The survival of the mobile-only business model is looking shaky and that is driving MNOs to seek fixed-line acquisitions and partnerships – mainly based around fiber, but with some operators also placing their hopes on 5G fixed wireless access (FWA) as a way to deliver fixed/mobile and quad play bundles, at lower capex cost.
In the UK, mobile-only operators like Telefonica O2 and Three say they are evaluating FWA, while infrastructure provider Arqiva recently acquired millimeter wave spectrum in the 28 GHz band in London, and carried out demonstrations of a potential wholesale 5G fixed network, working with Samsung.
But the main concern of MNOs in the country is to have an alternative to Openreach – BT’s wholesale arm – when it comes to fiber. Not only is Openreach the largest source of fixed lines to rent for backhaul, fronthaul or access, but it is owned by the same company as the largest MNO, EE. Regulator Ofcom has recently taken various steps to ensure the businesses are at arms’ length from one another, but companies like Vodafone remain resentful of their reliance on a rival’s network.
This is opening up opportunities for alternative wholesale fiber providers, from the leading cableco, Virgin Media, to more specialized concerns like CityFibre. Last week, Vodafone announced a strategic deal with CityFibre to reduce its need to work with Openreach. It will fund the smaller firm’s expansion in return for exclusive access to its fiber for consumer broadband purposes. The pair will jointly invest £500m in the initial phase of the project, which will take gigabit broadband to one million premises by 2021. Vodafone will guarantee to lease at least 20% of the lines and will have exclusive access during the build phase.
This could help Vodafone build up a consumer quad play offering – and support backhaul for small cells – without the expense of buying a cable operator, as it has in Germany (where it acquired Kabel Deutschland) and other markets. In the UK, it did buy Cable & Wireless Worldwide, but that is focused mainly on the enterprise market and has limited residential coverage. There have been persistent rumors of a merger or joint venture between Vodafone and Liberty Global, owner of Virgin Media (the two companies have a JV in The Netherlands), but that would entail regulatory scrutiny and complexities in execution. Adding other partners to the OpenReach deal could fill the gap more simply.
For CityFibre, it will bring the means to accelerate its city-by-city roll-out and fund its planned expansion into consumer fiber – its current business, which will still be open to other customers, is focused on backhaul, fronthaul and city networks.
Vodafone says the new alliance will bring fiber-to-the-premises (FTTP) to one million UK homes by 2021 and 5m by 2025. The network will be built and owned by CityFibre but Vodafone will have an undisclosed period of wholesale exclusivity and will commit to taking at least 20% of the connections. CityFibre is spending between £500m and £600m in the first phase of its 12-city ‘rolling build’ and has identified 50 further cities to be built out.
“Construction for the initial phase will commence in 2018, peak in 2020 and is scheduled to be complete within four years,” said Greg Mesch, CEO of CityFibre. “We have a framework in place to extend this to 5m homes, which would deliver 50% of the UK government’s target to deliver full fibre broadband to 10m premises.”
“The UK has fallen far behind the rest of the world, trapped by the limited choice available on legacy networks,” said Vodafone UK CEO Nick Jeffery. “We look forward to working with CityFibre to build the gigabit fiber network that the UK needs and deserves.”
He added that the deal “provides us with access to a superior product at a lower cost and with better service conditions than the regulated wholesale terms offered by the incumbent operator for access to its legacy copper telephone broadband network.”
Mesch added: “The time has finally come for the UK to join the more advanced markets of Europe in the full fiber league tables. The market has been held back for too long, with everybody being forced to consume from one major incumbent operator. If you look around the world, in every country where fiber has been rolled out, it’s always an alternate operator that starts the process.”
Meanwhile, the excitement around FWA as a likely first application for 5G – at least in the US – is being tempered by rising awareness of its limitations. It may be cheaper than pulling fiber, especially in rural areas, but research by Analysys Mason shows that its operating costs can often be higher (more maintenance, in particular, and ongoing site leasing).
And while FWA has traditionally been a rural play, there are also doubts about 5G’s suitability in this environment. Tod Sizer, VP of the wireless research program at Nokia Bell Labs, told Light Reading at its recent 5G Transport event that the technology is suitable for cities and dense suburban areas, but “the idea of this solving the rural problem is folly. There are too many trees.” Foliage is a major problem for signals in bands such as 26 GHz and 28 GHz, which are set to be commonly adopted high frequency bands for 5G worldwide. Verizon recently asked for an extension to its 28 GHz test licence but has not yet revealed many details of its trial results.
It is working with Nokia Bell Labs to improve performance in urban areas at least. The two organizations are co-developing a window-mounted radio to help the 28 GHz signal go through walls and windows more readily. Sizer said in a panel debate: “We can’t change the laws of physics. Millimeter wave doesn’t propagate well in buildings with concrete, or brick, walls, and covered in glass windows.”
The new transceiver sits outside a window and connects wirelessly to a WiFi router inside, which then distributes the signal around the home. Sizer said the first unit used a 2.6 GHz signal to connect the 28 GHz radio to the indoor unit, so that no hole in the wall or window was required, but a second iteration uses a higher performance optical connection.
More words of warning about the limitations of FWA as a fiber alternative came from CableLabs, the US cable sector’s R&D arm, in a new report co-authored with set-top box maker Arris. It is called ‘Can a Fixed Wireless Last 100m Connection Really Compete with a Wired Connection and Will 5G Really Enable this Opportunity?’ and examines a range of fixed 5G scenarios, using different bands (from 3.5 GHz CBRS to 70 GHz) as well as line of sight and non-line of sight deployments, and different types of walls or foliage.
Cablecos have been showing strong interest in 5G FWA, both as a weapon for potential rivals, but also as a way to extend their own fixed broadband reach cost-effectively. However, the report urges caution. “We have come a long way in the drive to 5G—but as the saying goes, there is still a long way to go,” it concludes. “As cable operators move fiber deeper, going to an all passive coax network, the ability to deliver multiple Gbps of capacity to a single home seems an easier path than building out a FWA millimeter wave architecture.”
The report highlights challenges such as interference, coverage, backhaul and deployment obstacles, which will offset the strong speeds and potentially capex efficiency of 5G. It predicts that fixed 5G might be helpful to support apartment blocks, or to provide a quick way to build out a new area until fiber could be added.
The report went on: “For FWA to be able to compete with wired solutions, it has to be able to provide Gbps peak rates on both the downlink and the uplink. To achieve these peak data rates in a wireless access network the only wireless bandwidth available with sufficient contiguous spectrum to meet 3+ Gbps service group downstream service—if the spectral efficiencies remain under 10bps/Hz—lies well into LOS-delivered (and near-millimeter wave) frequencies (28 GHz, 37 GHz, 39 GHz, 60 GHz and 64-71 GHz). These frequencies offer huge amounts of bandwidth (the unlicensed bands alone in the 60 GHz range can deliver 128 Gbps). But these frequencies present some clear technical, operational, and aesthetic challenges.”
According to the report, which cites fiber supplier Corning, it costs around $1,153 to deploy fiber to the average US household in a “dense” area, defined as about 880 households per square mile. That figure rises to $2,050 per household in areas with 175 households per square mile. Where there are only 72 households, it costs $2,499.
For 5G, in a 3.5 GHz deployment, a six-meter tower broadcasting a signal over 800 meters would cost roughly $3,750 per year plus around $300 for the receiver inside the home, calculate the authors. And in a millimeter wave deployment, a 30-meter tower broadcasting a signal over 200 meters would cost $26,700 per year plus around $650 for the receiver. The millimeter wave figures assume a taller tower to help avoid obstacles like trees.