IT Shops Coping With Overloaded 2.4GHz WiFi Band

alphadogg writes "Of the 470,000 Wi-Fi connections made on a recent day at Abilene Christian University, fully 94% used the 2.4GHz band, representing an extreme example of how today's surging number of Wi-Fi clients is crowding the band least able to accommodate them. At ACU, this is not considered a problem, at least not yet. In part, that's because of careful wireless LAN design and capacity planning. And partly because a goodly percentage of mobile devices that can run on the alternative 5GHz band, do so: on that same day, 47% of the school's laptops and desktops, and two-thirds of its iPads cruised on 5GHz, via either 802.11a or 802.11n. Yet relatively few of today's Wi-Fi clients support 5GHz."

Comments

wrong title

[Anonymous Coward]

this title should read "IT shops unable to cope with overloaded 2.4GHz WiFi band" IMHO

Re:wrong title

[Anonymous Coward]

Considering TFA is about an IT shop that is coping with the overloaded 2.4GHz band, in my honest opinion, your honest opinion is fucking stupid.

Re:wrong title

[Anonymous Coward]

Wow. Two commenters with exactly the same ignorant misunderstanding of correct English!

Fully 94% ?

[ksd1337]

fully 94% used the 2.4GHz band

100% of 94% ? Or 94% used the band fully?

Re:Fully 94% ?

[Anonymous Coward]

94% of the devices making a connection used the 2.4GHz band to do so.

Re:Fully 94% ?

[Anonymous Coward]

No less than 94%

Re:Fully 94% ?

[Anonymous Coward]

Sex Panther - 69% of the time it works all the time.

Re:Fully 94% ?

[Gilmoure]

*golf clap*

WTF??

[Anonymous Coward]

94% used the 2.4GHz band


47% ... cruised on 5GHz,

Re:WTF??

[Tubal-Cain]

Doesn't 802.11n use both?

Re:WTF??

[Anonymous Coward]

94% of everything on 2.4GHz. 47% of laptops and desktops on 5 Ghz. 66% of iPads on 5 GHz. 100% reader failure.

Re:WTF??

[Lev13than]

At least RTFS - 94% of all connections used 2.4GHz, while 47% of iPads used 5GHz. Most devices are either G only or 2.4GHz N. People generally avoid 802.11a and dual-band 802.11n often isn't turned on. So those numbers are not surprising.

Re:WTF??

[loufoque]

94% of the 2.4Ghz bandwidth was used, not 94% of the people were using 2.4Ghz.

Re:WTF??

[tokul]

It was before they performed carefully planned emergency :) upgrade of infrastructure to add n/a support.

Re:WTF??

[Anonymous Coward]

94% of total devices (an umbrella including laptops, desktops, phones, tablets, etc.). 47% of the laptops and desktops, 33% of iPads. Which means that over half the laptops and desktops and 77% of the iPads were included in that 94% 2.4GHz number.

Reading comprehension for the win.

Re:WTF??

[Anonymous Coward]

It *can* use both. In practice most routers use one or the other.

Re:WTF??

[Anonymous Coward]

Did you even read the summary?

"on that same day, 47% of the school's laptops and desktops, and two-thirds of its iPads cruised on 5GHz, via either 802.11a or 802.11n."

Are you assuming that all of the school's laptops and desktops are composed of iPads? At least try to make sense when you post and read the summary yourself before accusing people of not doing it.

Re:WTF??

[ceoyoyo]

Except you read it wrong too. 47% of laptops and desktops (somehow desktops got called "mobile devices") and 2/3 (i.e. 66%) of iPads were on 5 GHz. Which suggests that the biggest class of devices on their network is smart phones, even assuming none of them were making 5 GHz connections.

Re:WTF??

[MichaelKristopeit421]

94% of all devices used 2.4 GHz...

47% of the school's laptops and desktops used 5 GHz...

which simply means most of the connections to the school's network are made by devices that the school doesn't own.

you're an idiot... you've been labeled "Insightful" by the other idiots that moderate this internet website chat room message board.

slashdot = stagnated

Re:WTF??

[Obfuscant]

94% of the 2.4Ghz bandwidth was used, not 94% of the people were using 2.4Ghz.

No, of the 470,000 connections, 94% of them were made using 2.4GHz. Not "94% of the bandwidth was used". "Fully" is an adjective adding stress to how high the number 94% is.

They then switch units from "connections" to "systems" and report that 47% of the laptops and desktops used 5GHz. Also 2/3 of the iPads.

I suspect that laptops and certainly desktops account for very few "connections", since the connection is made once and held open. There are also a very large number of smart devices that connect and disconnect repeatedly during the day, like my fancy new smartphone. It changes connections when I walk from one end of the building to the other, and when I walk outside it changes depending on how close I am to which node. You can also count every connection from every smart device that is carried onto campus by every visitor, even if they don't authenticate. That's why the percentage of connections is so large, but I bet the percentage of bandwidth used is much different.

Re:WTF??

[Hymer]

Nope... 802.11n may either run on 2.4GHz (b/gn) or 5GHz (an) but needs dual radios to run on both frequencies on the same time.
True dual band AP's are quite expensive, even switchable band AP's cost much more than b/gn AP's: a Zyxel NWA-3166 has a list price of USD 200.

Re:WTF??

[suomynonAyletamitlU]

Laptops, desktops, and iPads do not make up a majority of connections. Nearly half of that subset can (and does) access the 5GHz band, but even so, 94% of the total network load is on 2.4GHz

Re:WTF??

[kelemvor4]

D link sells a number of dual band dual radio ap's for under $100. Pretty sure they are not the only ones.

Re:WTF??

[Anonymous Coward]

94% used the 2.4GHz band

Occupy 2.4 ghz band. We are the 6%.

5 GHz sucks

[stevegee58]

At least on my home network. Much slower and shorter range.

Re:5 GHz sucks

[Anonymous Coward]

You must be doing it wrong...

Re:5 GHz sucks

[sexconker]

No, hes not doing it wrong. 5 GHz is shit for anything but open-air.

Re:5 GHz sucks

[Anonymous Coward]

I've always heard - and with my RF knowledge would have figured myself - that 5GHz has less range, but for some reason my house chooses to be the exception to the rule. I've always had trouble getting 2.4GHz to reach to certain portions of the house, but when I added 5GHz (Apple Airport Extreme) I can now easily get a signal from anywhere in the house and even outside.

In my neighborhood, I definitely do much better on 5GHz, I can see 5-6 (check that, right now it's 9!) neighbors' APs on 2.4GHz, but so far I'm the only one on 5. I see slow-downs or at least "hiccups" periodically on 2.4, smooth sailing on 5.

Re:5 GHz sucks

[element-o.p.]

...but for some reason my house chooses to be the exception to the rule. I've always had trouble getting 2.4GHz to reach to certain portions of the house, but when I added 5GHz (Apple Airport Extreme) I can now easily get a signal from anywhere in the house and even outside...

There's the phenomenon...

...I can see 5-6 (check that, right now it's 9!) neighbors' APs on 2.4GHz, but so far I'm the only one on 5...

...and there, I'd say, is your explanation for it.

