​WiFiMon: Wireless Crowdsourced Performance Monitoring and Verification 


Why do Institutions need WiFIMon?


WiFi access (and in particular eduroam access) is now accepted as an essential part of the wider campus experience.  Not being able to connect to network resources can damage user experiences and can result in users having to use mobile 4G data connections rather than WiFi.


Most WiFi monitoring services merely provide information on the Availability of WiFi and can't measure the actual performance of the service.  A poorly performing WiFi service affects user experiences and can also be a trigger for IT to invesitgate wider performance issues.  But how can performance monitoring work in a situation where users are, by definition mobile?  


Within eduroam facilities the situation can be even more complicated as users can have multiple eduroam access points within range - making reporting of problems harder.


For this reason GÉANT are developing WiFiMon - a crowd-sourced, distributed performance monitoring and measurement service.


What is WiFiMon?

​GÉANT's ongoing activities show a traditional approach to PM&V, looking primarily for infrastructure information by implementing hardware probes on the wireless campus network and answering the question "WHAT have we done and is it working?", but this does not offer any statements about HOW well something was done, and how the end user is experiencing it.


How it works 

WiFiMon works by using an app installed on users' tablets or phones to measure the real-life performance of WiFi by checking access speeds against well-used sites (for example the webmail service) this provides realistic and repeatable tests across different users and different access points. 



WiFiMon Functionality an​​​d Potential

WiFiMon's proposed approach to wireless crowd sourced performance monitoring and verification is based on an architecture design using JavaScript and NetTest/Boomerang server, implemented on essential, highly frequented web sources.  As the term implies, the accuracy of our measurements largely depends on the number of "complete" data sets. From a statistical point of view, the large amount of measured data (by end-user activities) allow us to analyse performance of a wireless network in a campus/conference environment. As it stands today, WiFiMon  furthers the network monitoring development cycle. Particuar focus is placed on  improved measurement verification, GUI development, mobile device app deployment, and WifiMon as a Service (service automation).

​ ​wifimon.png

WifiMON Architecture Building Blocks

 

Find out more about how WiFiMon works here


GEANT WiFiMon March 2018.pdf

WiFi Performance using enduser measurement.pdf


How NRENs provide WiFiMon to their Users​​


The experience gained so far – mainly through the implementation of WiFiMon on various campuses and conference locations, showed measurements on  specific parameters of a wireless network through the use of  JavaScript, and correlated these measurements with the information provided by the various log files of the data source (see WiFiMon Architecture Building Blocks).

 

To found out how to provide WiFiMon to end users click here​.

​​ 



​​