OK, so Ofcom – the UK’s equivalent of the FCC – didn’t mention us by name, but in their latest study they do show up some of the failings of the networks. More interestingly, the study notes that “third party websites that aggregate consumer information could play a role” in helping consumers – in particular to compare the coverage of different providers.
The study, Mobile coverage information for consumers, highlights a number of fascinating points – one in particular that might surprise you is that 2G phones often do a better job of making calls in rural areas than 3G phones. That is 2G and 3G phones on the same network, this is due to the greater complexity of the 3G hardware. 3G phones do a lot of work “handing over” (transferring) between 2G and 3G and this can cause to dropped calls.
Testing in Devon, the difference in performance between 2G and 3G phones is staggering: the former has 97% call robustness, the latter only 95%. What’s 2%? Actually it means you’re 66% more likely that any given call will fail on 3G. That’s a big difference, and makes me a little nostalgic for my old Nokia 3310, well just a little bit. No word on 4G vs 3G here, as there’s no 4G in the UK yet.
As a quasi-non-governmental-organisation Ofcom is in a privileged position to get information from the networks, consequently the study contains some noteworthy tidbits. We learn that the operators coverage checkers are based on ‘propagation models’ this suggests that the operators use the known power of their antennae and then model the signal falling off as it is attenuated and spreads out. This is in contrast to directly using signal readings either crowdsourced (as we do) or taken by ‘drive-testing’ – where reading are systematically taken as contractors drive around with measurement equipment.
Ofcom found this methodology to be fairly robust – “underlying operator coverage data is reasonably accurate for outdoor predictions”. We do, however, wonder how well the propagation-model approach works where there are lots of obstacles (hilly terrain, buildings) to take into account. Even accepting that the “underlying data” is good, Ofcom found that the way networks present that data left much room for improvement.
“Our work has highlighted presentational differences between mobile operators’ coverage checkers and differences in the processes they use for translating propagation models into coverage checkers.”
What does that boil down to? It means that looking at Vodafone’s coverage checker and O2′s coverage checker isn’t a very useful way of comparing their coverage – even if the data is similar for the two it could appear rather different.
This is where Ofcom suggests that sites like ours could come in:
“There may be some additional value to consumers being more aware of coverage information and being able to access it in a more comparable way. To do this we believe third party websites that aggregate consumer information could play a role.”
We hope so!