Our local installation in RASH consists of two Ubuntu 18.04 servers. One is installed as the Analysis and the other one as the Test server. Both of the servers are located at the same place, in RASH datacenter. Since we have locally configured all the necessary components for WiFiMon to function, we can now monitor our Wi-Fi network on a daily basis. Our implementation of WiFimon gathers two types of data which respectively are: the gathered from the WiFi users (soft probe data) and the hardware probe data. In order to make all the right configuration for the hardware probe to function correctly you may follow the steps in the Hardware probe configuration.
In our office we are using an old laptop that is serving as a hardware probe. It is set up according to the same steps for the Raspberry Pi hardware probe. Since the distance between the hardware probe and the access points is important because of the signal attenuation and the impact of the physical obstacles we have decided to measure the case in which two components are in close range with each other in order to get maximum throughput. So the hardware probe and one of the access points that we are using are located in the same office and the real distance between them is approximately 2.5 meters. The other access point is located in another office and the distance between these devices is 7 meters with a thin wall between.
The number of employees in each of these offices is respectively three (first office) and five (second office). We have embedded the JavaScript code in four websites that we often use and the results that we get are the number of tests that are made each day including the download and upload rates, the number of client browsers that are used and also the number of types of OS that are being used. The number of tests made varies each day depending on how often the websites that the java code is embedded, are used.
To get a clear view of the measurements during the day you may find below a screenshot for the measurements about download and upload rates. These are the data that we do get from our probe. You can see that the traffic started at 09:00 AM and during the day there may be some variations on traffic, this depends on the number of users. In the graphs below you may also see that the traffic dropped significantly during 13:10 and 13:50 PM, during the lunch break which many used to browse various content on the Internet.
Figure 1. Information about download and upload rates
Another perspective about the graphs above is presented below where we have selected the last hour and you may see the variations. With WiFiMon it is possible to zoom into a specified time period, in order to have closer look at the graph, for the time period that you are interested.
Figure 2. Graph about the last hour
Additional information can be gathered from the hardware probe. With WiFiMon you can collect information about the bit rate, the link quality, the signal power and the transmitted power. A screenshot from our environment is presented below. As it can be seen, these technical measures as seen on the hardware probe do not show the moments when the perceived drop in the throughput was observed on the hardware probes – signal level, power and quality were on a more or less constant level during the whole day.
Figure 3. Hardware probe related information
The graphs that are presented above are only a few of examples that are offered in WiFiMon. You may also find specific information about the number of the OS that are mostly used the types of web browsers and also related download rates. An example is presented below:
Figure 4. Download rates for OS, test tools and browsers
This was just an overview about WiFiMon and we hope that we have helped you a bit to follow the necessary steps to configure and get results from WiFiMon. If you do have any questions please do not hesitate to contact the WiFiMon team at gn4-3-wp6-t3-wifimon-request@lists.geant.org.
Good luck!