ROME -- Executive Summit -- Standardized 5G networks are on course for a proper commercial launch in 2020 thanks largely to a push from pioneering operators like Verizon and South Korea's KT. Yet there are still concerns about the fragmentation that could occur as different players jostle for leadership.
Aggressive launch plans by Verizon Communications Inc. (NYSE: VZ) and KT Corp. , both of which aim to bring some form of 5G service to market in the next two years, have provided a major spur to the industry this year, says Gabriel Brown, a senior analyst at Heavy Reading . "They are trying to get ahead of the market as a whole but that is pushing everyone else along," said Brown, during a presentation at Light Reading's Executive Summit in Rome earlier today. (See Verizon CFO: Eat Our (Fixed) 5G Dust!)
The moves by Verizon and KT, which are basing their initial plans around the use of 28GHz spectrum, have prompted a reaction from companies including AT&T Inc. (NYSE: T) and Ericsson AB (Nasdaq: ERIC), which are keen to ensure that a unified global standard eventually takes shape, says Brown. Japan's NTT DoCoMo Inc. (NYSE: DCM), whose efforts have been vital to 5G development so far, similarly wants to avoid any industry fragmentation along regional lines, he says.
An additional risk comes from China, which could decide to press ahead with its own take on 5G -- as happened in the era of 3G networks. "The Chinese are now at the point of having the biggest 4G networks in the world and there is an immense hunger for R&D," says Brown. "What if the Chinese block goes ahead with its own version of 5G?" (See China Mobile Stamps Intentions on 5G .)
Despite the concerns, the Heavy Reading analyst is optimistic that China will not develop its own flavor of 5G and that a global standard based on an OFDM (orthogonal frequency division multiplexing) air interface will be introduced commercially in 2020. This "Phase 1" specification, which will be included in the 3GPP's Release 15, will also feature more capability than was originally envisaged, he says. "It was going to be a bare bones release but it now looks like there will be a systems design [in Release 15]," says Brown.
A major boost to the introduction of 5G recently came from chipmaker Qualcomm Inc. (Nasdaq: QCOM), which unveiled its first 5G modem in October. Expected to start shipping in 2018, this seems aimed at supporting the 28GHz-based use cases that Verizon and KT are initially targeting. Yet Qualcomm will be as eager to guard against standards fragmentation as anyone. "Qualcomm very much benefits from having a global standard and it does not want to have more limited regional markets," says Brown. (See Oh Snap! Qualcomm Unveils X50, Its First 5G Modem.)
Among other things, there is also more clarity on the spectrum that will be used to support 5G services. While much of the attention is focused on the millimeter wave spectrum that will support the highest-speed services, lower bands will have an important role to play, says Brown. Those include the 700MHz spectrum that will become available in Europe at the same time 5G is being commercialized, as well as spectrum in the 3.4GHz to 3.8GHz ranges.
According to Brown, these mid-band airwaves could fill the need for a global 5G band. "With advances in antenna processing you will be able to deploy mid-band systems without having to make huge investments in towers and concrete to get it going," he says. "That's down to advances in MIMO processing in the last few years."
Even so, while players in the US, Japanese and South Korean markets plan to use 28GHz spectrum to support gigabit-speed mobile services, those airwaves will not be available in Europe. Operators there may have access to 32GHz frequencies, says Brown, but they will not be able to take advantage of the same chipsets that Verizon and KT are using. (See Spectrum Hurdle Could Trip Europe in 5G Race.)
Another fragmentation issue is whether operators go with the non-standalone or standalone mode of 5G. The former uses the LTE core network and radio to provide the control plane and to set up and manage sessions, while the latter would require a next-generation core. "The idea of non-standalone mode is very attractive and that has been the lead model for most operators," says Brown, while noting that support for standalone has been growing. Non-standalone specifications are expected to appear at the end of 2017, with standalone ones due in mid-2018.
The concept of network slicing is also provoking a few jitters. Through a core technology called the network slice selection function, operators hope to be able to provide a number of differentiated services on the same 5G system -- a low-latency service for the healthcare sector, for example, and a high-bandwidth one for car makers. "The big challenge is how to operate this across an entire end-to-end infrastructure," says Brown. "You have radio, transport and backhaul and the core network, and to give operators the best chance to serve adjacent industries you need a way for slicing to work across everything."
— Iain Morris, , News Editor, Light Reading