Wireless Access

Currently i have a customer complaint that the wireless signal fluctuation is too many and too frequent. The wireless signal fluctuation was more than 20dbm. Customer was using the mobile wifi analyzer to check the signal. 

Now, the customer request to set the signal fluctuation only with +- 5dbm. This is because currently the customer is implement the automatic robot device in they factory. This robot unable to work with too high fluctuation signal and it will make the robot stopped. 

Is there any recommand configuration can advise? 

We had tired to disabled to the ARM and fix all AP into single channel so that the wireless signal not automatic change based on the environment. I know that disable ARM and fix single channel are not recommand in Aruba. 

Thats is a fairly large fluctuation and is probably and indication of some sort of interference in my expierence.

Rather than fixing the channel you can alter the minimum and maximum TX EIRP under the ARM settings which will reduce the amount of increase/decrease possible. Fixing the channel if there are interference issues can actually cause more issues than it solves.

Alternatively I would also have a look at some spectrum monitoring and see if there is something else going on that is causing the significant increases. I have seen cranes cause wifi issues in the 2.4Ghz spectrum and depending on what requency range the robot is using you may see something like that.

---

@niv lee wrote:

Currently i have a customer complaint that the wireless signal fluctuation is too many and too frequent. The wireless signal fluctuation was more than 20dbm. Customer was using the mobile wifi analyzer to check the signal. 

 

Now, the customer request to set the signal fluctuation only with +- 5dbm. This is because currently the customer is implement the automatic robot device in they factory. This robot unable to work with too high fluctuation signal and it will make the robot stopped. 

 

Is there any recommand configuration can advise? 

 

We had tired to disabled to the ARM and fix all AP into single channel so that the wireless signal not automatic change based on the environment. I know that disable ARM and fix single channel are not recommand in Aruba. 

---

The signal seems to fluctuate in areas with alot of reflections and co-channel interference:

- What access points and what version of code are you using?

- What kind of environment is this and how are the access points mounted?

- What band are the devices operating on ? (2.4ghz or 5ghz)

- What is your ARM min/max power?

Also, before you go making wholesale changes to your WLAN config, what wifi signal analyzer, what device, etc? Some analyzers, especially free or cheap ones, don't do a good job of normalizing the readings when the device chipset goes in and out of scanning. So the result is it looks like the radios are turning on and off when in fact, it's just the analyzer software and device wifi chipset giving you bad info. 

Seeing signal fluctuate from 0dBm to 20 usually isn't seen, unless you are running DFS channels close to an airport and the DFS channel protections are being triggered, OR the APs are rebootstrapping ALOT. 

In this case, which Wifi analyzer or tools is recommended from Aruba to conduct the quality checking ?

If controller based, you can check the arm activity for an ap to see the history of the channel/power changes and the reasons why.

show ap arm history ap-name <name>

Anything that suports a calibrated, known-value sensor or NICs (AirMagnet's products, Ekahau survey software and sensors, Fluke's WiFi testers, etc). While the freeware/software for laptops and androids are handy, the lack of validation of their performance across literally hundreds of different hardware platforms means that sometimes they don't work well or give bad data. Case in point, we had a government customer complain about poor wifi performance. And when he ran InSSIDer on his laptop, saw that signal would only be up for a second or so, then down for 4-5 seconds, then on again. Come to find out the laptop chipset and driver would take 500-700ms to scan each channel, and when in scan mode or trying to roam, we would see the client radio drop off for literally seconds at a time. Other laptops with InSSIDer, and our test tools, showed solid wifi and when he updated the driver after a TAC ticket with the laptop provider, the problem went away.

So if you see this with calibrated types of capture tools, or can show this occurring on multiple devices (increase the sample size of the freeware tools) with multiple types of software, then we would really want to investigate what is going on. 

---

@jhoward wrote:

Anything that suports a calibrated, known-value sensor or NICs (AirMagnet's products, Ekahau survey software and sensors, Fluke's WiFi testers, etc). While the freeware/software for laptops and androids are handy, the lack of validation of their performance across literally hundreds of different hardware platforms means that sometimes they don't work well or give bad data. Case in point, we had a government customer complain about poor wifi performance. And when he ran InSSIDer on his laptop, saw that signal would only be up for a second or so, then down for 4-5 seconds, then on again. Come to find out the laptop chipset and driver would take 500-700ms to scan each channel, and when in scan mode or trying to roam, we would see the client radio drop off for literally seconds at a time. Other laptops with InSSIDer, and our test tools, showed solid wifi and when he updated the driver after a TAC ticket with the laptop provider, the problem went away.

