Forum

Dear Specialist,

I have two questions about the RF signal fluctuation and I hope you can help me to find an answer.

If I use a tool such as inSSIDer I can see the signal fluctuate 6-8dB without ever moving the measuring device. Do you know why such behavior occurs? Yesterday I made some tests and what I noticed (a notebook with measurement tool was not connected to a wireless network):

1. No clients connected to AP – signal fluctuate 6-10 dBm (insider and Fluke measure tools)
2. Some client connected to AP – signal doesn’t fluctuate (1-2 dBm is a tolerance)

Next situation/scenario:

A guest in a hotel room has too low RF signal and can’t connect to a wireless network (1 bar in MS Windows tool). When he goes to a hall (where AP is mounted) the RF signal is great and he can connect to the wireless network (5 bars in MS Windows tool). He goes back to the room and he can normally works and the wireless network is available and the signal strength is good (3 bars on MS Windows tool).

Kind regards,

Marek

What are the actual signal levels, not just the relative numbers?  If you are already at low RSSI numbers I would expect more variation.   Remember that InSSIDer is reporting RSSI values, not dB, and you likely don't have any idea of the real signal level seen by the radio.  

However, I think it's more likely that their are other effects going on, especally people moving around between the PC and the AP.   It might also be the radio switching between its diversity elements.   Are these devices indoors or outdoors?  Have you checked for interference with a spectrum analyzer ?

As to your second question, many radios will connect to an AP, up close, where they are getting a good signal, and then  be able to travel quite a distance without losing the connection.   The same device may not connect initially at the longer distance.  

Personally I would call this a firmware or driver problem, rather than a hardware issue, but that probably doesn't help the end user.  

Because of this, I always start range tests at the maximum required distance, and move in from there only when necessary.    

Thank you for your response.

All tests were made in a small room with all the hardware on a table. One AP, two notebooks - one with testing software, second used as a wireless client.

I wonder why the RF signal stops fluctuate when some wireless client is connected to the AP.

Of course, the second problem is a design issue but I would like to know why such a problem can exists and why the signal is stronger in the same place when the wireless client is connected.

There are probably many reasons.   Does the AP have diversity?   It might be switching between antennas.

Is the AP set to automatic channel selection?  

The devices may even  be too close to each other.  Can you set the AP power to a lower value?

Roaming and DRS algorithms might cause this.  Can you change roaming aggressiveness settings on the client?   

How many AP's does InSSIDer see on this, or nearby, channels.   How strong are they compared to your AP's signal?   

Do you have much Bluetooth activity in the area?   Have you disabled BT on the PC's?

Thank you again.

I have asked 2 questions, now I have several:) Let's back to the subject.

1. It was MSM 466 with almost all in default state, so automatic channel selection and other features on.
2. The AP power set to 10%
3. There were a lot of AP around because I did tests in the office. A signal strenght fluctuated between -27dBm - -36dBm, after a client connection it was flat between -28dBm - -29dBm. Other APs signal -47dbm - -75dBm.
4. The BT is switch off by default in my notebook, so I can assume there was no BT active around.

For the time being I can't repeat tests but I will try do it in a close future to check your suggestions.

If the AP is using automated RF, based on the environment it keeps changing the power. It should not change that much (May be algorithm issue). When a client is connected, the algorithm also considers the clients signal strength to calculate the power to be assigned to the radio.  Hence you are seeing much less variation. To narrow down the issue can you set the power manually and check whether you are seeing same variation (if Yes then its entirely different issue).

There was not too much on the HP site on that AP - There probably is more somewhere but I couldn't find it quickly.   For example, I didn't find its true Output Power - only a statement that it is controlled by the country authority.

It does say it runs with automatic radio power levels. 

There are at least three areas you should consider:

1.   Does it share its radio with its Spectrum Analysis function, or is it a seperate device inside the AP?

2.   The power level variation could be due to its Beam forming mechanism.

3.    Since it is multistream-n,  it might have fewer antennas running until something is associated, or it might have a lower power setting in stand-by mode to market it as a "green" device.

Please let us know what you find out.

Thank you for your responses.

1. I have no idea - no information.

2. TxBF is not an issue - I checked both settings (with and w/o) and it works the same way.

3. Maybe but I can't check it.

EIRP for Poland is 26 and the AP works with such a power.