Re:5 GHz sucks

[Miamicanes]

The main problem is that 802.11A/N is totally the wrong kind of protocol for crowded environments where you have lots of individual users who need reliable and robust connectivity that doesn't necessarily have to be individually fast. 802.11N in particular was optimized for the questionable use case of making your neighbors hate you so you can avoid running cat5 between your living room and den, instead of enabling lots of adjacent users to peacefully coexist. Hand-offs are still kludgy, and it's rare to find someone who doesn't do wireless networking for a living who can literally walk from the back yard down to the basement and upstairs, then out to the front yard without having his connectivity break (and break *hard*) at some point along the trip.

What do I propose? "802.11u" (as in, "UMTS"), running on unlicensed 5.8GHz. Stick private picocells in every room, hallway, and area. Multiple ones, in large areas. Wire them together with ethernet, or optionally create a peer to peer mesh network for the backhaul (5.8GHz where good links are possible, 2.4GHz for links between floors or "difficult" walls). Then take more or less canonical UMTS, and implement it as micro cell sites in devices with the approximate form factor of a smoke detector.

The nice thing about UMTS is that it's CDMA, just like 1xRTT (and ironically, unlike EVDO). You don't have to do local spectrum planning. If an area has poor throughput, just buy and add more femtocells or picocells to the congested area, let the devices negotiate lower power levels, and watch CDMA work its magic. By linking the PicoFemto/cells together, they'll magically work the same way a real cellular network does (sharing their bitstreams, and allowing them to reinforce each other and cancel out local/directional noise).

5.8GHz private UMTS wouldn't be fast (it would probably max out around 2.5mbit/sec at the cheap end, and max out around 10 or 20mbit/sec for a high-end corporate network where cost was no object), but it would be perfect for places like college campuses, offices, etc. where you have lots of people who need mainly internet access, but no single user necessarily needs GIGANTIC amounts of individual bandwidth. In other words, the exact places that are a total clusterfuck mess today, because you have hundreds or thousands of users stomping all over each other trying to use a wireless protocol optimized for streaming HD media from a point source to a single hungry consumer.

It might even be possible to make a cheaper "home" version that ran with uplink & downlink sprayed across the entire existing 2.4GHz wifi band -- the idea being that if Qualcomm has a chip designed to do 1900 & 2150MHz uplink/downlink UMTS, extending it to add ~2.4-2.5GHz would probably just be a straightforward next-gen upgrade that would be easy to add to any Android phone (or iPhone) that already had UMTS capabilities.

I have no idea whether this could possibly be cheap enough to justify doing at home, but I suspect the costs would be fairly competitive to what businesses, colleges, and hotels spend kludging around with high-end enterprise WiFi *anyway* today.

Re:5 GHz sucks

[Anonymous Coward]

Wow, I wonder what an orgy of 5GHz networks all fucking merrily away in NYC looks like?

Re:5 GHz sucks

[smash]

Exactly. Lower frequency travels through walls better. 5ghz only advantages are less contention for frequency as it is new, and your microwave doesn't interfere with it (as much if at all).

Re:5 GHz sucks

[smash]

5ghz doesn't go through solid objects as well. I too noticed better wifi signal when ugpgrading from a netgear AP to an airport EXPRESS running on 2.4ghz only. I put it down to the hardware simply being better - or wireless N (old router didn't support N) being better in general.

Re:5 GHz sucks

[adolf]

I like your idea, but you started off talking about home users in the first paragraph, and then launched into some CDMA fantasy that no home user will want to pay for -- let alone install.

For the common home user in the US (in a normal home, not a McMansion or a city apartment that is densely packed amongst others) the best answer to limited coverage is to install one or more higher-gain antennas at the (singular) access point. And then...well, that's all. There is no "and then."

There aren't any hand-off/roaming issues. There aren't multiple infrastructure devices to try to figure out how to configure cohesively. There is no need to install a myriad of difficult or unsightly (pick one) runs of Cat5. Nor any reason to actively avoid self-interference on one's own network.

Just buy a better antenna, plug it in, and call it done -- just like folks used to do with TVs before everyone had 57 channels and nothin' on. No magic required. (I've seen perfectly reasonable antennas on the shelf at Wal-Mart, even.)

The general case in other countries might be different. But here in the US, things are usually relatively spread out, there are generally at most just a few Wifi devices in a given home which are capable of consuming large amounts of bandwidth, and most of that bandwidth will be bottlenecked at the relatively slow connection to the Internet. 802.11's typical CSMA works very well for this scenario, especially since the vast majority of typical traffic is in exactly one direction: From the access point, to something else, perhaps with some TCP ACKs coming back the other way.

Local bandwidth isn't usually a big deal, unless the local geek is transferring huge files between systems or performing backups or whatever. And most homes don't have a local geek anyway. But if they do have a geek then 802.11a/b/g/n allows for such processes to be very speedy indeed on a one-to-one basis (compared to any other widely-used and inexpensive wireless method).

Meanwhile, in my own house: If I'd just spent a pile of money and effort refitting the house with a bunch of 802.11u CDMA femtocells, and the maximum bandwidth per device was only 2.5mbit/sec, my wife would kill me because her Netflix movies would turn ugly and her WOW machine would take even longer to patch on Tuesday.

We get far better performance than that right now (with our not-atypical usage) out of a single ancient WRT54G, and don't have to pay Qualcomm a dime in CDMA patent royalties. (I'd be also be very happy to jump onto the 5GHz 802.11n boat, but we currently own zero client devices which can use it. I -could- upgrade my laptop, but I'm happy enough to plug it in when I've got big things to transfer.)

That all said, I'm not quite so dense as to suppose that you're only talking about the common home use scenario. I understand that even a busy coffee shop might be sufficient to throw a monkey-wrench into the works. And certainly, I'd be pleased as hell if I could actually get a consistent 2.5mbit/sec connection at a coffee house (or a library or a university or...).

But please realize that while CDMA is very good for some things, it can be just as lousy for other things, and that most of CDMA's practical advantages in a limited environment can be realized about as well with CSMA, RTS/CTS, QoS, and per-device limits. (The fact that virtually zero consumer devices support 802.11u is a different topic altogether.)

Re:5 GHz sucks

[AmiMoJo]

"802.11u" (as in, "UMTS"), running on unlicensed 5.8GHz. Stick private picocells in every room, hallway, and area.

Better yet use an optical link. Confined to one room, no pollution outside it. Gigabits of bandwidth.

Re:5 GHz sucks

[Miamicanes]

Except then you'd render things like infrared remote controls and 90% of the cheap IP surveillance cameras from China unusable. Most cheap (sub-$100) wireless IP PTZ cameras either omit the IR filter, or use one that intentionally allows most near-IR (~880nM, I believe) light to pass through, and quite a bit of the adjacent spectrum as well, because the equally-cheap CMOS video sensors washed by IR interpret it as "white" (possibly with some software assistance). Ultraviolet light wouldn't fly, because it's too hard to focus and would be a lawsuit magnet. Even if it ended up being 100% safe, people are so neurotic about "UV", anyone who tried selling a UV-based solution would be sued into oblivion regardless of merit just from the sheer bulk quantity of opportunistic lawsuits.

Light has its place, but 5.8GHz is better. It has enough penetration to be useful, but not so much that you can't easily attenuate most of it away if that's what you'd prefer.

Proper LAN Design

[Deathnerd]

I attend a state funded technical college in Kentucky. We just had a technology audit/overhaul done over the summer break. Before summer, I could always connect to the wireless without a hitch. Sure it was a little slow, but it worked. After we cane back from the summer, I can't even get my wifi to associate with the router. Turns out, they redid the subnet ting and only allow 255 ip addresses to be leased at a time. 255 ip addresses. On a school network. Where everyone has a laptop and a smartphone. What the hell? I talked to the it guys and they said they're waiting to hear from the ISP so the can raise the number of leased up addresses. That doesn't make a whole lot of sense. Someone explain that one to me?

