Connecting the Unconnected
Mobile phone technology has been successful. When the first deployments started in the 1980s, the mobile phone was more of a luxury than a new Mercedes Benz Car. Now we take mobile phones for granted.
According to the CIA Fact book, (the font of all knowledge) there were 7,323,187,457 people on the planet in July 2016 and there were 7 billion mobile phone contracts. But this doesn’t tell the whole story.
Parts of the world have no mobile phone service. There may be billions of people who have no access to any kind of communications: no phone of any kind, no internet, nothing.
Access to communications is not just good for the users, but it speeds the development of an economy.
Conventional Radio is designed to serve affluent urban and suburban areas, and it does this very well. For rural markets, the business case for deploying it just doesn’t stack up.
The challenge is to provide communications (probably Mobile phone service) to as many people as possible.
In the past, we thought adapted large conventional cells were the answer. They provide the cheapest contiguous coverage. While this is the essence of the mobile phone, contiguous coverage is an expensive luxury. Is there a need to provide coverage in a wilderness?
Large cells require high towers, large bandwidth backhaul, large amounts of power, air-conditioning and proper shelters with foundations. Contiguous coverage implies complicated operation for handovers and management of interference. This makes the equipment complex and requires costly planning and optimization.
Having access to a connection is more important, even if the user can only use the phone in the centre of his village.
If we are prepared to give up contiguous coverage, small cells offer a way forward. Properly placed, they can concentrate coverage where facilities and customers are: the simplest and most cost-effective connectivity per user.
What about other aspects?
Backhaul is a thorny enough subject for regular deployments, but for these ultra-rural ones it becomes critical. In basic terms, backhaul is quantified by length and breadth. How do we connect even a small BTS that’s distant from any existing point of presence? We may need many hops of microwave. Each site will require a tower, equipment, power…. How about fibre? We’d have to dig a trench all the way. We could consider Satellite backhaul. That doesn’t need us to deploy a lot of equipment, but will be expensive especially for large bandwidths.
High speed data (or even UMTS and LTE) needs abundant bandwidth. GSM’s requirements are more modest. Unless there is point-of-presence nearby, our rural-coverage we’ll use GSM with satellite backhaul.
Once we thought that the cost and availability of handsets would limit ultra-rural deployments. There simply wouldn’t be anyone to use the coverage we have provided. Now there is reliable supply of handsets. Some new but many grey market and used are cheap and readily available.
A standard network has architecture defined by 3GPP. Some of the function is needed for basic ultra-rural and some not. Moreover, the network elements, and the interfaces between them, were standardized several decades ago. As a result expensive, often bespoke, hardware was specified. It would nice not to rely on structure as old as these.
Licensing and Spectrum
The model for licensing cellular is the a country leases some spectrum to a mobile network operator. The MNO gets a monopoly for the use of the spectrum in return for money. the MNO operators are not interested in covering rural areas which they have deem would yield low returns, and so don’t cover them. Would they object provided coverage in these places? That will have to be seen. It becomes a political and legal issue rather than a technical one.
A new model
If Mobile Network Operators can’t make the business case for ultra-rural coverage work, perhaps we need to go back to basics, and have another look at the requirements:
There are two alternative models.
This allows fully 3GPP-compliant architecture. The coverage is provided by the entity licensed to to provide it. MNOs typically don’t want to provide coverage to rural, low-ARPU regions. The costs are too high and the potential revenue is too low. Calls are routed through a complicated legacy core equipment.
Let’s call any provider who isn’t an MNO an ISP. He won’t assume he’s providing mobile service to an urban population and he won’t have legacy equipment.
Specialized Radio Access will be key to connecting the unconnected.
Whether the coverage will be provided by conventional Mobile Network Operators (the OPCOs) or someone else, remains to be seen.