Ofcom coverage map: 7/10 – must try harder next time Ofcom has released a map of where you should have mobile phone coverage. Based on computer models and limited testing, however, it’s not a map of where you do actually have mobile coverage. We searched one of our favourite blackspots – a point in Hampstead where the networks deliberately reduce coverage to avoid the signal …
Carter sees the map as helpful, but said: “To make it an accurate and useful tool, Ofcom needs to regularly test networks both indoors and outdoors around the UK, using only data collected in real life, and not a mash-up of testing and computer simulation."
So he wants Ofcom to send some poor sucker out to cover the entire land area of the UK, cities, villages, fields, mountains, swamps and all, stopping every 100 metres, to measure the actual reception on every network? The phrase 'job for life' springs to mind...
I no longer give out my EE mobile number for calls to come to me. If I am at home I use the land line, outside I carry the phone in the hope that if I need to make a call I will be able to do so.
Apparently the mobile service only works incoming to me for PPI, new mortgages and ambulance chasers.
The 'map' of course shows that it is perfect, in who's dreams?
I can help with my house, too, and between myself and my various friends and relatives we've tried all the networks. The signal out in the street is OK. Indoors, downstairs is a black hole. Upstairs has a couple of places near windows where you can pick something up. The house is 1970s brick construction just outside Cambridge (in one of the areas marked "OK" on the map in the article), not some Aberdonian granite castle with walls three feet thick.
Fortunately, mobile phones now have apps that let you piggy-back on the domestic wifi. Hurrah for land-lines!
Fortunately, mobile phones now have apps that let you piggy-back on the domestic wifi. Hurrah for land-lines!
Except that on the map, the area between Dereham and Fakenham has ok phone coverage, but in reality it is 1 bar and GPRS on a good day for 02 and Voda, 2 bars for EE, 3 is so bad on anything other than 3G data at 2 bars, and most houses there are still dial up as they are over 4Km from the exchange
>So he wants Ofcom to send some poor sucker out to cover the entire land area of the UK, cities, villages, fields, mountains, swamps and all, stopping every 100 metres, to measure the actual reception on every network?
Don't have to do that - the operators have tools that get their punters to do it for them. Now, if Ofcom regulated the desired output of such tools, and the amount of data to collect, it would be relatively simple for the operators to comply and provide data sets periodically (although to a degree it also requires some effort from the network equipment vendors to support it).
There could be an app for that, couldn't there? It wouldn't give accurate power readings due to different phones' sensitivities and the reception bars aren't calibrated, but the fact that I've got no bars, or GPRS only or 3G etc is a reflection of a user's actual experience and might be better than nothing in terms of understanding what real-world phone coverage is like.
Put an app on the phone, let it report signals every few hours or once a day, or store them in a file for uploading when signal returns.
Privacy worries would be understandable, but shouldn't be insurmountable. The networks would hate it because many smart phones are deaf (my Sonim sucks signals out of space when iPhones all round me are showing no signal) and some areas where they claim good coverage would probably report bugger all signal - but that's the real user experience.
In a previous life I spent months driving a car around verifying that cellular coverage matched computer predictions. The predictions were very accurate and that was 30 years ago. What surprised us most was how many deadspots there were in terrain like the Fens.
Telcos know full well what coverage will be like down to about 10 metre resolutions. They choose not to divulge it for commercial reasons.
I'm not sure why Ofcom is reinventing the wheel. There are a number of crowdsourced coverage maps they could/should be using as input and the only reason why they're not would have to be "Not invented here"
"So he wants Ofcom to send some poor sucker out to cover the entire land area of the UK, cities, villages, fields, mountains, swamps and all, stopping every 100 metres, to measure the actual reception on every network? The phrase 'job for life' springs to mind..."
Odd, to me, the phrase "Can you hear me now?" sprang to mind.
It's all very well bitching that the coverage map is at least partially based on predictive models, but if a particular location attracts no punters to make calls at it, then they can't very well say what the call quality will be at that point and are forced to rely on prediction models.
Additionally, anyone relying on a "green" pixel to say they will have good quality coverage in their basement faraday cage is also rather foolish.
