‘Mobile misery maps’ offer new route to better services on the move
The common experience of swearing at your mobile phone in the train or car as the connection drops out for the umpteenth time could soon become a thing of the past.
Vodafone Ireland has conducted one of the first live trials of a technology that can pick out millions of fast-moving phone subscribers from their Call Detail Records (the logs of time, date, place, duration and type of communication that phone companies collect every time we use our mobiles) and produce live maps of how they are using their phones (and how rubbish the experience is).
The idea behind ‘mobile misery maps’ is to help phone companies improve their services around pain points on roads and railways, with one eye on the future of increasingly autonomous vehicles.
The maps mark each cell site along a route in ‘traffic light’ colours, from red to green. These can be filtered for specific call-quality indicators, including problems like low-quality VoLTE calls, slow video streaming or frequently dropped connections.
Integrating such data into car satnavs could potentially let us plan journeys around quality of mobile experience rather than the shortest or fastest route.
Or to put it another way, who wants a self-driving car if you can’t sit back and stream the latest series of Fargo on Netflix without interruption?
Mobile sessions that run smoothly at speed are considered a minor miracle by the mobile industry. During its trial, Vodafone Ireland focused on 60km of roads that run in and out of Dublin and found that it could reduce dropped-call rate by 28 per cent by tuning networks around the user experience maps.
The Vodafone Ireland exercise was part of a collaborative research study with the data analytics company CellMining, which has run a similar study with a phone operator in Shanghai.
CellMining’s mapping software effectively regards subscribers as hordes of network testers. It can distinguish moving from stationary subscribers because Call Detail Records (CDRs) show the location of each basestation used and how quickly we pass them.
“With CDRs, we can detect how fast people are going as well as the services they are using, such as video, audio or plain voice calls. We can separate who is driving on the highway and who is sitting on a fast train nearby within a few metres,“ said Omer Geva, president of CellMining.
“When a phone call lasts only 30 seconds and the call made afterwards also lasts only a few seconds, the shortness of these calls suggests a problem with voice services."
Service quality on the move depends on many factors, including the number and type of basestations along the route; the statistical mix of LTE-capable versus 3G-only handsets that attach and require services from the basestations and how congested the data pipes are out to the wider network.
These factors could be automatically ‘tuned’ using customer experience measures in a feedback loop, says CellMining, which developed its mapping software from a product it sells for Self Optimising Networks (SON).
Potentially, CDR-based ‘mobile misery maps’ could predict which customers are the most fed-up and likely to leave, giving mobile operators an extra incentive to tune up their networks.