This does not surprise me that they ran in to this. The 802.11b/g spec is designed as a carrier sense collision avoidance network where all users are within range of each other so they can be polite and not transmit over top of another. When you have all sorts of users and multiple access points things get confusing to the protocol very easily. What further causes problems is that the access points cause the interference to each other or cause them to hold off to wait for other traffic on the same channel, mainly due to their high gain antennas on both locations and being mounted in a higher location. On the EarthLink Philly project we were able to overcome some of these problems with clutter. The three story row houses and dense trees made life difficult to design a network, but at the same time contained the signal to smaller areas for each access point, thus minimizing some of the traits displayed in this article. I've seen too many office or college campus situations where they don't like the throughput or coverage of their wireless network. They throw up more access points thinking that will solve the problem. It rarely does unless they figure out how to better manage the signal patterns of each access point. While 802.11 b/g gives you a choice to mix and match equipment, proprietary protocols usually handle this collision problem much better. If you will never load up a network with this type of traffic, you can get away with it. If you need to squeak out every bit that you can from the available spectrum (and all WISPs will when video and voice become the norm on Ethernet), some sort of managed on air protocol will work out the best. Right now with bursty internet usage patterns, things don't seem to be that bad. When the internet patterns change to a more constant bandwidth demand, different technologies will fair better than others. When that happens I wonder about pricing models that rely on over subscription..........
Thank You, Brian Webster -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of CHUCK PROFITO Sent: Wednesday, February 13, 2008 10:21 PM To: 'WISPA General List' Subject: [WISPA] WLAN stress test uncovers 802.11 performance problems WLAN stress test uncovers 802.11 performance problems The tests confirm two troubling issues for high-density nets Everyone on this list probably already knows this, especially if you have read Jack's book, but John Cox from Network World did a good job explaining it. Usage vs. self interference and scaling up the access points. Interesting, where they had the antennas folded flat, it worked better, but they don't know why...he forgot it became an unintentional sector. I bet if you put in Alvarion with packet and VoIP priority it would blow this test out of the water. Maybe? What do you folks think? http://www.arnnet.com.au/index.php/id;1526372596;pp;1 Or http://tinyurl.com/2cebwk Chuck Profito 209-988-7388 CV-ACCESS, INC [EMAIL PROTECTED] Providing High Speed Broadband to Rural Central California ---------------------------------------------------------------------------- ---- WISPA Wants You! Join today! http://signup.wispa.org/ ---------------------------------------------------------------------------- ---- WISPA Wireless List: wireless@wispa.org Subscribe/Unsubscribe: http://lists.wispa.org/mailman/listinfo/wireless Archives: http://lists.wispa.org/pipermail/wireless/ -------------------------------------------------------------------------------- WISPA Wants You! Join today! http://signup.wispa.org/ -------------------------------------------------------------------------------- WISPA Wireless List: wireless@wispa.org Subscribe/Unsubscribe: http://lists.wispa.org/mailman/listinfo/wireless Archives: http://lists.wispa.org/pipermail/wireless/
