I take it from your response you are running APs in DFS channels and I do agree that in many situations it is completely fine to do so. Where you will run into trouble is where an AP has one 5 GHz radio and its on a DFS channel AND a client doesn’t support that channel, you have an issue. Some environments wouldn’t lend itself to that and others would.
And here’s the thing about Wi-Fi design that has been mentioned; there is no one size fits all for design. Are there cases where you could have one AP per classroom with dual 160 MHz channels? Sure. If your building and walls are made of solid concrete, go for it. If implementers stay disciplined and keep deployed channel sizes in check then yes, co-channel interference shouldn’t be an issue. But we all know that won’t happen. As an example, think of a multi-story deployment. Even with 18 40 MHz channels, could there be co-channel interface? Sure. Dual 5 GHz APs, rooms above and below… won’t take much time at all to have co-channel interference again. Even as we get more spectrum we are going to continue to find ways to use (and abuse) it. :-) GT From: The EDUCAUSE Wireless Issues Constituent Group Listserv <WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU> on behalf of "Jeffrey D. Sessler" <j...@scrippscollege.edu> Reply-To: The EDUCAUSE Wireless Issues Constituent Group Listserv <WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU> Date: Friday, May 27, 2016 at 2:00 PM To: <WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU> Subject: Re: [WIRELESS-LAN] Beacon Intervals My experience in 5GHz so far been that co-channel interference isn’t a problem, even in one AP per classroom situations. Simply stated, the propagation of 5 GHz is so poor thru our building materials and the number of non-overlapping channels is so plentiful as to render it a non-issue. I do agree that 5 GHz APs that don’t support DFS channels would be challenging, but then why would anyone not use DFS and/or purchase an enterprise AP that didn’t support them? Are folks still running into significant client devices that don’t support them? Additionally, with FCC Order 14-30, we immediately gain a few more channels and can run some of the existing at higher power. Once the future channels are released and clients support them, we’ll have a total of 36 non-overlapping channels (20 MHz), 18 (40 MHz), 9 (80 MHz). Coupled with vendor RRM and dynamic channel-width selection, I’m hard pressed to see where co-channel interference is going to be an issue in 5GHz except in outliers such as stadium deployments. Jeff From: "wireless-lan@listserv.educause.edu" <WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU> on behalf of GT Hill <g...@gthill.com> Reply-To: "wireless-lan@listserv.educause.edu" <WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU> Date: Friday, May 27, 2016 at 10:07 AM To: "wireless-lan@listserv.educause.edu" <WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU> Subject: Re: [WIRELESS-LAN] Beacon Intervals There is a fine line here. Signal strength best practices can’t be arbitrarily increased just because 256 QAM requires it. Added signal strength is of course achieved with either increased Tx power or added access points, both of which will contribute to increased co-channel interference. An increase in modulation rate is never a good idea if it increases co-channel interference. As an added comment, dual 5 GHz APs that don’t support DFS channels have limited advantages. Each AP will be occupying 2 of the 8 available channels which makes design look like a 4 channel model. While better than 2.4 GHz for sure, not nearly what is possible with DFS support. With DFS support each AP will have one radio on a non DFS channel and the other radio on a DFS channel. This allows for increased bandwidth while still accounting for client devices that don’t support DFS channels. GT From: The EDUCAUSE Wireless Issues Constituent Group Listserv <WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU> on behalf of "Jeffrey D. Sessler" <j...@scrippscollege.edu> Reply-To: The EDUCAUSE Wireless Issues Constituent Group Listserv <WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU> Date: Friday, May 27, 2016 at 11:30 AM To: <WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU> Subject: Re: [WIRELESS-LAN] Beacon Intervals So basically, you have someone who’s livelihood depends on charging you for site surveys and design, telling you why using a simple one AP per room is a bad idea. I certainly wouldn’t disagree with some of his findings when when designing around 2.4, but when designing around 5 GHz, and especially .11ac, many of the concerns necessitating comprehensive surveys (even in one AP per room deployments) tend to evaporate. Focusing on 5-GHz: His argument is then about installing more AP’s then necessary and wasting money, which is 