Re:Proper LAN Design

[Deathnerd]

Parent here. Sorry for the typos and lack of structure. iPhone's don't make for friendly slashdot commenting companions.

Re:Proper LAN Design

[Anonymous Coward]

Sounds like they moved from NAT to global addresses. while NAT is a hack, it does have valid uses, public wifi being one of them...

Re:Proper LAN Design

[Anonymous Coward]

I attend a state funded technical college in Kentucky. We just had a technology audit/overhaul done over the summer break. Before summer, I could always connect to the wireless without a hitch. Sure it was a little slow, but it worked. After we cane back from the summer, I can't even get my wifi to associate with the router. Turns out, they redid the subnet ting and only allow 255 ip addresses to be leased at a time. 255 ip addresses. On a school network. Where everyone has a laptop and a smartphone. What the hell? I talked to the it guys and they said they're waiting to hear from the ISP so the can raise the number of leased up addresses. That doesn't make a whole lot of sense. Someone explain that one to me?

If they need to talk to their ISP to up the leased address space, that would seem to imply they are using public space vs private space. That is just stupid and irresponsible. If they are using public space, they generally can only get issued a class C at a time if they have a large network requirement. What they should be using is the 10.0.0.0 space and NAT at the edge..

Re:Proper LAN Design

[sexconker]

Parent here. Sorry for the typos and lack of structure. iPhone's don't make for friendly slashdot commenting companions.

It's not your fault. And for once it's not even Steve's fault.
Slashdot doesn't make for friendly Slashdot commenting.

Re:Proper LAN Design

[thatskinnyguy]

WTF were they thinking? They could use a class A or B for more IPs or go so far as to use VLSM if they want a classless headache.

Re:Proper LAN Design

[Anonymous Coward]

Sounds like they budget cut their competent network admin, brought in the lowest bid consultant, and the lowest bid consultant made sure to leave a mess that would require a second visit.

Re:Proper LAN Design

[skids]

Either they have a small global address allocation to start with (and are trying to disable NAT), in which case they have to get more from the ISP or apply for more as an institution, or they decided to "outsource" their WiFi so that it is run as a turnkey system by an ISP, and that ISP did not allocate enough addresses for the need.

I could also see them wanting to segment WiFi if they for some reason decided not to turn off broadcast forwarding entirely (the saner thing to do if you have no mission-relevant installed application base that needs broadcast/multicast.) However, in that case the ISP wouldn't be getting involved, so it is likely one of the above two scenarios.

Re:Proper LAN Design

[NevarMore]

Go on, name names. http://www.kctcs.edu/Colleges_and_Campuses.aspx

Re:Proper LAN Design

[dasherjan]

There isn't enough info for us to know for sure. Assuming the parent poster is living in dorms. The deal with the ISP probably has the AP's VLAN'ing the SSID's back to their controller, and the university is probably just allowing the ISP to run their cabling on the uni's property. This is becoming a common setup for cash strapped organizations like colleges and hospitals since it cuts down on net admin costs. Well...in the short term anyway.

Re:Proper LAN Design

[Deathnerd]

Didn't wanna finger point but eh, alright. It's the cooper campus located in Lexington

Re:Proper LAN Design

[mcavic]

NAT may have been a hack to start with, but today it should be the default for any local network. Even if you need inbound access, you can route a public IP through to your private IP. Admittedly, that can cause problems where DNS lookups are concerned, and it makes VOIP more complicated. But the advantages for security and address conservation usually make it worthwhile.

Re:Proper LAN Design

[Anonymous Coward]

alot of smart phones are capable automatically joining unsecured Wifi, and smartphones are the way of the future, especially among early 20s students

Re:Proper LAN Design

[smash]

The only reason to use NAT is if you have no public IP space available and you KNOW that you never want to run services exposed to the internet. There is ZERO advantage for security and in fact if you are running IPSEC, NAT breaks things fairly badly, and NAT traversal to try and undo some of the brain damage is LESS SECURE than straight IPSEC without NAT-T. NAT is not a firewall and people need to stop believing it is one.

Re:Proper LAN Design

[smash]

And before it starts: yes, most people do not have public IP space available to run their entire lan on IPV4 public IP any more. Hence, your only option is NAT. This is not because NAT is a good idea if IPs are plentiful - IPv6 will fix this problem.

It shouldn't be slower.

[khasim]

The shorter range means that you won't have as many other access points conflicting with yours.

The only problem I've seen with 5GHz is that fewer end-user devices are supporting it.

Re:It shouldn't be slower.

[Alamais]

Yeah, I'm a little annoyed that my Nook doesn't support 5 GHz. It's basically the only reason I still keep a 2.4 GHz network running at home.

Re:It shouldn't be slower.

[skids]

The only problem I've seen with 5GHz is that fewer end-user devices are supporting it.

...even the ones that have 5GHz radios -- they will choose a 2.5 signal over a 5 even if they can get full speed over the 5, even if the 5 offers dual band 11n and the 2.5 does not. They will ride that crowded band into the ground when there is a perfectly usable 5 band right there for the using. The drivers don't seem to be able to distinguish correctly between a crowded band with a string signal and an uncrowded band with a weaker, but still fully capable signal.

So tip to geeks -- if you want speed, turn off your 2.5G radios when you are in range of a 5G AP. Even in cases where you can't get a 54Mbps signal, as long as you aren't down in the 1-4mbps range, you'll be one of like three people using the band so it will be faster for you.

Re:It shouldn't be slower.

[hawguy]

So tip to geeks -- if you want speed, turn off your 2.5G radios when you are in range of a 5G AP. Even in cases where you can't get a 54Mbps signal, as long as you aren't down in the 1-4mbps range, you'll be one of like three people using the band so it will be faster for you.

I split my SSID's across frequencies - I advertise one 802.11b/g SSID at 2.4Ghz and a different 802.11a SSID at 5Ghz, so I connect to the 5Ghz SSID with devices that can take advantage of it, and let the devices that can only handle 2.4Ghz use the other SSID. I regularly experience network drops at 2.4Ghz, but no problems at at all on 5Ghz. I can see about 30 - 40 AP's on 2.4Ghz from my apartment.

About the only thing I use Wireless for is to connect to the internet, so 802.11a works fine for me, I don't need the extra speed 802.11n would provide.

Re:It shouldn't be slower.

[Anonymous Coward]

You won't have any conflicting access points regardless, as the 5 GHz band has many, many more non-overlapping channels than the 2.4 GHz band's whopping 3 (or 1, if you're trying to use a 40 MHz WiFi-n channel).

Re:It shouldn't be slower.

[Tanktalus]

I had to completely disable the n support on my 2.4GHz band just to get my ASUS android table to connect. Which I thought was nuts. To be fair, I haven't tried again since upgrading from Android 3.0 to 3.1 and then 3.2.

I left the 5GHz band on (obviously n) for my laptop to use. Not much in the way of interference that way, but for the likelihood of collision, it's somewhat of a waste.

wi-fi needs an evolutionary upgrade

[sl4shd0rk]

It's one of the ubiquitous technologies which always seems to have something crippling it. Interference, noise, hardware compatibility, exploits, proprietary protocols. The 2.4Ghz problems will evolve the same way on the 5Ghz band. The only reason it's not a problem now is the reason stated in the article. Not that many devices are in the 5Ghz band.

