neilmac

Wow, I stand corrected. That's a great find. I still see "b" devices in industrial settings, but I haven't seen these before, and definitely not anything like this is a school. I need to get me on of these for training classes. This is exactly the thing that will significantly affect throughput. These will definitely be a problem. Any one of these connected is going to put a major portion of the network into protected mode. I couldn't find a certificate for them from the WiFi alliance, so there's no way to know if they even operate correctly. The saving grace is that they probably transmit really low power levels, but the damage is the same as the protection triggers when it associates to an access point. If you must have these, then get as many devices as you can into 5 GHz, and reduce power in 2.4 GHZ, turn off the 2.4 radios on access points where you don't need it. Wow... If you rely on good throughput for your network, with these dotted around, you're gonna have a bad time.

I haven't seen a "b" only capable device in a long time, so I doubt anyone is really using them. It's just not relevant any more. Does anyone have anything like that ? I would be uber surprised.

Much easier, just tell me the make/model of devices you use and I can show you.

Sorry, you're right - I didn't mean to be so short but I am struggling for time, I'm in the middle of something. By asking about devices, if you cam back with an answer then I would have been able to answer by referring to your device, it would have made the explanation easier. Basically, you won't find client devices that support 40MHz in 2.4 GHz, and many access points either won't support 40 MHz or won't enable it if they detect ANY non 40 MHz transmissions. It's all to do with co-existance, and the realisation that if you put up 40MHz you will probably interfere with someone, somewhere, so 40 MHz is just not deployable in 2.4 GHz. If you want 40 MHz you MUSt go to 5 GHz. You will see the same thing with 802.11ac, people rushing to deploy it without ANY of their client devices able to make use of it. If Chazzy2501 tells me what devices he used in his test, I can point out why it would not have been 40MHz - not to embarrass him, but to show how to find these things for your own devices. Or, Psydii, tell me what you commonly deploy (Ap model and client type) and I will explain for your devices. NM

Yep. Find me any client device that's capable of 40 Mhz in 2.4 GHz. Tell me what devices you used in your test and I will reply with an explanation.

You don't get 40 MHz in 2.4 GHz. Tell me what device you were testing with ? There's a lot of misconceptions about what 802.11n offers, in 2.4GHz it's all about reliability.

timtom - You don't have any devices that are capable of ONLY 802.11b, so protection is not likely to be the issue. (There are circumstances where protection may be triggered - neighbour AP's, rogue wifi enabled printers, etc, however I would think unlikely here). What you described can have many possible causes. If it's a serious problem you need to have someone run spectrum analysis and a frame capture and look for: 1) Layer 1 problems - non wifi interference, adjacent channel overlap, transmit power 2) Layer 2 problems - co-channel interference, neighbour access points. To rule out protection, look for a beacon frame from your AP and see if the USE_PROTECTION bit is set to 1 and if the NON_ERP_PRESENT bit is also set to 1. Look specifically for the device data rates in use compared to the expected rates for the given signal to noise ratio. Get a value for retry rates and CRC errors. Capture the conversation of a specific problematic device, and if possible look for the point that triggers the device to try to roam. Only then can you begin to think about what to do to put things right. If I had to guess I would say that you probably have too much power in the system. I don't know how many access points you have, but one starting point would be to reduce the transmit power. I would also be curious to see how you have the AP's deployed. One point, on you use of frequencies, the access points should automatically steer devices on to 5 Ghz that are dual band capable. You should plan your deployment from 5GHz perspective, then only activate 2.4GHz on the access points you need to have on to support the number of 2.4 GHz devices you have. Again, if this is an important network, and you are having these kind of problems, then pay to get a certified WiFi guy with the correct tools to make an analysis. He should be able to narrow down the problem in half a day if he is any good. Before engaging someone, ask them for their wifi certifications/qualifications and what tools they use. Ask to see examples of previous jobs. I hope this helps you to get a solution. NM

Well, it's irrelevant because OP does not have any "B" only devices, so protection is not the issue. I am happy to get into the technical side any time you want, but you are wrong on so many levels I wouldn't be able to easily answer you. I would need to run a couple of days of classes to even scratch the surface. FWIW, disabling "B" rates is a critical mistake. If you want to stop "B" devices from associating, then make 6Mbps a mandatory rate. Honestly, it's nothing personal, you just don't understand wifi well enough, though you are clearly making an effort to learn. You shouldn't take offence, but the harsh reality is that you are giving out BAD advice from an untrained viewpoint. Your advice is manifestly wrong, I am afraid to say, and I call you out not to embarrass you, but to prevent the OP from making a bad situation worse. Until you know better, please don't give out this kind of advice. Keep going, by my estimation it will take you about 2 years of full time WiFi design and analysis to truly understand how these things play out in real networks. That's not a reflection on you - it took me 6 years to get my CWNE, this is working as a full time WiFi engineer and trainer. I also write courseware for AirMagnet and teach advanced analysis classes. What you're referring to is probably the most difficult things to troubleshoot and diagnose, many many senior engineers have problems understanding the big picture. NM

