A potential breakthrough on virtualization of the radio access network (vRAN) could fast make it the hottest topic in mobile telecom.
But it's a breakthrough that may owe little to the traditional equipment vendors, or to groups like the 3rd Generation Partnership Project (3GPP) that have traditionally represented them. In fact, they might not even want you to know it is happening.
Instead, the plaudits will go to Facebook's highly disruptive Telecom Infra Project (TIP) and an equally disruptive vRAN startup called Phluido, which has emerged as the driving force behind a new TIP working group on vRAN.
So what is vRAN, exactly, and why all the fuss about it?
In most of today's 4G networks, the baseband processors that manage radio functions are located at the radio sites themselves. But what if you could take those processors and move them into a data center? You could run your entire RAN off any centralized X86 server and turn your distributed radio sites into a bunch of dumb, inexpensive radio units.
It's a transformation that could hasten the demise of the "black box" -- the dedicated bit of kit that vendors such as Ericsson AB (Nasdaq: ERIC) and Nokia Corp. (NYSE: NOK) have traditionally flogged to their customers -- and make operators look more like high-performance athletes than infrequent park joggers. It could even improve the economics of network rollout amid concern about the cost implications of 5G.
But despite its attractions, there has been relatively limited action on the vRAN front in the 4G era. The big hurdle, just about everyone agrees, is a constraint in the "fronthaul" part of the network -- that is, the connection between the baseband processor you're sticking in a data center and the antenna you’re leaving at the radio site.
Historically, a protocol known as the common public radio interface (CPRI) has been used to support that connection. But CPRI places huge demands on the underlying transport infrastructure. It basically means you need to run a whole lot of fiber out to individual cell sites, and that means spending a sizeable chunk of change. It pretty much rules out vRAN for many operators. (See What the [Bleep] Is Fronthaul?)
That's where Phluido comes in. Founded in early 2014, the small company has developed a proprietary fronthaul technology that compresses data and allows it to be carried over much lower-cost transport links, including Ethernet and WiFi connections. If Phluido's technology lives up to that promise, vRAN suddenly becomes a very economical option for all sorts of players.
Since it first set out, Phluido has made some progress on licensing its fronthaul technology and vRAN software stack to other vendors. But when Facebook launched TIP in early 2016, it quickly grabbed the attention of Alan Barbieri, Phluido's CEO. Aimed at spurring network innovation, TIP seemed like an opportunity for a startup to bring its technology to the attention of a much broader community, and especially the mobile operators flocking toward TIP.
Yet at first, things didn't pan out as Barbieri had hoped. "We felt that everyone was waiting for something to happen," he tells Light Reading. "The key change happened about three months ago when we took the initiative and formed a new working group on vRAN fronthaul."
Unlike the Facebook-led OpenCellular group, which says its ultimate goal is to produce an "open source" wireless access platform, the vRAN fronthaul group looks fully committed to an approach based on RAND licensing (or the licensing of intellectual property based on "fair and non-discriminatory" terms). "It is not about open source, but it is about unbundling and open APIs [application programming interfaces] and the ability for small companies to be a part of the ecosystem," says Barbieri.
Not just small companies, though. And that's one thing that makes the vRAN fronthaul group so interesting. Its chair is UK telecom incumbent BT Group plc (NYSE: BT; London: BTA), while its co-chair is mobile giant Vodafone Group plc (NYSE: VOD), which has previously carried out trials of Phluido's technology. Another service provider involved with it is Telecom Italia (TIM) . That is an impressive lineup, but "tens" of other companies, whose identities have not been revealed, also appear to be group members.
Phluido doesn’t want to lose control of its technology, and it has zero interest in open sourcing its intellectual property. That means steering a fine line between exposing interfaces and ensuring the "secret sauce remains 100% the property of Phluido," says Barbieri.
Next page: Taking care of standards business
Taking care of standards business
What makes it all worthwhile is the group's goal of coming up with new open specifications that would support standards-based vRAN deployments in both 4G and 5G networks. "We want a lower PHY [physical layer] split of the RAN, which is something the 3GPP is not currently taking care of," says Barbieri. "With the power of the companies backing up the group, the hope is this will become the de facto standard."
At some point, the TIP group may have to start coordinating its activities with the 3GPP's, acknowledges Barbieri, to ensure there is alignment on 5G-related work. But whether or not Facebook's big network initiative gives rise to a future vRAN standard, the momentum behind the technology is undeniable.
Just a few days ago, another vRAN player called Amarisoft became one of four startups that Orange and Facebook said they would nurture as part of a new incubator program, after considering applications from a total of 22 companies. "One of the reasons we did this [TIP] was to hear from mobile operators what their needs are and we hear that vRAN is very high on the list," said Steve Jarrett, Facebook's head of infrastructure partnerships for Europe and the Middle East, during an investor briefing at Orange's offices in Paris. (See Orange, VCs Commit $113M to Network Startups as 'Black Box' Frustration Mounts.)
Germany's Deutsche Telekom AG (NYSE: DT), which is heavily involved with TIP, has also talked up the attractions of RAN virtualization from a 5G perspective, even suggesting it could lower the costs of deploying 5G networks. And during previous network trials by China Mobile Ltd. (NYSE: CHL), operating expenditure, capital expenditure and power consumption fell by 50%, 15% and 71% respectively compared with "respective values in a traditional mobile system," according to a 2015 research paper from China's Chongqing University. (See The Growing Pains of 5G.)
What also seems irrefutable is that vRAN will pile further pressure onto equipment vendors already wrestling with the consequences of the software and virtualization revolution. "There are the usual big incumbents that are a little resistant to these changes as you can imagine, but there is a trend from mobile operators to break this oligopoly and have more flexibility to cherry-pick from different vendors," says Barbieri.
Indeed, Orange has even suggested that Amarisoft could in theory replace Ericsson in national networks if its technology can be appropriately "scaled." (See Startup Could Replace Ericsson, Says Orange.)
Bruno Jacobfeuerborn, Deutsche Telekom's chief technology officer, has offered perhaps the bluntest assessment of the upheaval that is underway. "This will change the landscape and some vendors will disappear and some new ones will come into the game," he said during a conversation with Light Reading about 5G and virtualization earlier this year. "It is going to be a disruptive moment in time."
— Iain Morris, , News Editor, Light Reading