Re:wi-fi needs an evolutionary upgrade

[skids]

The 2.4Ghz problems will evolve the same way on the 5Ghz band

Not really. 2.4G has 3 bands. 5G has, depending on the country has 4 to 8 times as many. Make that 2 to 4 when you turn on dual channel bonding, and then you are talking worst case 6 channels. That means that you can tesslate the 5G band much closer together, without APs that are on the same channel getting as close to each other (and no, dropping Tx power isn't a perfect solution to that problem on the 2.5G band.)

So it's much easier to microcell on the 5G frequencies, in fact AP density can be cranked up absurdly high.

Re:wi-fi needs an evolutionary upgrade

[dayton967]

First thing, because of the nature of the beast wifi is basically like going back to the old Ethernet Hubs, with the added benefit of it being Collision Avoidance, and not Collision Detection, but because it operates like a hub, you are only as fast as your slowest user.

There are big differences in 2.4Ghz and 5Ghz, first 2.4Ghz has only channels 1, 6, and 11 (in North America, elsewhere there is also 14) that can handle 20Mhz frequencies without interference, at 40Mhz there is 1 or 2, depending on location. The 5Ghz range has about 20 channels that have full 20Mhz ranges available (some countries have more, some less). This leads to a lot less crowding.

Secondly 2.4Ghz has a lot of other devices on this range causing interferences, such as cordless phones, leaky microwaves, video senders (and some of these use frequency hopping, with high power, but using 10Mhz), bluetooth, cordless keyboard and mice, and car alarms are ones off the top of my head, and I think there are some baby monitors in the range as well. 5.0Ghz does have anything in the range as far as I recall, with the exception of radar (in the US), but those ranges are not usable in the US.

The biggest issue with the 5.0Ghz is the penetration power, it may not pass through concrete walls or water, as well as 2.4Ghz. But I really don't plan to use an ipad while I am underwater in a pool.

Re:wi-fi needs an evolutionary upgrade

[ridgecritter]

That 5 GHz is actually 5.8 GHz, and there are lots of cordless phones in that freq range. I have three of those in my home, and I have wireless digital headphones (Amphony) that live on the same freq but use a proprietary protocol. I don't think my 5.8GHz wifi gets impaired by the cordless phones or the headphones, but the headphones get clobbered by the wifi. So far, I am the only detectable person on 5.8GHz wifi in my neighborhood. *smug grin* as I look at AirRadar's readout on the high local usage density on 2.4 GHz...

Re:wi-fi needs an evolutionary upgrade

[dayton967]

No the 5Ghz wifi range actually starts around 5.18Ghz, and 4.915Ghz in other countries. Even taking out the top end of the range from 5.7Ghz and up that 15 channels.

channel frequency(MHz)
36 5180
40 5200
44 5220
48 5240
52 5260
56 5280
60 5300
64 5320
100 5500
104 5520
108 5540
112 5560
116 5580
132 5660
136 5680
140 5700
149 5745
153 5765
157 5785
161 5805
165 5825

Re:wi-fi needs an evolutionary upgrade

[Anonymous Coward]

Thanks for this, I was misinformed.

Re:wi-fi needs an evolutionary upgrade

[EvilJoker]

First thing, because of the nature of the beast wifi is basically like going back to the old Ethernet Hubs, with the added benefit of it being Collision Avoidance, and not Collision Detection, but because it operates like a hub, you are only as fast as your slowest user.

CSMA/CA is not an advantage. It takes A LOT more overhead to avoid a collision than to detect it. However, on wireless networks, detection is problematic; it's more of a workaround than an improvement.

Unreliable sources:
CSMA/CD CSMA/CA Comparison

Re:wi-fi needs an evolutionary upgrade

[dayton967]

In the case of WIFI it is an advantage. In CSMA/CD you attempt to send your frame after waiting for the line to be free, and if a collision occurs, everyone on the network backs off, and you must resend the complete frame over again. With CSMA/CA, you send a smaller packet stating your intention to send, if there is a collision there, then you back off, and try again. When a shared access network becomes heavily loaded, with collisions in CD, it almost always causes no end of problems.

This is generally not an issue with switched Ethernet, unless there is a speed and duplex misconfiguration, which will happen either if one side of a link is forced and the other isn't, or some network cards and switch combinations can have problems negotiating on occassion, but it's becoming less and less common. (unless you are using the old Baystack 450's)

Wireless N doesn't imply 5GHz

[watermark]

I find it largely annoying that "Wireless N" doesn't imply support for 5GHz. Many "Wireless N" devices only support 2.4GHz and most are bad at labeling whether they support 5GHz or not. It makes it difficult if you're looking for devices that support 5GHz "Wireless N".

From a cost perspective, I understand why they might only support 2.4GHz. I just wish they called it something else, like "Wireless NS" or something.

Re:Wireless N doesn't imply 5GHz

[Alamais]

Never happen. Who wants to label their device as lacking a feature?

Re:Wireless N doesn't imply 5GHz

[watermark]

You're right in that you can't expect the vendors to do it themselves. It's going to have to be baked into the spec (ie. they can't call it wireless N unless it supports the following features). Obviously too late for Wireless N, but a point for future specs.

Re:Wireless N doesn't imply 5GHz

[msauve]

"most are bad at labeling whether they support 5GHz or not."

In my experience, it's the "not" that's the problem. If a device supports 5 GHz, it will almost always say so (usually labeled "dual band"), because it's a marketable feature. If they don't support that band, they say nothing.

Re:Wireless N doesn't imply 5GHz

[skids]

...which is why you look for products labeled "abgn" since you'd probably be hard pressed to find a product that supported both a and n but didn't support n on the 5G radio.

Re:Wireless N doesn't imply 5GHz

[Mia'cova]

But even then, some require manual configuration to pick a band, others can switch on-the-fly, and others can use both 2.4 and 5 at the same time. So even among devices which do support 5 GHz, there's a range of support.

And it is getting worse

[chill]

Look at the number of smartphones, e-readers, laptops and Android tablets out there that don't support 5 GHz. With the premium price of some of these gadgets I'm surprised vendors are trying to shave expenses by getting 2.4 GHz-only 802.11n chipsets.

If you get an opportunity, let the vendor or salesman know you one of the features you look for is 5 GHz capability. Make a point of it.

Re:And it is getting worse

[Alamais]

I know, I've been pondering a tablet, but it's hard to tell what even supports 5 GHz. Do any of the current 10"-class tablets?

Re:And it is getting worse

[chill]

Motorola Xoom is the only one, I believe.

Re:And it is getting worse

[Anonymous Coward]

Cheap devices suffer the worst. Roku 2 & Kindle Fire don't support 5GHz for example. Both companies make efforts to hide this information.

Re:And it is getting worse

[93 Escort Wagon]

I know, I've been pondering a tablet, but it's hard to tell what even supports 5 GHz. Do any of the current 10"-class tablets?

The iPad does. My wife's iPad can connect to either our 802.11b/g 2.4GHz network or our 802.11n 5GHz network.

I'd like to kill off the 2.4GHz network, but we've got too many devices that still need it - including my new-ish Android phone.

