On-site WIFI Feasibility Study

Question

Can anyone suggest a protocol or testing procedure for evaluating WiFi traffic/interference on-site? This isn't something I regularly do so I am hoping that this will not require the purchase of expensive test gear that won't be used.

I'm deploying some custom wireless technology in a Govt space that appears to have many different wireless networks running concurrently. Thus far I've been spoiled and never had to "complete" with other wireless networks and haven't had to do too much in the way of troubleshooting.

I just want to know if it's feasible or if additional considerations need to be made. I believe i can use 5GHz or 2.4GHz.

Answer 1

There are many freeware applications suited for this task, some can even run on your phone.

Check out this list by Network World:

http://www.networkworld.com/article/2183790/wireless/8-free-wi-fi-stumbling-and-surveying-tools.html

Furthermore, a handy Android app is "WiFi Analyzer":

https://play.google.com/store/apps/details?id=com.farproc.wifi.analyzer

The above app will show you graphical representations of shared channel overlap, as well as other stats like signal strength, etc. - very handy for mapping signal strengths.

In terms of technique for your survey, a solid course of action would be:

1: Submit your Plan of Action for approval: Since this is a .gov facility, you'll likely need to comply with their in-house Information Assurance & Change Control guidelines - this will entail putting together your plan-of-action (things you will do) and if need be, a back-out plan (how you'll restore service(s) if something goes wrong). Note that this plan of action should also include the software you'll be running to scan WiFi signals, so that they understand you're not doing something nefarious like sniffing without permission.

2: Obtain a floor-plan, and note the zones which your wifi will be required to cover: Mark these zones with unique identifiers like "Zone A", which will allow you to easily name and track your WiFi mesh nodes ie: "GOV-WAP-A2" - this will facilitate ease of maintenance in case someone else is called in to work on your setup.

3: Using your above zones, map the signal strengths of each existing WiFi signal in the areas you're supposed to cover - choose the WiFi channel which is least-used in that zone to be the channel used by that AP (or multiple APs, depending on your type of coverage / service-set). Record this zone data as part of your survey and note it in your design document.

4: After your APs are configured with the proper channels / etc you desire, do a second site-survey and note signal strengths of all APs in each zone as needed, and do a full walk-thru with a constant "ping" running, to watch WiFi handoff between APs, and adjust accordingly. Note all signal strengths in your final report, so that they can see what you've delivered.

NOTE: With .gov projects, documentation is a huge part - depending on the agency, you may be held accountable in the future if your work is not up-to-spec with what they expect.

Good luck!

Answer 2

Since non-802.11 traffic can impact the utilization of any given channel as well as other 802.11 networks, you really need to use a tool that measures channel utilization.

This comes down to some solution that will have at least basic spectrum analyzer functionality. You can find these in a number of forms including (but not limited to) handheld devices, deployable sensors/probes, or USB adapters/software. Functionality and costs will vary.

If you only want to concern yourself with 802.11 channel utilization, you can use capture software like WildPackets OmniPeek or AirMagnet WiFi Analyzer which will give you reporting tools that can calculate utilization based on the traffic it captures.

Keep in mind that any tool will only give you information at the the time and place at which you are testing. This is something that many people new to 802.11 troubleshooting often overlook. Some examples:

• Time of day can have a significant impact on any measurements you take. If you take your measurements "after hours" when no one else is around, this won't reflect that actual working conditions when everyone is around.
• The location at which you take any measurements will only be able to make use of any usable signal at that location. This means that you may "hear" additional devices/traffic in the next room that was too weak at the current location. This also often means that you can gather a lot of lower data rate transmissions because the higher data rate transmissions are not usable where you are located.
• The environment is always changing. Maybe there is a heavy 802.11 user that happens to be out of the office on vacation when you do your testing. Or maybe next week someone gets a fancy new pair of those "WiFi enabled pants" (i.e. any 802.11 device that may not have been in the environment before).