Testing AP Transmit Power Using AirMagnet

If there's one recent change in the Wi-Fi world that has brought some heat and light to your humble blogger's cold, dark heart, it's the awareness of device differences.  But there's a difference between understanding that Wi-Fi devices have differences and creating an infrastructure that supports them.  And an important step in setting up a good Wi-Fi infrastructure is finding out whether the APs' transmit power is too high, too low or juuuuuuust right. 

It has long been the recommended to have matching transmit power for devices engaging in two-way wireless communication, but with Wi-Fi that rule is often broken.  Breaking the rule is understandable because Wi-Fi environments are often vendor-neutral.  Most Wi-Fi deployments are not like the beaches of Normandy, where everyone had the same model of radio.

Sometimes we need really, really good Wi-Fi, and matching the transmit power levels of APs and stations can go a long way towards fulfilling that need.  Matching AP and station transmit power is great because it facilitates two-way applications.  And lots of users like to use two-way applications nowadays.

(At this point it should be noted that if the applications you support are primarily one-way applications [that might include IPTV, educational video streaming or Internet radio], then you probably should ignore what I'm saying.  If your applications are primarily downlink applications, then your optimal Wi-Fi infrastructure probably is going to have AP transmit power be quite a bit higher than typical transmit power levels for smartphones and tablets.)

The tough part about matching AP and station transmit power is that it is often hard to find out exactly what power level stations are using.  FCC documents can help (try searching "[my device's name] FCC" and you might find some leads), but the fact that a lot of devices use a transmit power level below their registered maximum makes it tough.

My technique for finding out which transmit power level works best on APs is to test connections using a Wi-Fi protocol analyzer.  Any protocol analyzer that can use Monitor Mode will do.  I've used Wireshark, OmniPeek and AirMagnet WiFi Analyzer, and they all work great.

Here's how to use AirMagnet WiFi Analyzer (I'm going to call it "AirMagnet" from here on out for brevity's sake) to test whether AP and station transmit power is balanced:

1. First, open AirMagnet and start capturing only what your AP is transmitting and receiving.  To do that, select the "AP" tab in the Start screen of AirMagnet, then double-click on your AP to get to the Infrastructure screen of AirMagnet.

2. Next, configure AirMagnet so that it filters only traffic being transmitted by your smartphone/tablet/laptop.  To do that, click the "[+]" to the left of your AP while in the Infrastructure screen of AirMagnet, and then click on your device (by MAC address, hostname or IP address).

Once you've clicked on your device, go to the lower right corner of the screen and use the dropbox (an unconventional looking dropbox, but still a dropbox) to select "Tx Total/% Total".

Check out the percentage of Retry frames that your device is transmitting by looking under "Frames/Bytes".  (Hopefully your Retry percentage will be a lot lower than what I saw from my iPod when I was Periscope-ing the creation of this blog post.  12.18% is way too high a Retry% for an ordinary residence during work hours, even in a relatively crowded neighborhood of Los Angeles.)

3. The last step is to configure AirMagnet so that it filters only traffic being received by your station, and then compare the Retry percentages.  The dropbox that was previously configured to "Tx Total/% Total" just needs to be flipped to "Rx Total/% Total".

(In my case, the received Retry% for my iPod was much more acceptable than the transmitted Retry%.  Anything under 5% is usually considered good in a crowded residential environment.)

By comparing the transmitted Retry% with the received Retry%, you can figure out whether your AP transmit power is too high, too low or just right.  When the transmitted Retry% for a station is higher than the received Retry%, then AP transmit power is too high, as it is at my place.  (Unfortunately, I have an old Apple Airport Extreme wireless router that does not allow its transmit power to be reduced in the configuration GUI.)  When the transmitted Retry% is lower than the received Retry% (as it often is in modern enterprises, due to the unfortunate "picocell" phenomenon), then AP transmit power should be raised to avoid a meltdown when the Wi-Fi network gets busy.  And if the transmitted and received Retry percentages are approximately equal?  Great job!  That means that you already have AP transmit power set to the level it should be at.

One last note I should add is that a big difference in transmitted and received Retry percentages most likely means that there is a transmit power mismatch between AP and station, but there are exceptions. It is possible for Retry percentage differences to be caused by unusual device behaviors.  Most of the time, however, the Retry percentages are a great way to test whether the transmit power of the AP is best for durable Wi-Fi performance.

***

If you like my blog, you can support it by shopping through my Amazon link.  You can also donate Bitcoin to 1N8m1o9phSkFXpa9VUrMVHx4LJWfratseU or to my QR code:

Twitter: @Ben_SniffWiFi

ben at sniffwifi dot com