Re:And it is getting worse

[JamesP]

Then on the next couple of years we get congestion at the 5GHz band

It's not about band, it's about intelligently using the resources. Unfortunately, 802.11g is very bad at this, 802.11n is better, still, I think there's a long way to go still...

Re:And it is getting worse

[adolf]

I've had an Intel 802.11a/b/g wireless card in my laptop for well over half a decade, and have never encountered an access point in the 5GHz band that I want to use when out and about. Nobody, in my not-so-limited experience, operates a public 5GHz hotspot.

To be clear: I do use the various 5GHz ISM bands for all kinds of long-range communication in my day job, and really appreciate the available spectrum and general lack of interference as compared with the near-universal mess that exists at 2.4GHz.

But as a consumer, I'd have been better off getting the cheaper 2.4GHz-only adapter and spending the difference in price on a bottle of mid-grade tequila.

Re:And it is getting worse

[Anonymous Coward]

It's almost impossible to congest the 5GHz band without resorting to high-power equipment. To achieve the same levels of congestion in a large area you need almost twice the number of APs.

Re:And it is getting worse

[chill]

Congestion at the 5 GHz band will be harder to accomplish. The 2.4 GHz band has only 3 non-overlapping channels for use in the United States. The 5 GHz band has 21 non-overlapping, 20 MHz-wide channels.

The 2.4 GHz band also has to deal with Bluetooth and Zigbee traffic, in addition to additional background noise created by microwave ovens and fluorescent lights.

Finally, signal transmit distance in the 5 GHz band is shorter than 2.4 GHz. This means in low-to-moderate density areas there won't be as much overlap from neighboring stations. For example, my computer can detect my neighbor's 2.4 GHz Wifi base-station from inside my house. I had to adjust the channel of my unit to not contest. However, I can't detect his 5 GHz signal unless I leave my house and cross half the yard. Thus, it doesn't interfere with 99% of my usage.

Re:And it is getting worse

[chill]

Out and about? I agree. At the office or home, however, I prefer the 5 GHz band so as to not contend with all the clueless noobs and their cable box/wifi hotspot crushing the 2.4 GHz spectrum.

Re:And it is getting worse

[Tomato42]

it has 3 non-overlapping channels everywhere except Japan, where it has 3 and a half non-overlapping channels...

Re:And it is getting worse

[m50d]

I think you have that backwards; you get 1-11 (i.e. 1, 6, 11 for non-overlapping) in Japan, and 1-13 everywhere else. No?

Re:And it is getting worse

[tlhIngan]

I know, I've been pondering a tablet, but it's hard to tell what even supports 5 GHz. Do any of the current 10"-class tablets?

Easy. Look for 802.11a support.

All good tablets will have one of two labels for the WiFi - 802.11bgn, or 802.11abgn. (If the tablet doesn't have 802.11n support, back away from it and pick another one).

802.11a only works on 5GHz. If you don't see it on the list, then it's 2.4GHz only (including N). If you see a, then it's 2.4 GHz (b/g/n) and 5GHz(a/n).

Smartphones I can understand not having 5GHz - it takes another antenna and more power because you have to drive a 5GHz amplifier and all that. Tablets, not so much, especially since they should be following the iPad example and supporting 5GHz out of the box.

Re:And it is getting worse

[Tomato42]

No, Japan = 14 channels, EU = 13 channels, US = 11 channels, all starting at 2412MHz.

Japan has also extra 5GHz channels.

http://en.wikipedia.org/wiki/List_of_WLAN_channels

Re:And it is getting worse

[ridgecritter]

Just curious - assuming you've got an Apple wifi - Does the dual-band Airport Extreme Base station, which has 2.4GHz and 5.8GHz chipsets, alleviate the 2.4GHz blues?

Re:And it is getting worse

[93 Escort Wagon]

Just curious - assuming you've got an Apple wifi - Does the dual-band Airport Extreme Base station, which has 2.4GHz and 5.8GHz chipsets, alleviate the 2.4GHz blues?

My 802.11n Airport Extreme was the earlier, single-band version - so I've got that handling 5GHz and an older Extreme (the one that looks like a white Hershey's kiss) taking care of 2.4GHz. But I can't imagine a dual-band Extreme wouldn't still be subject to the fundamental limitations of the 2.4GHz band.

Re:And it is getting worse

[adolf]

I live in a big house which, while definitely in-town by any measure, sits on a fair bit of land. There are something like 16 access points shown on my Droid while standing in my driveway...but four of them are mine. (I've currently got two access points that I actually use, just because I can. The PS3 also becomes an access point when it's turned off for Remote Play(tm). And I use AT&T's provided access point only to broadcast obscure and sometimes dirty SSID beacons, again just because I can...)

Of the remaining twelve, two are open. Their signal is so weak inside of my house that I cannot connect to and use them without extraordinary means.

One of them is an illegally-amplified access point operated by the local municipality, located about five miles away. (I really should get around to complaining to the FCC about that.)

Almost all of the rest of them are on one channel, so I use the other two non-overlapping channels.

The rest of them just constitute a bit of noise, which seems OK: My 802.11g is as fast as any practical benchmark or testing arrangement says that it should be, and my portable devices don't seem to suffer undue battery drain when I leave their WiFi on. In terms of signal-to-noise, I think it's working fine.

And so, while I don't see much detrimental interference in the 5GHz bands when using a spectrum analyzer around here, I also simply don't experience any at 2.4GHz as a consumer even though I can see that many of my neighbors are sharing that spectrum with me.

And, of course, just because you see a plurality SSID broadcast doesn't mean that they're actually using any meaningful chunk of available wireless bandwidth. Most folks are just the usual occasional Youtube watcher who otherwise does light web browsing, and that doesn't generally have much effect on things. 802.11 is pretty good about being fairly quiet when it doesn't have actual data packets to transmit, making it a rather good neighbor (unlike some other things such as Motorola Canopy).

(YMMV, of course. I'm just one data point.)

Okay that is so odd.

[LWATCDR]

There are only around 5.000 students at ACU based on their website. Even if their 5000 faculty and employees you are averaging 47 logins per person.
Does half the town of Abilene use the university as a hot spot or what. Even with an average of two devices per person that seems like a lot of users.
of course you could always throttle which ever band gets say more than 70% of the traffic.

Re:Okay that is so odd.

[Anonymous Coward]

Many smartphones are set to autoconnect to the nearest open WiFi network. Perhaps people driving by the school are within distance of the signal?

Re:Okay that is so odd.

[LWATCDR]

Only if they are open. If so then that maybe an a solution in part. Password protect the WAPs that cover the road way.

Re:Okay that is so odd.

[Pharmboy]

Keep in mind that Abilene has 117k residents of all ages (2010 census). That sounds like a whole lot of connections for my former home town. Plus they are on the NE side of town, not near most of the shopping or traffic, which is on the south side, so incidental traffic can't the be cause. If you discount infants and old timers, you are left with a total possible universe of 60k-80k people that would be in the right ages to even USE the internet, half which likely don't, with 95% of the population never getting close enough to ACU to get a connect, and...well, you can see where this is going.

I can see 470,000 http requests, but 470k actual wifi connects (devices connected for X minutes) seems impossible for that location. If there really was 470k connects, even on an open network, something is very, very wrong. Time for someone to pour through the logs.

Re:Okay that is so odd.