 

So if you see this with calibrated types of capture tools, or can show this occurring on multiple devices (increase the sample size of the freeware tools) with multiple types of software, then we would really want to investigate what is going on. 

---

Thanks for the information !

Hi Jerrod, 

I had done a site survey using airmagnet. Attachment is the document for the site survey using airmagnet.

The signal for the coverage is ok, however, customer still not satisfied on it. Customer need a gurantee from Aruba that how to prevent the signal fluctuation would not be too many and provide a recommanded signal fluctuation range to allow their robotics device function. 

Please correct me if i'm wrong. For my understand, signal will fluctuation more when the AP discover nearby have many same channel or too many interference, AP will change the channel or re-adjust power transmit.

Attachment(s)

lvl2-airmagnet.pdf   2.17 MB 1 version

---

@niv lee wrote:

Hi Jerrod, 

 

I had done a site survey using airmagnet. Attachment is the document for the site survey using airmagnet.

 

The signal for the coverage is ok, however, customer still not satisfied on it. Customer need a gurantee from Aruba that how to prevent the signal fluctuation would not be too many and provide a recommanded signal fluctuation range to allow their robotics device function. 

 

Please correct me if i'm wrong. For my understand, signal will fluctuation more when the AP discover nearby have many same channel or too many interference, AP will change the channel or re-adjust power transmit.

---

The signal from the device itself will not fluctuate; when there is co-channel interference or reflections, the signal will seem to fluctuate.  Questions:

How high do you have these access points mounted?

How far apart are the access points mounted?

What is the output of "show ap radio-summary"?

Thanks, that helps alot. Colin has you on the right track. APs don't have to change power or channels to find slight variations in power output, as that is due to reflections and nulls in the RF due to reflectivity.

You can disable ARM and do a static channel plan, or you can set the min and max ARM power to the same and just let it change channels if/when needed. But to know what is best, the info Colin is looking for will be the starting point. 

Thanks alot. I will suggest the configuraton to my customer for a testing. 

Just to confirm, there is no any gurantee that the signal fluctuation will be stable on a value and not keep change?

Based on your experience, what is the signal fluctuation range consider normally behavior from wireless? +-15dbm or +-20dbm? 

The fluctuation has little to do with the equipment.  It has everything to do with the environment it is being measured in.  Most likey you need to reduce the power on your access points, because you have too much coverage.  Co-channel interference and collisions are probably the culprit here.  The Airmagnet survey gives clues about coverage, but it does not say specifically what coverage is being provided by what access points, so it does not get to the root of your problem.

Here is what I would do:

The ARM profile by default has a minimum of 18 and a maximum of 127.  That is much too high for a warehouse environment where everything is open.  RF can propagate hundreds of meters in open space and that can cause exactly what you are observing.  I would make the minimum 12 and a maximum of 15, and then take more measurements to see if the "fluctuation" continues.

Access points with  ARM do not change channel or power any sooner than 8 minutes, so if there are any fluctuations in power, that is not ARM; that is the environment.  Depending on the access point, it can only output power of up to 21 dbm, so a swing of 15 to 20 dbm could not be the hardware, unless the access point is being turned on and off.

If you could overlay where the access points are mounted in the diagram and how high, that would be a good start for us to figure this out...

Hi,

The high is around 3 meter high. 

Below is the AP's location that deployed. Each red small square represent the pillar. Each pillar distance is around 5 to 6 meter. In this floor location, there is alot of machines that running on testing micro chipset. 

That placement is good.  Do you have the Airmagnet Survey of just a single access point?  Can you filter it to just one?

Hi Jerrod Howard,

For your information, my customer is purchasing small auto robotics device that assisting them to carry the boxes around. This robotics device is a self design by a AGV company based on the customer requirement. This robotics device is using a usb wireless adapters with external antenna to receive the wireless signal for the robotics device. 

Hi cjoseph,

- What access points and what version of code are you using? 

  AP-205 with AOS 6.4.2

- What kind of environment is this and how are the access points mounted?

 The environment is micro chipset manufacturing lab.

 The access point is ceiling mount

- What band are the devices operating on ? (2.4ghz or 5ghz)

 The device is using 2.4GHz

- What is your ARM min/max power?

 Currently we had disable ARM due to customer need no fluctuation more than +-5dbm. 

 In the controller the max Tx ERIP is 127, as i know.