100% valid, but does that theoretical waste exceed the cost of the survey? Is the one AP per classroom over designed for the legacy, current, or future standards? The write up also shows how quickly best practices change when new standards emerge e.g. where -65 dB was sufficient for the top rate in 802.11n (20-MHz channel), in .11ac that requirement is now -59 dBm in 20-MHz, and -49 dBm in 160-MHz. This is why I say it’s inevitable that we’ll get to one (or more) APs per room if the desire is to maximize performance. It’s a really interesting conversation especially if one starts to consider tuition cost per minute of class time as a factor in a design. Last but not least, some of his assumptions don’t necessarily hold up in higher-ed, and more so between private/public. I’ll also go out on a limb and say that with the advent of flex-radios, we’re well on our way to solving 2.4 GHz problem. 2.4-GHz can gracefully ride into the sunset while those one AP per classroom devices slowly transition to running fulltime as two 5-GHz radios. Jeff From: "wireless-lan@listserv.educause.edu" <WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU> on behalf of Samuel Clements <scleme...@gmail.com> Reply-To: "wireless-lan@listserv.educause.edu" <WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU> Date: Friday, May 27, 2016 at 7:25 AM To: "wireless-lan@listserv.educause.edu" <WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU> Subject: Re: [WIRELESS-LAN] Beacon Intervals Sure, but there is a great writeup on that exact topic that does a good job in my stead: http://www.wlanpros.com/wp-content/uploads/2014/04/Why-One-AP-Per-Classroom-Approach-is-Wrong-.v3.pdf In short, that may be a design you end up with, but assuming it's correct to begin with is a premise that should not be used. Proper WiFi design (including disabling radios or converting them to 5GHz radios if you have hardware that can do that) is of paramount importance in any environment that believes their network is of any measurable importance. Remember that disabling lower data rates & changing beacon intervals can *mitigate* poor design - but there is always a trade off (client compatibility being chiefest). I don't necessarily disagree that in some environments, one AP per classroom is what you would net, but I've seen far too many environments where they over bought and a 1.5 classroom per AP (or some other measure) would have supported the load just fine. I hate to see people waste money when it could have gone to some other area of technology to further the end goal - education. -Sam On Fri, May 27, 2016 at 9:18 AM, Jeffrey D. Sessler <j...@scrippscollege.edu> wrote: Sam, would you please explain your position on one AP per classroom being a mis-design? Do you have data on this you could share? In my environment, I’ve found that in order to properly deploy 5 Ghz and .11ac, it’s pretty much inevitable that we’ll get to one AP per room, especially if one desires consistent and universal coverage. Data from existing spaces clearly show gaps in 5GHz coverage when using an every-other room scheme. Now if you are talking about 2.4 GHz I may agree with you, but even there, with removal of lower data rates, and a low-power microcell design, the data suggests it’s working very well. Jeff From: "wireless-lan@listserv.educause.edu" <WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU> on behalf of Samuel Clements <scleme...@gmail.com> Reply-To: "wireless-lan@listserv.educause.edu" <WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU> Date: Thursday, May 26, 2016 at 6:38 PM To: "wireless-lan@listserv.educause.edu" <WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU> Subject: Re: [WIRELESS-LAN] Beacon Intervals Remember folks, there is such a thing as too much RF and in the edu space, this occurs quite commonly due to the One AP per Classroom mis-design advice that was making the rounds some time ago... -Sam ********** Participation and subscription information for this EDUCAUSE Constituent Group discussion list can be found at http://www.educause.edu/groups/. ********** Participation and subscription information for this EDUCAUSE Constituent Group discussion list can be found at http://www.educause.edu/groups/. ********** Participation and subscription information for this EDUCAUSE Constituent Group discussion list can be found at http://www.educause.edu/groups/. ********** Participation and subscription information for this EDUCAUSE Constituent Group discussion list can be found at http://www.educause.edu/groups/. ********** Participation and subscription information for this EDUCAUSE Constituent Group discussion list can be found at http://www.educause.edu/groups/. ********** Participation and subscription information for this EDUCAUSE Constituent Group discussion list can be found at http://www.educause.edu/groups/.
