More than half the world's population does not have Internet service, according to an ITU report published in July 2016.
In the global effort to connect the unconnected, companies have proposed a wide range of schemes to help bridge this divide: Drones from Facebook Connectivity Lab and balloons from Google X's Project Loon are two of the more eye-catching initiatives. These are interesting, and potentially important, projects that aim to fundamentally change the economics associated with wide-area coverage.
But could part of the solution lie closer to home?
One option is to deploy existing mobile broadband technologies. Because they are later in the development cycle, equipment costs for 3G and 4G networks are now much lower and there is a wealth of accumulated experience in how to deploy and operate these networks, at low cost, in many different environments.
4G LTE is in many ways an attractive option. Performance is good relative to 3G, low-cost handsets are increasingly available and the skills to optimize and operate LTE are now also much more available. All things being equal, an operator would generally deploy 4G over 3G. There are some markets, however, where 3G might be a better option due to the handset cost and spectrum allocation. In many unconnected regions, a proportion of the subscriber base -- by some estimates this could be as high as 20% -- already owns a 3G handset.
Spectrum is also a critical issue in some emerging markets. Operators often only have a limited allocation, which is currently used for GSM and cannot easily acquire new spectrum for mobile broadband.
LTE uses 10 MHz or 20 MHz of spectrum (and occasionally 5 MHz), which in spectrum-rich markets is not a problem. In some emerging market countries, however, operators are often much more restricted and have only a small amount of GSM spectrum available -- typically a 5 MHz to 6 MHz allocation at 900 MHz. This frequency has good coverage characteristics and is widely supported in devices. If it could be used for mobile data, operators and their customers could benefit significantly.
These operators can't countenance switching off GSM and refarming it for 3G or LTE when so many customers are reliant on it. One interesting proposal, therefore, is to develop solutions that can run 3G in a 5 MHz to 6 MHz GSM channel such that both GSM and 3G can operate simultaneously in the same frequency band. This means legacy GSM devices continue to be supported alongside 3G services. Typically, the operator would look to introduce such a capability when it is ready to refresh or upgrade its GSM radio network.
The ability to run 3G and GSM simultaneously in 5 MHz to 6 MHz of spectrum offers a high degree of flexibility in network capacity allocation and will help transition users from voice to data communication and bring the benefits of Internet connectivity to more communities.
You won't get a Gigabit with 3G, but you will get access to critical information, social networks, entertainment and the worldwide web.
-- Gabriel Brown, Senior Analyst, Heavy Reading
This blog is sponsored by Huawei