It depends on how serious you are about the reason for the survey and the information you need to collect. If you start with asking those questions first then decide on the survey software to meet your needs. If you need a proper survey done then you should call someone in who is trained and knows how to use survey software. What's your situation ? NM

Good spot ! Naturally, I meant a professional WiFi person / company :>)

Sigh.... 1) Start with your requirements, work out what the ipads are capable of and what you want to do with them. Browsing ? Streaming ? How many simultaneously? Will your requirements today change in 1 yr, 3yrs, 5 yrs ? How will you handle security ? 2) Match those requirements to the capabilities of an access point. 3) Rule out vendors that do not match your requirements. Remember, budget is a constraint so get a decision on how much money you will have available to have a finished set up. Don't forget to calculate day 1 costs, then 3 year cost then 5 year cost (in case you get a licensed model). Do you have capex or opex choices ? (Buying or leasing). 4) Select the vendor from those that do match your requirements, and buy the best that your budget will allow. Metrics to judge "best" are not only performance. Vendor choice should NEVER be the first question. PS - this sounds like a small project but as it's for a school, it could have the potential to get bigger. Any project in education deserves to be treated as important and approached with the same attention to detail as bigger projects. Ask yourself if you really know enough about wifi to make an educated decision. If not seek professional help, but make sure the person advising you does know what they are doing.

What is the retry rate and the crc error rate ? What data rate are the frames being transmitted at ? How many other devices are visible and how many more are visible and active as you do your test ?

MIMO was introduced with 802.11n. A spatial stream is a transmission from a single antenna, 2 spatial streams = 2 transmissions. So true MIMO means you have to have 2 antennas/radios at the transmitter and receiver. Many client devices only support 1 spatial stream, and in 2.4 GHZ there are no 40MHz channels, so the BEST data rate available is 72 Mbps, which is a long way from the 300 mbps advertised. Nm

OK, quick answer - OFDM (Orthogonal Frequency Division Multiplexing) is the modulation method used in 802.11a, 802.11g, 802.11n and 802.11ac. OFDM uses distinct sub carriers within a channel, each subcarrier will transmit a wave that has an amplitude and phase difference when compared to a pilot wave. The difference denotes a bit value, the meaning of which is derived from the specific coding and error rate. It's an advanced topic,covered in depth in the advanced CWNP classes (CWAP/CWDP). You don't need to know how OFDM is technically performed, more important is to understand why a specific modulation rate is in use when you would expect a different modulation. The antenna is the means by which a radio creates an energy wave in the air. Antennas can be internal or external. Antennas are used to shape the energy of the wave. The CWNA class has a whole section dedicated to antenna theory, and it's important to understand the characteristics of the antenna pattern when planning a network. When it comes to spatial streams, the first number usually denotes transmit radios or antennas, and the second, receive radios or antennas. So, a 2 x 3 would be transmit on 2 simultaneously and receive on 3, for example. I thought about making videos but decided against it. For a living I teach WiFi classes and help end users understand wifi planning and deployment. If I were to produce videos, it would really be to help promote my work. With all due respect to everyone here, I don't think there is any interest in training videos. Most people are happy just to get by without professional level consultancy, which is fine. Education has pretty tight budgets and the decision makers usually wear many hats and have lots of conflicting responsibilities, usually being knowledgeable on WiFi is not a high priority. If you personally really want to learn the best place to start is by studying, and possible taking a class. I recommend the CWNP Certified Wireless Technology Specialist (CWTS) track, it's a great course for beginners. I am happy instead to step back and just help out when people would like to understand a technical point a bit better. NM

What was the question again ?

