Jonathan Jensen – Roaming

This week Jonathan looks at the old issue of mobile coverage and asks what the networks are doing to improve it.

It’s been a busy week with Nokia’s launch of the new S60 E71 and E66 handsets. However I’m not going to be covering them here as Ben Smith is already doing them justice elsewhere for SMS Text News.

Something that’s been bugging me recently is the question of coverage. I’m increasingly frustrated that we’re not seeing anything dramatic from the networks to tackle the problem of patchy and inadequate coverage. At home, despite living about half a mile from the M25, coverage is flaky and my handset jumps from cell to cell. I can see the signal strength indicator going up and down and switching between 3G and GSM. Having discussed it with my operator they accept that in building coverage is poor here and yes they do have plans to improve it, but not for a couple of years! Great, but that really isn’t good enough! My wife recently stayed at Center Parcs in Wiltshire and was unable to use her handset in most of the village. Standing on one leg on top of the wardrobe improved things marginally but proved less than practical! Even Victoria station in London suffers from poor coverage on some of the platforms.

So what are the networks doing to create a step change in coverage improvement (and I don’t mean installing the odd new cell site here and there)? 3 and T-Mobile, and Vodafone and Orange have announced infrastructure sharing agreements. These are designed to make it easier and cheaper to improve coverage by giving access to each others cell sites, thereby sharing costs and reducing the need for planning applications for new masts. To be fair it’s early days for these agreements but it seems likely that over time this will improve coverage for customers of these networks (not sure what happens to O2 here!).

There’s been a lot written about femotocells which provide localised in building coverage and route the mobile service via a broadband connection back to the operator’s network. This means the call or data traffic does not use the radio portion of the operator’s network. Benefits here are much better in building coverage and freeing up capacity in the radio portion of the network. These devices are currently undergoing test and it remains to be seen whether they will become a viable proposition for regular users, in terms of simplicity and cost.

Something I’d like to see would be roaming agreements between the operators to provide coverage where individual operators have gaps. Currently the only national roaming agreement is between 3 and Orange (previously with O2) where Orange’s GSM coverage fills gaps in 3’s 3G coverage. Maybe this is an area that OFCOM should look at? Whilst I generally don’t favour more regulation, I think 26 years is quite long enough to wait for ubiquitous, or even just better, coverage!

Jonathan’s also at Sevendotzero.

9 replies on “Jonathan Jensen – Roaming”

Hi Jonathan,

Don't underestimate the importance of the Voda/Orange & 3/T-Mo sharing arrangements. These will deliver very noticeable improvements for many customers.

The best examples will be where the conjoined networks (e.g. Voda & Orange) don't currently share masts in a poor coverage area. If you are a VF customer with poor coverage but Orange have a site just up the road, from next year you will get great coverage as you connect to a site you were previously barred from.

In many instances MNO's are forced to make hard decisions on site placement to give the best bang for their buck, whilst fitting into the existing network design and local planning laws/community wishes. Not for nothing do MNO RF engineers often have Masters or PhD's. And grey hair.

Here in the UK we have some of the best coverage in the world. If you look at a map of 2G coverage there is hardly a place in the country that isn't covered. Notable exceptions are Central Wales, Peak District, Yorkshire Dales and Highlands mountains. In these places, if you have line-of-sight to an urban area or motorway you will almost certainly get coverage. You go to these places for isolation and unspoiled wilderness, so I have absolutely no problem with no coverage there – long may it last. This story is replicated across the globe, with varying coverage based on overall population density and national wealth.

In-building vs outdoor is a major issue, especially for 3G where the frequencies are much higher and therefore the signal doesn't penetrate brick & tile as easily. Is Femto the answer? A lot of players are betting on it, but as you say there are a lot of issues to be worked out, so don't expect anything soon.



Can you clarify something for me Mike42,

These sharing agreements, are they sharing the physical mast, land and associated costs or are they sharing the actual radio equipment and backhauls?

With regards to in-building vs outdoor, I think a lot of faith is being mistakenly put into femtocells to fix a problem that can already be solved with a dumb repeater (which will, no doubt, upset your grey haired RF engineers)

Hello DanLane

What happens is that the 'foreign' network you previously could not access because you didn't show up in their records as a valid customer (I'm wildly oversimplifying here) will now allow you to connect. So the tie-up between the core networks is critical. When you do a manual search of available networks, you see all networks listed. If you try to select one you aren't subscribed to, it attempts to log on and is then rejected. Once the network sharing is complete, you should only see one network instead of two, and your phone will automatically log into it.

Generally the actual sites get sold to a new company set up to manage the sharing, so you could think of it that each MNO becomes an MVNO on infrastructure they previously owned outright, but now have shares in. MNO's contribute to network upkeep according to complicated metrics around customer numbers, throughput, etc.

No doubt there will be some clever clogs who could put it much better, but that's a basic starter.

Re repeaters, they are evil, and are banned in many countries because they can kill coverage for others and because they are very complicated to get right, often don't work as sold. Using something that emits in an MNO's licenced band without explicit approval is just a tad illegal, and will see you in the dock.


Ah, very interesting Terence, thanks. Site sharing is in many ways very hard as well, as rearranging existing kit (BTS, feeders, antennas etc) is by no means easy. You do still remove a significant lump of CAPEX in terms of power/tower/property etc.

Will be interesting to see how well 3 & T-Mo customers react when that gets turned on.


The only place I don't get 3G reception in and around my home town is the shopping mall in the town centre – the worst spot being just metres away from the network's shop!

Coverage on trains and along the south coast is between poor and terrible on 3!

Thanks for some great feedback on site sharing. For me the killer is still in-building coverage. I appreciate that the higher up the radio spectrum you go the worse the propagation but that's all the more reason for the networks to tackle the issue (femtocells?). In many respects it's better when there's no 3G coverage rather than flaky coverage – my handset spends it's time jumping between 3G cells at home & rarely sticks on Orange GSM. If I disable 3G then I get great in-building coverage via Orange GSM! Just have to remember to enable 3G when I go out, to get decent data rates. However that sort of stuff is not Normob friendly!

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.