Adding fibre to national networks
In line with its conviction that open access networks are the most effective model for achieving high speed broadband penetration, FibreCo Telecommunications is aiming to roll out an additional 1 800km of advanced fibre networking across South Africa’s southern coastal regions over the next two years, connecting Cape Town to Durban.
FibreCo Telecommunications, a joint venture between Cell C, Internet Solutions and businessman Andile Ngcaba’s Convergence Partners, builds and operates open access fibre networks across South Africa, and plans to extend its operations to develop terrestrial links across the continent within the next few years.
Chief executive, Arif Hussain, says the open access model makes advanced network rollout viable for operators such as FibreCo and its partners, and as such, is a model that would help assure universal access for the country. With more than 2 000 km of fibre network now connecting Cape Town to Johannesburg and East London to Bloemfontein, Hussain says work on the connection along the coastal route from Cape Town to Durban is expected to begin towards the end of this year.
With a core ring of fibre connecting South Africa’s major cities, FibreCo believes that extending this state of the art network to reach smaller centres along the major routes will become increasingly feasible. “It’s important to note that not all fibre is equal,” he says. “It may appear that much of the country is connected to a fibre network, but unless that network is of the right quality, it will not deliver the 100Gbps connectivity we hope to see in South Africa.”
Likening the fibre networks to roads, Hussain says one might feel a town connected by a sand track is connected, however reaching it would not be quite as easy as it would be using a freeway. With fibre connectivity widely accepted as the best solution for high-speed connectivity, Hussain says the belief that fibre is too costly to reach smaller and rural areas is a misplaced one, to some degree. “With fibre access, the biggest inhibitor to investment is demonstrating demand,” he says.
“The real challenge is defining the business case that makes it viable to extend a network to a smaller or rural area. Open access encourages a model where the cost of the infrastructure is split between as many customers as possible, therefore where the customer demand is aggregated, rollout becomes viable.”
He points out that by committing its regular connectivity spend to a private open access network operator for a set term, government can help make advanced fibre network infrastructure development into rural areas a viable proposition, and so come closer to achieving its access goals without duplicating efforts or spending additional public funds. This scenario would go some way to closing the existing network gaps, Hussain believes. He notes that Telkom has about 145000km of fibre networks, Broadband Infraco has about 14000km and other players, such as FibreCo and other telecommunications operators, have about 5000km of open access long-distance networks in total.
“So there is a lot of fibre in place. There are areas where there are gaps — such as the southern coastal route and the north-west to north-east of the country, and there are some areas too remote for fibre to be viable.” According to research published by the Department of Communications, more than 86% of the population is within 10km of a fibre node. Hussain notes that connecting more households to these nodes will depend not only on fibre, but also on other technologies, such as long-term evolution (LTE).
However, maximising the potential of LTE mobile connectivity depends on freeing up of the necessary spectrum. “My personal view of the spectrum debate is that the vested interests of stakeholders make it difficult to resolve the debate. You see the same situation around the world, where on the one hand, there is the option of releasing spectrum at a low cost with the hope that this will make services to consumers more affordable, while on the other hand, there is the option of using the release of spectrum to generate additional revenue from operators to boost the fiscus.
"There are good arguments in favour of either side. But I believe that picking any course of action, which results in LTE spectrum becoming available to operators in the near future might be better than to keep delaying universal access through ongoing debate.”
High speed universal access
Achieving meaningful universal access means more than just connecting homes to the internet — access needs to be delivered at speeds appropriate for the use of popular applications. And work to enable this has to start now, says Reshaad Sha, chief strategy officer at Dark Fibre Africa (DFA).
Sha notes that five years ago, 2Mbps connectivity was sufficient for running the consumer applications of the day. “But today, the applications available to consumers, such as those that utilise rich media, Facebook, YouTube, Instagram, video on demand and more, require connection speeds of at least 10Mbps. If this trend continues, and we expect that it will, we can expect to need a minimum of 100Mbps to run popular applications five years from now.
"Delivering on the 2020 vision demands that the connections rolled out to households allow them to use all the apps of the day. Delivering 100Mbps connectivity could be achieved through a combination of fibre and wireless technology, such as long-term evolution (LTE). But the rollout of LTE needs to start on a large scale now, to bring down the cost of the technology over the coming years, and make it viable for rolling out to rural areas five years from now.”
DFA installs open access fibre networks in major metros and secondary cities, building infrastructure that is then shared by telecommunications and internet service providers. With 7800km of infrastructure in the ground at a cost of approximately R5-billion to date, DFA is still expanding its network. Sha believes that with the long-haul networks put in place by the likes of Broadband Infraco, Fibreco and Telkom, and metro networks rapidly being expanded across the country by the likes of DFA, much of the groundwork has already been laid to support a universal access programme.
However, fibre is not the answer for rural connectivity, he notes. At a cost of up to R800 per metre to lay fibre, and an associated and ongoing maintenance cost, rolling fibre out to rural communities is simply not economically viable. “Today fibre to the home is not viable even in traditional suburban areas in South Africa, due to the distance between houses,” Sha explains. “It only becomes truly viable in high density residential settings, and in new estates under development, where the cost of ducting becomes almost negligible.”
Connecting rural areas will need to involve a combination of technologies, with wireless playing an important role, says Sha. “But a major impediment telcos face is that they don’t have access to the spectrum they need. Spectrum needs to be allocated as a matter of urgency and we need to start delivering LTE to major metros now, so that when the time comes to extend it to rural and outlying areas, the cost of doing so has been reduced, much as we saw with 3G.”
Sha believes the latest national Broadband Policy includes the necessary principles to support universal access, such as open access models and the principles of non-duplication of infrastructure.
“However, with my background working with broadband policy analysis in other emerging markets, I believe South Africa is a little late in stepping up to address access. Whether the 2020 vision will be achieved depends to a large degree on the political will to get it done, and an environment that supports private sector investment in the necessary infrastructure,” he says.
This article has been paid for and signed off by KPMG and its business partners.