The Facebook-driven Telecom Infra Project (TIP) held its third annual Summit this week, for the first time outside Silicon Valley. In a well-attended and lively gathering in London, the TIP community made several announcements of significance for those who seek to disaggregate the telecoms network, and therefore disrupt the established order of suppliers and drive down cost.
There were new working groups, including one focused on edge compute; a rapprochement with the Linux Foundation’s Open RAN (ORAN) Alliance; and recognition of the vendors which had best responded to the first TIP-based RFIs (requests for information), issued earlier this year by the OpenRAN Project Group.
Facebook more aligned to WiFi than cellular community:
The proceedings were heavily focused on the wireless network, and especially on the looming prospect of 5G, and here is where doubts creep in about TIP. Like the Linux Foundation’s various open carrier network projects, TIP is heavily supported by some flagship operators, looking to drive down the cost of 5G through open platforms. However, the technologies which are closest to market are not cellular or 5G, but have their roots in the WiFi/WiGig community.
Of course, it makes sense that unlicensed spectrum – in this case, 60 GHz – should be at the heart of an open networking effort, but there is no prospect of 5G radios running, in a standardized way, in shared spectrum for several years (Qualcomm is sponsoring a 5G-Unlicensed activity in 3GPP, but it may not even make the next re-lease of standards, Release 16).
That means that the 3GPP bands and radio technologies remain separate from those in unlicensed spectrum, whether the IEEE’s WiFi and WiGig, or TIP’s Terragraph 60 GHz reference platform. In the 4G era, the boundaries had started to blur because of the various LTE implementations in unlicensed bands, but in the early years of 5G, the communities are standing apart again.
The risk for TIP, and other open networking efforts which focus on shared spectrum, is that they will be sidelined by the cellular juggernaut. Initiatives like ORAN, and its predecessor xRAN, focus on cellular radios and licensed bands, but seek to encourage open architectures and interfaces within those. The aim is to disrupt the ecosystem surrounding the 3GPP standards, not create an alternative to them.
TIP is developing interfaces and specifications, in areas like virtualized RAN fronthaul, which could be applied to 3GPP environments. Though several commentators have likened it to WiMAX, it is not overtly trying to posit an alternative to 3GPP standards for the next wave of wireless broadband.
But the London Summit did reveal its schizophrenic nature. On the one hand, it is supported by traditional operators like Vodafone, Telefonica, Deutsche Telekom, BT and SK Telecom, which are interested in developing or incubating new, low cost solutions for the cellular network (and for wireline – TIP is not specific to 5G). They dominate many of the project groups, including the important OpenRAN.
On the other, the focus was very heavily on the 60 GHz band, and Terragraph in particular, and on the potential of such technologies to deliver broadband to the underserved – in line with Facebook’s Internet.org program – and also to dense urban environments, supporting either access or small cell mesh backhaul. This harks back to the first days of TIP, before the big telcos signed up, when Facebook set the ball rolling by contributing its own OpenCellular small cell and Terragraph designs to open source, providing specifications for vendors to build low cost, open access points, MIMO antennas and millimeter wave radios.
That aspect of TIP’s work fits with the ambitions of Facebook, Google and Microsoft to drive a low cost, broadband platform which can bring their services to far more people, far more quickly and cheaply – bypassing the established operators if that helps.