There are tools around for the operators to produce these maps based on subscriber experience, but there is high variability in a variety of factors that influence the perceived worth of such maps:
a) geolocation accuracy - can be reasonably good in inner city, but out in the boonies if you only have coverage from one cell the accuracy can be a bit pants
b) statistical certainty - you need a goodly number of different mobes to make a statistical statement about the quality of the coverage in an area
c) environmental variables - leaves on trees cause problems, weather causes problems, so time of year will influence the results
d) overall network load - it's conceivable that a horde of punters doing something in a small area will have a detrimental effect on their surrounding area
Yes, you can normalise for all of the above, but it adds a lot of smoothing that makes the results less meaningful.
Short of the operators being forced to generate very detailed time series data for what happens to their network in every area, these maps are always going to have to be taken with a generous pinch of salt.
I agree that being able to do a network comparison would be quite nice, though.
Geolocation is far more precise out in the boonies than in a city.
The urban canyon is pretty much the worst case for a GPS receiver.
GPS is amazingly precise and accurate when there's no buildings to confuse the receiver.
The problem with the OFCOM stuff is that they use calculated for almost everywhere, and only measure raw signal strength in the few places they've measured.
Thus they don't spot the overloaded cell that couldn't carry a connection of any type, or the noisy link that can't carry anything more than a text message.
Real geolocation works by analysing layer 3 messaages and trianglulating the results. It's quite good but not perfect and doesn't work too well out in the sticks. (I know this as I've just spent the last 2 weeks eveluating a geolocation tool)
Geolocation looks at all the messages,even when not on a call/active data session so it's a lot better than GPS which only works when a user has GPS switched on and is engaged in the trial.
But they get points for trying :-/
Yes, they do.
But you can help make it more accurate by submitting your response to the feedback request. For myself I've suggested that it would be quite feasible to have a form to submit a user experience report asking for the reception strength reported by the mobile phone? That'd then modify the predicted data, and over time the map would better and better shows the truth (although I'm not suggesting a user-editable map).
And I suggested they differentiate the "amber" category into upper and lower divisions of performance.
We've had our account manager in multiple times to discuss the options for improving signal coverage in our urban area because the signal is so bad. The carrier has even performed their own surveys which confirmed the poor coverage. Yet the OFCOM website says I can get good indoor and outdoor signal! The map is utter rubbish.
Or is there a total lack of link to this map in the article?!
As for the accuracy, why not get your customers to do it? Apple and google have already built wifi/cell based AGPS databases from the data logged by the phones, why not do the same with signal quality...
(Yes, I know there are already such 3rd party tools available, but they need more feet on the ground).
Curiously, I have a very very small blackspot near me... It's about 20 feet long... on a straight flat road with fields on both sides... Oddly enough it's right near a black-spot eliminator mini-mast... Unfortunately it's not a mini-mast for my network, so I suspect the strong radio transmissions from it are completely desensitising the receiver's AGC in my phone when I'm very close to it.
I tend to agree that these OFCOM coverage maps don't bear any relation to what happens in reality.
If they want to do this it really seems that they need to tap into something like rootmetrics, or slap some equipment onto a fleet of delivery vans / Google Streetview cars to actually monitor what the reality is.
Though Noel Coward put the words "very flat, Norfolk" into one of his plays, its only the western part that is Fens. The rest is gently rolling land (with a terminal moraine running across it) and substantial number of watercourses and lakes (The Broads).
And I suspect some of locals don't care for masts in some parts of the natural environment. Though we do have many many church towers (famously round) they could hang their antennas on.
This post has been deleted by its author
Ofcom implies that houses with thick walls are an oddity that they can't be expected to consider. But here in the Peak District (and in may other rural areas), thick stone walls are the norm. By ignoring such needs, Ofcom neatly understates the extent of coverage problems - and so reduces the pressure to do anything about risible rural coverage (eg domestic roaming). How convenient.
I checked out a few places where I have personal knowledge of the signal levels (O2 and 3) and so far as I can see the map's "predictions" bear virtually no resemblance to the reality.
For example, I sat on a SouthWest Trains mainline service from Southampton to Waterloo the other day and the average data connection success was below 10% for the entire journey, and the on-board WiFi was no better at delivering the bits either.
