Whether the virtualised RAN (vRAN) will be adopted as a mainstream basis for 5G is an important question – but another question may need answering first: what form should the fronthaul link between the virtualised, centralised baseband unit (BBU) and the radio/antenna units at the cell sites take in order to enable the vRAN?
Whatever the standard protocol, fibre will be essential to most fronthaul architectures. A recent forecast from Rethink Technology Research indicates that long-range fronthaul links for macrocell vRANs will remain heavily fibre-based because of the capacity and low latency required to transfer traffic from cell site to a faraway cloud baseband. Wireless fronthaul will account for only 7% of long-range fronthaul links until 2023.
However, copper-based technologies like G.fast –which has been tested as a fronthaul approach by BT in the UK – will grow from 4% to about 16% of fronthaul connections between 2018 and 2022.
In early 2016, BT announced tests in the UK which showed copper lines handling the speeds necessary for mobile vRAN fronthaul achieving speeds of between 150Mbps and 200Mbps over G.fast connections within a vRAN environment, in the operator’s Adastral Park labs in Ipswich. At the time, Dr Tim Whitley, managing director for research and innovation at BT, touted the results as a major breakthrough for the future of mobile data networks.
“These technologies will play a key role in 4G networks and will be fundamental to 5G architectures. The trials are another step towards a fixed and mobile network which will support customers’ increasing demands for data,” he said. BT, which conducted its tests with chip provider Cavium, said back in October 2015 that it had reached speeds of 5Gbps over copper using an advanced version of G.fast, XG.fast.
And next year, G.mgfast (the new term for XG.fast) is set to be standardised by the ITU. There has been a resurgence of R&D in the segment, stimulated by the prospect of small cell backhaul for 5G. The plan is to double the 212 MHz baseband ceiling on copper twisted pair to 424 MHz first, and then to 848 MHz after that. We have already seen trials of aggregate (combined speed in both directions) throughput approaching 2 Gbps and this should boost this fivefold, and also add in full duplex mode (full bandwidth in both directions), making 10 Gbps each way.
This is a very ambitious re-write of the copper wire broadband standards predicated on the need to include copper in the 5G picture. One MNO CEO estimated that it would cost €400 billion to lay fibre in just one large European country (Germany) to reach all small cell locations, so a cheaper copper alternative would be attractive in the early days of dense 5G.
Could copper be a fronthaul front-runner for 5G vRAN? The xRAN Forum, among others, may have something to say about that, but it is certainly an interesting prospect.