Wireless PCI adapter for GNU/Linux

Hi,

I'm setting up a wireless network in my home to share cable internet. I
already have a D-Link wireless router (DI-624+) which I am just using as a
non-wireless router now.

Could someone provide some recommendations as to good wireless PCI or USB
adaptors that are compatible with Debian GNU/Linux. I am happy to upgrade
my kernel to the 2.6 stream if necessary. I prefer drivers licensed
under GPL.

Is using a wireless access point the way to go? This way I don't need to
worry about drivers.

Does anyone have any opinions on the D-Link DWL-2000AP+?


Thanks

On Thu, 13 May 2004 16:12:38 +1000, Glenn Meehan wrote:

I also need a wireless PCI or USB adapter for another computer running
Windows 2000. I have read horrendous reports about he DWL-G520+. Could
someone please recommend some alternatives.

Which gives greater range. 802.11b or 802.11g.
Which standard is less resistant to brick walls?

Or does it depend on the card in question?


Thanks

"Glenn Meehan" wrote in message ..
>
> I also need a wireless PCI or USB adapter for another computer running
> Windows 2000. I have read horrendous reports about he DWL-G520+. Could
> someone please recommend some alternatives.
>
> Which gives greater range. 802.11b or 802.11g.
> Which standard is less resistant to brick walls?
>
> Or does it depend on the card in question?
>
>
> Thanks


B & G use the same frequency so they have the same characteristics as far as
distance and penetration of walls.

Don Woodward

In article <D6Noc.1266$(E-Mail Removed)>,
News Account <(E-Mail Removed)> wrote:
>"Glenn Meehan" wrote in message ..
:> Which gives greater range. 802.11b or 802.11g.
:> Which standard is less resistant to brick walls?

:B & G use the same frequency so they have the same characteristics as far as
:distance and penetration of walls.

Yes and no. They do use the same frequencies, but the wavefronts are
used differently. G uses a lot of sub-carriers each of which is held
for a longer time. That reduces multipath losses quite a lot because
most of the signal bounces will arrive while the bit is being
held, whereas for B the bounces were often arriving and interfering
with one of the following direct-path bits.

If the building material actually absorbs the 2.4 GHz frequency
then or if the building material is essentially a homogenous mass
and you thus only considering loss through attentuation, then, Sure,
B and G will have the same issues. But building materials are usually
*not* homogenous: you get reflections off of surfaces, and you get
different propogation characteristics of brick versus mortar,
and you get signals propogated along wires or nails -- so in practice,
G's better multipath characteristics has the advantage. [But put
a thick enough wall inbetween and it's not going to matter.]
--
Oh, yeah, an African swallow maybe, but not a European swallow.
That's my point.

Great info - thanks!

I'm very familiar with multipath effects in voice communications in the 70cm
(440mhz) ham band living here in the mountainous North Georgia region -
never really thought about how multipath could affect WiFi.

Don Woodward


"Walter Roberson" <(E-Mail Removed)> wrote in message
news:c808dl$j9q$(E-Mail Removed)...
> In article <D6Noc.1266$(E-Mail Removed)>,
> News Account <(E-Mail Removed)> wrote:
> >"Glenn Meehan" wrote in message ..
> :> Which gives greater range. 802.11b or 802.11g.
> :> Which standard is less resistant to brick walls?
>
> :B & G use the same frequency so they have the same characteristics as far
as
> :distance and penetration of walls.
>
> Yes and no. They do use the same frequencies, but the wavefronts are
> used differently. G uses a lot of sub-carriers each of which is held
> for a longer time. That reduces multipath losses quite a lot because
> most of the signal bounces will arrive while the bit is being
> held, whereas for B the bounces were often arriving and interfering
> with one of the following direct-path bits.
>
> If the building material actually absorbs the 2.4 GHz frequency
> then or if the building material is essentially a homogenous mass
> and you thus only considering loss through attentuation, then, Sure,
> B and G will have the same issues. But building materials are usually
> *not* homogenous: you get reflections off of surfaces, and you get
> different propogation characteristics of brick versus mortar,
> and you get signals propogated along wires or nails -- so in practice,
> G's better multipath characteristics has the advantage. [But put
> a thick enough wall inbetween and it's not going to matter.]
> --
> Oh, yeah, an African swallow maybe, but not a European swallow.
> That's my point.