in the estimation calculations.
MAP allows for the fact that the parameter vector Θ can take values from a distribution that expresses our prior beliefs regarding the parameters. MAP returns that value for Θ where the probability prob(Θ|X) is a maximum.
Both ML and MAP return only single and specific values for the parameter Θ. Bayesian estimation, by contrast, calculates fully the posterior distribution prob(Θ|X). Of all the Θ values made possible by this distribution, it is our job to select a value that we consider best in some sense. For example, we may choose the expected value of Θ assuming its variance is small enough.
The variance that we can calculate for the parameter Θ from its posterior distribution allows us to express our confidence in any specific
value we may use as an estimate. If the variance is too large, we may declare that there does not exist a good estimate for Θ.
Viterbi algorithm Demo
En excellent tutorial on Viterbi Algorithm was fount at http://www.telecom.tuc.gr/~ntsourak/demo_viterbi.htm
| Posted in Linux | 1 Comment »