I was part of a team that deployed Navini Networks (now owned by Cisco) kit into Ireland in 2004 doing 'crazy' stuff like beamforming and creating No Line of Site connectivity up to 8kms from the base station to a mobile device at 2.4GHz and then later 3.5GHz. The 'only' reason it worked was due to the high EIRP (much higher than legally allowed at the time) that was emitted by the base station. The weak responding signal massively impacted upload speeds (as can be affected).
Ive also worked with people like Flarion Flash OFDM (later bought by Qualcomm) and IPWireless .... The laws of physics still remained. Ive seen some good stuff but there is no free lunch. There is no way to get 'thousands' onto single access points unless you completely rewrite the wifi stack which means that now you no longer have a proprietary system. That said, I have been out of the wireless game for a few years now, perhaps someone has invented a miracle solution .... On Tue, Dec 9, 2014 at 2:10 PM, Christian de Larrinaga <[email protected]> wrote: > Hi Gord, > > ? http://www.artemis.com/pcell > Be interested to learn if anybody has got under the veil of this one? > > Christian > > Gord Slater wrote: > > > > > > On 9 December 2014 at 13:32, Richard Halfpenny > > <[email protected] > > <mailto:[email protected]>> wrote: > > > > On 09/12/2014 11:37, Peter Knapp wrote: > > > I spent quite a while on this with a couple of larger multi-vendor > WAP > > > distributors and they have clients of the big exhibition centres > and > > > music venues, and they use extremely steerable APs with shielding, > much > > > as you do with an audio line array if you are familiar, and don’t > have > > > anywhere near the number of users per AP you are citing. > > > > +1, the antenna/RF design will make or break this kind of solution. > > > > > > as an RF guy, I approve this message. > > > > In fact, you might be forced to use "bad antennaes" like leaky feeders > > placed in metal troughs or spaced above them as corner reflectors under > > a raised floor to get stuff to work best, to limit the numbers of > > devices per physical radio. > > neat.. > > > > > I've had to pick up the total mess of this on several SCADA-type > > fitments. It can only be worse for one-off events where failure is not > > an option. I had to borrow over 300 humans to move around carrying > > sensors to prove one fitment didn't work the way the vendor said it did. > > Fine when you have minimum wage people that are bored. Some of the bods > > I had to borrow were on 150k+ a year and very pissed off to be in a cold > > icy refinery - one of my worst ever site visits; so, so glad I'm > > retired from site ops like that now. > > not geeks then > > > . > > Beamsteering, $latest-craze and "software solutions" are unhelpful above > > a certain empirical threshold. You simply cannot break the laws of > > physics by even a tiny fraction of a percent. If you do exceed the > > *known* laws, please document it, make it repeatable and become rich and > > famous. Otherwise, you've been conned :) > > > see above > > -- > > sent via Gmail web interface, so please excuse my gross neglect of > > Netiquette > > -- > Christian de Larrinaga > > ------------------------- > @ FirstHand > ------------------------- > [email protected] > ------------------------- > >
