The BBC conducted a survey of 3G coverage in the UK and recently published their results. They do warn it’s not completely scientific and that the data should be interpreted with caution, but it does point out shortcomings in 3G coverage around the country. The test used a smartphone application to measure and report 3G signal availability, with results collated centrally. Data testing firm Epitiro, who also supply wireline broadband performance measuring systems, provided the system itself. 44,600 volunteers  ran the application and contributed to the study, which operated across all UK networks. I expect similar results might be observed in many other countries.

One shortcoming of the survey was that signal quality or data throughput wasn’t recorded – only the presence and availability of a 3G service was measured. While the major cities show excellent 3G coverage almost everywhere, I suspect the data performance would be constrained in some areas by congestion.

Nonetheless, it highlighted major shortcomings in access to a 3G data service from all UK networks – a signal was only available 75% of the time. Some of the worst places are on the road and rail networks, where fast moving traffic may be difficult (ie expensive) to address. 

Two different problems to solve

For me, the study highlighted one clear issue but didn’t really show another.  It confirmed a problem with data coverage, but missed the effects of data capacity. These are two separate issues that both need to be addressed.

1: Data Coverage

It did show that a 3G service  simply isn’t available in many parts of the country, typically the more rural and remote parts. This is partly because 3G operates at a frequency of 1900MHz compared with 2G that (mostly) runs at 900MHz – the higher frequency reduces the range of the cellsite coverage area considerably. This dramatically increases network costs because the operator has to install two or three times the number of expensive cellsites to cover these wide areas as effectively.

One cost effective solution to this is to recognise that more mobile data service would be used in population centres rather than completely in the wilderness. The cost of providing local cellsite coverage to smaller and smaller population centres (i.e. small villages) has now become ever cheaper, helped by the availability of wireline broadband service in these areas combined with the low cost and practical availability of small public outdoor femtocells. We discussed this further in our recent article called “Femtocells and the (Digital) Village Pump”. Large numbers of these small cells in each hamlet and village would serve the needs of mobile data users, relieving difficult demands on the wide area macrocells in the area and increasing their performance for those who really need it.

2: Data Capacity

Unfortunately the test didn’t measure actual throughput – the data rate achieved when using other applications such as email, web browsing and other updates. I can accept that this would be a complex test, taking into consideration whether the destination server or other internet issues are restricting data throughput. However, my own experience is that frequently the availability of a 3G signal gives little confidence of any reasonable data performance.

Again, the technical solution here has to involve larger numbers of smaller femtocells, such as the so-called metro-femto and picocells. By taking this approach, mobile phones are closer to the cellsite and so benefit from higher data quality and speeds (and longer battery life), while the economics of deploying more/smaller cells has been shown to be beneficial to the network operator. In areas where backhaul is available (either wireless or wireline), the timeframe to deploy these solutions can be much shorter – there are often few of the planning restrictions or neighbour concerns found with much larger installations.

Summary

So while the survey told us what we already know – that 3G signal coverage is far from ubiquitous - it didn’t tell the full story about the lack of data capacity and effects on daily usage in many parts of the network.

For customers, there is a need better information and measurement tools to compare and contrast the likely service we would get from each network in our area. For network operators, femtocells and small cells provide clear solutions in both these aspects.

A full report about the survey can be found on the BBC website here.