[BLToday]

It was a very poorly written article. But the 470,000 connections on the single day probably means instances of connection. If I turn on my laptop (1 connection), put it to sleep, turn it back on (2 connections), move to a different AP (3 connections)....etc. Think of how many people have their computers set to go to sleep after 15 minutes of inactivity. And smartphones, some only turn on the WiFi when screen is on or there's something actively going on. In a single day, I probably turn on and check my smartphone (20 - 30 times a day), that's 20 to 30 instances of connections.

Re:Okay that is so odd.

[jdkramar]

I think it's there were 470,000 connections to the wifi network, so everytime a device disconnected from one wireless router and connected to the next router gets counted. So each time all the students with their iPhones on walk across campus they probably get counted 5-10 times.

Re:Okay that is so odd.

[egamma]

There are only around 5.000 students at ACU based on their website. Even if their 5000 faculty and employees you are averaging 47 logins per person. Does half the town of Abilene use the university as a hot spot or what. Even with an average of two devices per person that seems like a lot of users. of course you could always throttle which ever band gets say more than 70% of the traffic.

It says 470,000 connections PER DAY, divided by 5000, is 94. So if you go from your dorm room (1) to the business building (2) to the bible building (3) to the library (4) to chapel in the coliseum (5) to english (6) and back to your dorm room (7), your one device just made 7 connections. If you went back to your dorm room in between each of those events, it would be 20. Now, consider that there are multiple floors and multiple access points per building, and it adds up very quickly. Imagine if someone went jogging with their iphone wireless enabled--you could make a new connection every 10 seconds.

Re:Okay that is so odd.

[skids]

Likely one of two things here:
1) they are (defensibly) counting all roams from one AP to the next as their own connection or
2) they aren't exclusively running WPA (e.g. they are doing open +IPSEC) so every jerk who drives by the campus with his gadgets turned on is ending up in their connection table, even if they cannot use the connection. Which is also defensible, because all those negotiations do use the band and also the infrastructure resources.

Re:Okay that is so odd.

[jdkramar]

He got 47 per user by assuming there are 5000 faculty and staff members also connecting to the network. 5000 students + 5000 faculty/staff

Re:Okay that is so odd.

[jdkramar]

In response to #2, it is an open network. But that alone wouldn't be enough to get that number, so it is likely a combination of 1 and 2.

Re:Okay that is so odd.

[egamma]

He got 47 per user by assuming there are 5000 faculty and staff members also connecting to the network. 5000 students + 5000 faculty/staff

That's not what I was disputing--there's about 1000 faculty/staff, by the way--my main point was that LWATCDR confused "connections" and "devices" in his counting.

Re:Okay that is so odd.

[skids]

When I first got to this place we had an open SSID. We had 8 times as many "connections" present on the controllers at any given time back then. Since most of those were short-lived drive-bys I could easily see us reaching that number. We have about the same size population, but then, we are in a slightly more urban setting.

Re:Okay that is so odd.

[LWATCDR]

No I do understand but even with your going back to your dorm room you only hit 20 which is half of my 47 total. That would be a worse case. Jogging while having a smartphone in your pocket shouldn't case that many hits since it should be asleep for most of the time one is jogging. Still just seems really high to me. Plus does every student have a smart phone, tablet and Notebook at ACU these days.

It has been a while since I was in college and I never went to ACU. I knew a very pretty girl that went there... I have to say that she was interesting...

Re:Okay that is so odd.

[egamma]

No I do understand but even with your going back to your dorm room you only hit 20 which is half of my 47 total. That would be a worse case. Jogging while having a smartphone in your pocket shouldn't case that many hits since it should be asleep for most of the time one is jogging. Still just seems really high to me. Plus does every student have a smart phone, tablet and Notebook at ACU these days.

It has been a while since I was in college and I never went to ACU. I knew a very pretty girl that went there... I have to say that she was interesting...

I'm a former ACU student--and former student worker in their networking services (2006). Right now, every student at ACU has an iPod or iTouch, plus their personal computer. ACU is a Google and Apple technology partner or learning center or something.

Re:Okay that is so odd.

[LWATCDR]

Really wow things must have changed. When my friend went there it was the one of the most uptight strict place I had ever heard of. My friend would go out town so she could go to dance clubs. Remember I am old so this was a good bit ago.
Frankly she was really sweet but she had a kind of sad in the way she seemed to have faith but would only obey the word of the rules she said she believed in but not the spirt. I have not heard from her for a very long time funny how how a story can make you think about old friends.

Re:Okay that is so odd.

[egamma]

Really wow things must have changed. When my friend went there it was the one of the most uptight strict place I had ever heard of. My friend would go out town so she could go to dance clubs. Remember I am old so this was a good bit ago. Frankly she was really sweet but she had a kind of sad in the way she seemed to have faith but would only obey the word of the rules she said she believed in but not the spirt. I have not heard from her for a very long time funny how how a story can make you think about old friends.

The rules have indeed changed--students over 21 can drink off-campus. The new swimming pool is coed. And those are just changes since I graduated in 2006!

Re:Okay that is so odd.

[thejynxed]

It's not just that, the radio beacon will pulse on some sort of software or hardware-based (controlled by firmware) timer to locate the nearest cell radio node. It does this even if you have it turned "off".

They could easily hit those numbers with all of the cell phone/tablet devices just doing an "are you alive and where am I" connection to the nearest nodes.

Couple that with the built-in GPS in today's devices, and you'll have even more connections, even if you aren't actively using them.

Re:Okay that is so odd.

[LWATCDR]

Dang it almost sounds like a normal school now. Actually probably for the best IMHO. Just from observing my friend and her friends I get the feeling that excessively strict rules leads to less respect for all the rules. AKA following the word and not spirt but that is just my opinion.

NS or N5?

[tepples]

"NS" for 2.4 GHz-only would just be confusing. It'd look like "N5", which one might assume means 5 GHz.

Re:NS or N5?

[drinkypoo]

What would have not been confusing would have been to just not call anything that will only do 2.4 "N". But noooo....

I'm not sure what's new here

[sdavid]

I'm always surprised how often this is described as a 'new' problem. I have a home network in a very highly populated area, I can see over fifty networks from my apartment, and I switched to a dual-band router as soon as they came on the market. I would have thought that planning a university network to work on both bands would have been on the radar for a number of years.

Re:I'm not sure what's new here

[AngryDeuce]

I had the same problem in my apartment, there are between 30-50 networks (both residential and business) in range at all times crowding bandwidth, and to make matters even worse, it seems like a lot of them have that "smart switching" technology where they automatically go looking for the least crowded channel. The problem is that there are so many of those routers here that it's like they chase each other up and down the band all day long. I had no idea how bad the problem was until I downloaded a wifi app for my phone and watched it happening in real time; blocks of networks moving together because they all see the same "clean" channel...which they then immediately saturate so they all move to the next "clean" channel, totally unaware that the other networks are moving right along with them. It's maddening...

Going 5GHz 802.11N only was not a viable solution because so many of our devices require the 2.4 GHz band so I had to go out and buy 100' long cat6 cables and hard line every desktop in my apartment. It's hideous and I can't stand having cables run up walls and across the ceiling but it was either that, spend every waking moment of my day stuck in my living room, or not have internet.

There's got to be a better way.

Anybody got a suggested reference?

[Anonymous Coward]

Anybody got a good, concise reference recommendation for how to set up and administer a wireless network? I'm not as up on this side of networking as I should be, but I have no idea where to start reading.

Thanks in advance.

Re:Anybody got a suggested reference?