What makes you say this ? NM

This excellent video shows a completely independent and neutral stress test of WiFi in a schools environment, with different vendors tested. Well worth checking out to help understand the constraints of School WiFi and some hints about planning. Wi-Fi Stress Test is explained - YouTube Second Video explaining the results: (A bit technical) NM

All of this shows why it's even more important to understand wifi before making any buying decisions. Professional WLAN designers use design tools to balance all of these requirements in order to show the most effective design to meet the requirements, and it's sometimes tricky. It takes a big investment in time, tools and experience to become proficient at WLAN design. I wish that more people would consult with WLAN professionals before ordering enterprise equipment and just randomly sticking them up. NM

DHCP is given through Data frames, and you can only send data frames once you associate to ann access point. So, first of all check the logs and see if the client is being shown as associated. Next, run wireshark on the wired portion of the network to trace the DHCP D-O-R-A process. If you don't see the DHCP discover packet coming from the station on the wired, then you need to do a frame capture of the wifi. This is not a particularly wild problem in the world of wifi, with the right tools and knowledge you can easily and quickly see where it's failing. NM

I'll stand corrected, I must have looked at an older spec sheet - I was sure I saw 802.11n modulations only available in 5 GHZ. 2.4 GHz, 5 GHz - Switchable Tx 1 tested Spatial Streams 2.4 GHz Rx 1 tested Spatial Streams 2.4 GHz Tx 2 tested Spatial Streams 5 GHz Rx 2 tested Spatial Streams 5 GHz40 MHz operation in 5 GHz The max rate available then will be 72.2 Mbps, probably 65 Mbps without short guard interval, so very little benefit over the current 54 Mbps, though it's a more advanced AP and will use better signal processing and error recovery. You are right it, it is indeed a generous offer and will give the OP a chance to immediately see the impact of switching to N, though 2x2 in 2.4 Ghz would be better, though his devices may not support it. These will buy time and may even be enough to solve the problem, it's kind of you to offer. NM

It wouldn't matter as the total bandwidth available via the WiFi is only 54 Mbps, so 10 devices would end up with about 2 Mbps each on the wifi side, assuming all can modulate at 54 Mbps. The reality will be less. The ethernet wouldn't be a bottleneck in the current set up, (though 802.11n AP's do require a gigabit port) The free HP access points is a generous offer, but I think in this case the OP needs to get something more modern. NM

It doesn't take a lot to learn the basics of WiFi, and there are courses out there that are designed for people in your position (ie CWNP - CWNP | Certifications - look for CWTS, Certified Wireless Technology Specialist). You should understand WiFi enough to be making the kind of decisions you have to make. For access points, check out Indoor Dual or Single-band 2x2:2 802.11n AP (ZoneFlex 7372 & 7352) | Ruckus Wireless, but do also look at other suggestions from other members here. NM

First of all, you should learn a bit about wifi and how it works. In your classroom, only one device can talk at any one time, so it's not about how many simultaneous users you have. Access points can allow many devives to associate, that's not the problem. The issue is throughput vs capacity vs what's available. You have to get a dual-band solution in place and get as many devices as possible using the highest data rates possible. For a quick solution, go for the latest ruckus AP's - you can get 2 which will work as standalone, you don't need a controller. (we can debate the different vendor choices, but that's my first thought based on your description of needs). An immediate swap out will provide benefits. By using 802.11n, you will get a lot of benefits even if your stations are G only, however you must look to upgrade the stations to n as soon as possible and dual band (5 GHZ capable). What you have now is exactly as you describe, a cheap solution that was never intended for what you are using it for, so you should not be surprised it doesn't work. Secondly, you have to get out of this mindset that you can solve this for no money. Decent Wi-Fi equipment is an essential spend, you should not be expected to fight for a solution that works when you are trying to provide education. NM

Only 1 device can talk at once, so if your design is going to be critical you need to work from the very beginning with a good understanding of how WiFi works. The differences between vendors are not so great unless you know how to configure it to get the extra benefits. Manufacturers will give figures but you have to understand how AP loading works, plus how to make extra gains with QoS, balancing, toggling data rates and other tweaks that will change the environment. 10 internet users will behave vastly different to 10 VoIP phones, simply adding more access points does not increase performance and can often make things worse. It's never about how many devices per access point. You have to start with the client devices and work up from there. Complex network design needs a skilled professional, if you try and do it cheaply and with minimal skills then you design faults into the network that will be difficult to put right later. You should thoroughly analyse your client devices and their needs, this will tell you how many access points you have to install, then you have to try and balance the signals. Vendor selection is about in the middle on a task list of things to do when planning a network. Really, if it's an important network that will be used for more than just casual browsing, you need to plan correctly, and get professional help. NM

And we conclude, the wifi was behaving as it was designed to .... Great you got it worked out. NM