Until there's an incontrovertible volume of real-life data from actual users (such as I try to contribute via the Rootmetrics app), Ofcom and the operators are going to continue to sit fat and happy with the abysmal state of mobile coverage in UK.
This post has been deleted by its author
Just like everything they do, Ofcom's map is utterly hopeless.
I have a standard brick built house, and only Three works indoors. The other networks have next to no signal - I used to be with TescO2 and missed many incoming calls. I had to leave the phone on a sweet spot on the dining room table and dash upstairs to take a call. Calls would often drop out, even up in the study on the second floor.
Yet the map says that all four networks have a good indoor signal !
I live a mile from a motorway, in an area with solid green coverage for voice, 3G, and 4G.
Data is useless, voice coverage patchy. While this is a rural area, I get the same lousy sugnal outside the local Tesco.
Add to that the way the map doesn't even show something as fundamental as the woods and built-up areas. You can infer the presence of houses from the clusters of roads, but the trees also affect signals. and they have a habit of growing.
Conclusion: Ofcom are giving us the exaggerated claims of salesmen, with a flimsy veneer of official authority.
And where is there any site which tells us how sensitive is the hardware on sale to us. Maybe the iPhone would deserve a 5* rating for its aerial, maybe my phablet is solidly 1*, but can we ever know? And I am sure we all remember the complaints about new iPhone models and their reception performance.
The mobile phone business seems to have a surfeit of liars.
Many other countries derive their coverage information from users via Apps running in smartphones. These are FREE - and the data is damn sight more realistic.
Cellco's use trunk/boot mounted receivers attached to external antennae. Hardly the optimum way to gather cell coverage INSIDE homes!
What's with OFCOM - are they with the consumer or the trade groups?
There are cellphone apps that will try to collect data and generate a map. There are pros and cons compared to a computer-generated map. The pros, the areas I was in the coverage maps was accurate, and one of them has a speed map in addition to just signal strength. Cons: If no-one has been there it's not mapped. Some cities appear to have blanket mapping, but you zoom out much and probably 10% of any given carrier's coverage has actually been mapped by anyone.
One big problem with modelling is cell breathing. This is the tendency for the site's coverage to reduce under load.
GSM (2G) does not suffer from cell breathing, you either have free timeslots or you don't.
CDMA (as used here in the states) and WCDMA both suffer from cell breathing. For any user on the channel, every other users traffic adds to the noise floor they have to deal with. The devices then have to transmit at slightly more power to be heard over the noise, which makes the noise floor ramp up more. As this noise floor ramps up, users at the edge of the site coverage are no longer able to use the site at all, it's coverage shrinks. This cell shrinkage does affect the signal strength shown on the phone, bad cell shrinkage can mean 2 bars of service overnight and "no service" at peak times.
LTE doesn't have actual cell breathing, but can have a similar effect depending on how the site is tuned. LTE uses "resource blocks" (slices of the LTE spectrum and timeslots) to send data to your phone. If you get a clean signal you get highest data rate in each resource block, if not the amount of data you get in each resource block is lower. This is where site tuning kicks in, once a site hits full load it can be tuned entirely for speed, for range, or in between. A site tuned entirely for range will try to give every user a minimum mbps, this helps make sure speeds don't crap out until the signal is quite weak, but means when you are near a cell site, you might be getting considerably lower speeds than you would on a site tuned for capacity because distant users are using up many resource blocks. A site tuned just for capacity (maximizing site mbps), will give resource blocks to the closest users. If the site hits capacity, nearby users get the best possible speeds but distant users will then get few or no resource blocks (it'd still show the same signal strength but not get any data.) A policy in between would give you some kind of intermediate behavior.
If the mobile phone companies spent a fraction of the money they waste on marketing to add new base stations (or high powered versions of Vodafone's SureSignal for example), the UK would benefit from good coverage. We have had mobile phone 'service' in this country since 1986 and it's still abysmal. Just go across the water to the Nordic region, France, Holland or Germany and you will see how much better service is. Simply put, the mobile phone companies in the UK don't give a damn about the subscribers - in the rural area service is a myth.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