[k6mfw]

I have asked this same question. I get two responses: Either "google it" (a bankrupt answer as all stuff I find online is really simplistic or just aggregetes of vendor sites trying to sell crap) or a long discussion from "a expert" of all kinds of things about wireless stuff (but never telling me what I need to know).

Other day at a group of companies displaying tools to prevent network and computer virus/malware/etc. I asked one of them a book he would recommend on networks. He said enroll in community college class. It's low cost, you get experience, you get assistance, you can ask questions.

Re:Anybody got a suggested reference?

[cheros]

Tell me what scale you're looking at and I'll help you . maybe I'll write the HOWTO in the process because you're right, there isn't anything. Actually, I may put it in my book..

It's not that hard, but you need a little bit of base knowledge, mostly about TCP/IP based networking in general. I can steer you to an old Linux HOWTO that is actually quite good to give you the basics. WiFi/wireless setup is about two things: replacing the cable with a radio link (and all that comes with it such as reach, signal strength etc), and setting up a network (because that's what you're probably wanting to set up WiFi for).

So, TCP/IP, a bit about DNS and DHCP, and then knowing about radio. When you do this on a bigger scale you'll extend that set with considerations about repeaters (sometimes it makes more sense to set up separate cells as most devices now switch automagically).

So there.

i know where not to look...

[Anonymous Coward]

not that school

Channel 14

[snsh]

Anyone knowledgeable about the conflict is 2.5GHZ that led the US FCC to limit wifi from using channel 14 (2.484 GHz)?

According to the FCC spectrum chart the top of the 2.4 wifi band abuts the "Standard Frequency and Time Signal" Band at 2.5 GHz. What is that used for?

Re:Channel 14

[jonbryce]

Radio controlled watches and clocks use that to set the time. My watch checks that signal every day at 2am and adjusts itself to the correct time.

Re:Channel 14

[oneiros27]

I assume that would be WWV :

http://en.wikipedia.org/wiki/WWV_(radio_station)

Re:Channel 14

[egamma]

Anyone knowledgeable about the conflict is 2.5GHZ that led the US FCC to limit wifi from using channel 14 (2.484 GHz)?

According to the FCC spectrum chart the top of the 2.4 wifi band abuts the "Standard Frequency and Time Signal" Band at 2.5 GHz. What is that used for?

Let me Google that for you

Re:Channel 14

[Anonymous Coward]

But how does it know that it's 2am unless it already knows the correct time?

Re:Channel 14

[Anonymous Coward]

Actually, I think you were looking at the wrong part of the chart. There is no "Standard Frequency and Time Signal" band at 2.5GHz, but there is one at 2.5MHz.
Clocks and watches set themselves to WWVB at 60kHz and WWV and WWVH are both on shortwave on 2.5MHz, 5MHz, 10MHz, 15MHz, and 20MHz (only WWVH uses 20MHz); none of which are near 2.5GHz (2,500MHz).

Re:Channel 14

[jonbryce]

It does it when it thinks it is 2am

Re:Channel 14

[vlm]

Anyone knowledgeable about the conflict is 2.5GHZ that led the US FCC to limit wifi from using channel 14 (2.484 GHz)?

According to the FCC spectrum chart the top of the 2.4 wifi band abuts the "Standard Frequency and Time Signal" Band at 2.5 GHz. What is that used for?

You're a factor of 1000 too high for WWV.

The problem with channel 14 is if it were used it would pretty much wipe out the BRS/MMDS service right above the wifi band.

http://en.wikipedia.org/wiki/Multichannel_Multipoint_Distribution_Service

MMDS never really went anywhere, which is a shame. For at least 30 years some areas have had some MMDS gear; my local school district linked the schools together in the 80s. Back when a decent pro-grade VCR cost $2500 a $1000 MMDS link between schools to share the VCR sounds like a good idea.

You'd be crazy to set up a MMDS system now, with the wifi wanna be hackers trying to use channel 14 to get away from the noise and some microwave oven interference. So that chunk of bandwidth is kind of a wasteland that no one can use, more or less.

Advanced AV stuff like that was kind of the "ipad of the 80s" where merely spending dough on silicon would magically make the kids smarter, or something.

Re:Channel 14

[ericloewe]

The signal is always transmitted, but it would be a huge drain on the battery to keep it synched at all times, so most watches sync every night, so even if the watch is off by hours, it'll sync.

Re:Channel 14

[snsh]

Right, I was looking at MHz not GHz.

So it looks like there's some sort of radio-satellite and mobile-satellite communication going on at 2.85GHz. What's that used for?

The underlying question is, if everyone starts transmitting low power wifi over channel 14, what's it going to break? If it means somebody's satellite phone stops working, then I don't really care because I highly doubt my neighbors are walking around the neighborhood chatting on satellite phones.

Re:Channel 14

[adolf]

You'd be crazy to set up a MMDS system now, with the wifi wanna be hackers trying to use channel 14 to get away from the noise and some microwave oven interference. So that chunk of bandwidth is kind of a wasteland that no one can use, more or less.

Watch TV has used it for a long time for subscription TV service, and it's still a popular (though regional) alternative to cable or satellite. They're located not too far from me, and it's fairly typical to see their telltale parabolic reflectors on small towers near farmhouses and on multi-story commercial buildings.

In areas served by Watch TV, I have every expectation that 802.11 channel 14 is an unmitigatable sea of TV signals, and that any wifi wanna be hacker would quickly abandon the attempt except for at very short distances.

Re:Channel 14

[bzzfzz]

It's part of the MMDS spectrum, which in many communities has been relicensed for wireless data purposes using either 802.16e (WiMAX) or older proprietary systems from Alvarion or Nextnet.

Re:Channel 14

[Anonymous Coward]

2 am? So... if there is a problem with syncronization, it may set the clock completely off and the alarm won't wake you.

A better design would be to check, say, 2 pm, and then you would notice the drift while being still awake. Unless the clock's internal drift is so high it really must synchronize once every day.

Controller-based Wifi Network

[MoldySpore]

This might be troublesome to places that run stand-alone AP's, but anyone who runs a controller-based wifi network knows this isn't an issue at all, considering how easy it would be to create a new SSID on a 5Ghz band and push it out to all APs simultaneously. We run 5508 Cisco controllers, where I work, that support between 500 and 1000 devices connecting at any given time (only about 200 APs between the two controllers). For us to put out a new 5Ghz 802.11a or 802.11n-based SSID would take all of 5 minutes from creation to people using it live. The problem is, as the article suggest, the lack of devices that support 5Ghz. He were going to turn on WPA2 and 802.11n @ 5Ghz at the beginning of this year, until we found out all the phones and laptops everyone uses where we work (several thousand people) don't support WPA2 or 5Ghz. Have to wait until the end of the refresh to turn on all that. I have a feeling this is the case for a lot of other companies too, especially those tied to state and local government IT refresh cycles.

Re:Controller-based Wifi Network

[sigipickl]

Why wait to turn it on? You're right, it does only take 5 minutes. Even if only 5 of your clients connect @ 5GHz, you have 5 users that may see an improvement and may become advocates for you and what you are doing.

Re:Controller-based Wifi Network

[DarkOx]

No WPA2? How old is the kit you have most people using? like 7+ years?

Re:Controller-based Wifi Network

[Tomato42]

This, my N900 does support WPA2!

