[rescue] 802.11 + VoIP

[Geoffrey S. Mendelson]

Eric Webb wrote:

> I see a lot of opinions that 802.11 and VoIP are a bad combination, but I need 
> more proof and stats if anyone has anything.  (I'm kind of anti-wireless 
> anyway when it comes to wanting performance, which voice seems to demand.)

Performance is not an issue with VoIP, latency is. To get a decent VoIP 
connection you need to sustain 64k BITS per second of DATA, which means
about 96k-128k of IP throughput. Just about any WiFi network can do that,
but if the delay is too long in getting a packet out there, then there
will be echo and other problems.
 
> Also, do they even make 802.11-based VoIP handsets?  Can someone point me in 
> the direction of a few models?

I was just searching and haven't seen any. The closest you can come is
either a USB phone for your WiFI equiped laptop, or an ethernet phone
and a WiFi router. 

Or for the real gearheads, a Cisco or Packet8 ATA (analog telephone adaptor),
a WiFi router, and a "real" (POTS aka analog) phone.

This is not as far fetched as you might think. Sambo will be here in a week,
bringing WiFi access points and a Packet 8 ATA. I've got a Cisco 186 comming
around the same time with service on a competing network. We can do some
"real" testing in our spare time. :-)

I've spent several hours this week getting QOS (quality of service) routing
to work on my Linux based router/firewall machine. Tonight's installment
was to put traffic limiting on the LAN side, so my son does not swamp
the connection to the outside world with his downloads. 

He's doing intensive self study in the visual aspects of human anatomy
and sexuality. But with the QOS routing, it should not affect any VoIP
stuff I do.

Geoff.
-- 
Geoffrey S. Mendelson gsm at mendelson.com 972-54-608-069
Icq/AIM Uin: 2661079 MSN IM: geoffrey_mendelson at hotmail.com (Not for email)
Carp are bottom feeders, koi are too, and not surprisingly are ferrets.