Re:Controller-based Wifi Network

[Anonymous Coward]

Hospital sites are loads of fun. Medical device vendors buy the cheapest old WiFi chipset to make their products wireless, and medical staff buy into it hook, line and sinker. It seems that we will be forced to keep our 802.11b WEP SSID up forever.

Then there's the vendors that throw the cheapest integrated USB dongle adapter on their device, inside their metal cabinet. Gee, I wonder why the signal is poor? It's FDA certified, you can't change it.

iPhone 4S is a huge disappointment in this regard.

[qubex]

I for one was hugely disappointed that Apple’s latest iPhone refresh (iPhone 4S) did not address 5GHz WiFi connectivity. The iPhone 4 has supported 802.11n since it was released last year, but, unlike the iPad, it does not support the 5GHz band, constraining use to the already oversubscribed 2.4 GHz band.

The end-result is silly: for example, I’m running a 5GHz 802.11n network for all my devices at home, but I’m broadcasting an extra 2.4GHz signal for the sole benefit of my iPhone 4. I hoped Apple would address this obvious shortcoming but obviously they didn’t. Sad.

Re:iPhone 4S is a huge disappointment in this rega

[MoldySpore]

The iPhone 4S is a huge disappointment in several other regards too.

Re:iPhone 4S is a huge disappointment in this rega

[qubex]

Granted.

The greatest disappointment of all? That they somehow engineered a desire for it into me.

Re:Christian "University"??

[Tyrannosaur]

yeah you're an idiot

Re:iPhone 4S is a huge disappointment in this rega

[johnmat]

Smartphone designers struggle with the number of antennas required; and 5 GHz implies an additional antenna. They already have antennas for the cellular network, GPS, increasingly NFC, sometimes FM, WiFi and Bluetooth. The latter two are sometimes combined, but still, that is a lot to fit in a package that is very space constrained.

Band Steering

[sigipickl]

I don't know if apple fixed their wireless driver in IOS 5, but I have found that the iPad running IOS 4 does not 'steer' to 5GHz when presented with the same SSID on 2.4GHz and 5GHz. This has been a consistent experience using Cisco, HP (E-series), and Ruckus wireless networks. With some of my customers, we have had to create different SSIDs for the bands to get their fleets of iPads off 2.4GHz.

Re:Christian "University"??

[jdkramar]

If I had mod points, I'd mod you up for this.

WiFi was *never* intended for....

[Anonymous Coward]

...very large scale, concentrated deployments. The limited amount of RF spectrum available, combined with CSMA/CA means that WiFi was only ever intended for small, localized, sparse and isolated deployments, for network communication "convenience" and augmenting a primary wired network for a small number of portable devices... not for being the primary method of business network connectivity itself. Anything else is a losing battle against a very unforgiving conglomeration of compromises and tradeoffs.

802.11a

[xororand]

Look for 802.11a support. It requires the 5 GHz band.

Re:802.11a

[fostware]

Oh for a +1

802.11a (or 802.11abgn) *is* something that will be on the outside of a box, giving some certainty.

Except the cheap-ass Broadcom chips Samsung use top out at 72mbps . Yeah, it supports N, but it's only to tick a box.

Re:Channel 14 not really

[Anonymous Coward]

Me thinks your off by 3 orders of magnitude WWV @2.5 MHz megaHertz does not abut channel 14 @ 2.5 Ghz gigaHertz

Re:Dangerous.

[KingMotley]

Tinfoil hats solve that problem.

Two words:

[Anonymous Coward]

Baylor University.

You really are a special kind of moron if you think "Christian" and "University" are diametrically opposed... as if you have some warped perception that all of academia must somehow be atheist.

Have they tried renaming their networks?

[Anonymous Coward]

Have they tried to use different SSID's for those different channels? On a few events where I'm part of the organizers the network guys use -SLOW and -FAST suffixes for those 2.4 and 5 GHz bands and get about 50/50 usage between those bands. Which is good as software that makes those Wifi connections usually doesn't try to prioritize those 5 GHz bands.

Re:iPhone 4S is a huge disappointment in this rega

[sigipickl]

Well that AND battery life.

Re:iPhone 4S is a huge disappointment in this rega

[Marcika]

An HTC Desire?

ironically the HP Touchpad supports 5GHz

[Chirs]

I was fairly impressed when the touchpad detected my 5GHz network. Great buy for $99, will be even better when they port ICS to it.

WiFi cabled to switch-router ....

[OldHawk777]

WiFi cabled to switch-router, how about they get a better network designer.

If that many WiFi spots are cabled to the same nodal switch, then that is just one problem of many.

Has any one tried a flood-ping at any one of the WiFi spots, I suspect, it would be an avoidable single/few point DOS attack with a good network design.

Re:WiFi cabled to switch-router ....

[AmiMoJo]

Try using a de-auth attack. Makes it impossible for clients to maintain a connection. Leave that running for a few weeks and it will clear our that part of the spectrum.

Obviously don't do that, it's evil.

Re:Dangerous.

[geekmux]

These radio radiation signals are not good for us. Sure ethernet isn't as convenient - but isn't it more convenient than getting cancer or having changes of mood and though processes? I don't have a link handy but they have proven that high levels of EMF radiation can cause such problems with people. We are not designed for such high levels. They are dangerous.

Which is why it is an FCC-controlled/regulated device that must meet certain class requirements for transmit and receive levels. Sure, one of the more common "hacks" with 3rd party firmware like DD-WRT is to boost those signals, but there's still a legal limit, for both broadcasting and safety reasons.

Not doubting your claims here, just saying we don't exactly burn holes in the sky with Wi-Fi devices. And trying to avoid excessive radio signals and EMF in public these days is like trying to avoid oxygen in the air you breathe. Good luck.

why why why 2.4GHz!!!!!

[k6mfw]

I'm mainly on my Gripe Of The Month but I sure wish there was some little sliver of spectrum besides this for us commoners. I guess all the Corporation$ have hijacked the good stuff. Though there is white space here and there, though need to either spend lotsa bucks on RF equipment or design/build your own (last option time consuming).

Re:why why why 2.4GHz!!!!!

[Glendale2x]

Are you not aware of the 5GHz unlicensed bands?

Re:why why why 2.4GHz!!!!!

[LinuxIsGarbage]

Are you not aware of the 5GHz unlicensed bands?

I'm hanging onto my 900Mhz cordless phone for dear life. I can talk on the phone while surfing the net and using the microwave! Such advanced technology.

Re:iPhone 4S is a huge disappointment in this rega

[cthulhu11]

The end-result is silly: for example, I’m running a 5GHz 802.11n network for all my devices at home

For "all my devices"? How many things do you have that require WiFi? How many of them would be better connected via copper? The first thing that jumped out at me in the above summary was to ask why the hell desktops are using WiFi.

Re:iPhone 4S is a huge disappointment in this rega

[qubex]

I have a desktop, a server, a NAS, an audio streamer connected to my stereo, a pair of AppleTVs, and a pair of Ethernet-sporting flat-panel LCDs all wired with 1000baseT ethernet.

However, wherever possible, I’ve also configured the same devices to use WiFi if for some reason the cable comes out, so that, for example, the video stream does not fail if somebody trips over the AppleTV’s ethernet connection and yanks it out of the jack.

Then there’s the incriminated iPhone4, an iPad2 and a MacBookAir exclusively on WiFi.

Shhhhhh!

[Anonymous Coward]

You're ruining a good thing for us in 5GHz!