FTTH/B will continue to see deployments as gigabit broadband becomes common currency
The Fiber Broadband Association recently reported that fiber to the home (FTTH) deployments in the U.S. hit a record high during the 12 months ended last September (see "U.S. FTTH deployments set record: Fiber Broadband Association"). The drive to deliver gigabit broadband shows no signs of slowing, so I would expect we'll continue to see a healthy rate of FTTH deployment this year – although perhaps not at record levels again. Google Fiber looks like it's slowing down, and Verizon is likely not to launch any major new initiatives (unless it signs another city deal like the one with Boston or its 5G infrastructure rollout presents new opportunities). Altice USA has trumpeted deployment of FTTH over DOCSIS 3.1, so perhaps their efforts (if they come to fruition) will make up for Google Fiber's likely slowdown.
10G deployments will ramp
Most of this will come in the form of XGS-PON, deployments of which have already begun (see, for example, "Call Flow deploys ADTRAN XGS-PON for 10-Gbps FTTH infrastructure"). Meanwhile, all eyes in the U.S. will be on Verizon and its stated desire to roll out NG-PON2. That effort awaits affordable tunable ONT transceivers – or a willingness on the part of Verizon to accept less than full capabilities.
Cable operators embrace "fiber deep"
Cable companies, particularly here in the Americas, will push fiber deeper into their hybrid fiber/coax (HFC) networks to support deployment of DOCSIS 3.1 capabilities. Meanwhile, they'll quietly continue to roll out FTTH in greenfield applications. One area to watch here is uptake of 10G EPON. There are plenty of sources for 10G EPON systems, and some of these platforms have reached the field, I'm told (again, albeit quietly).
Robust alternatives continue to hinder the drive to all-fiber access networks
Carriers (with the exception of Verizon) remain loathe to replace their existing copper networks unless they absolutely must. And technologies such as Gfast (particularly for in-building applications, but for outside plant as well) and DOCSIS 3.1 for coax promise to provide a pathway to gigabit and beyond that will help such operators put off fiber deployments in brownfield applications for a few more years. Meanwhile, fixed wireless will become a major force, particularly in urban and rural areas where fiber deployments are difficult, either technically or financially.
Software-defined access networks gain steam
The access network will be one of the places where software-defined networking and network functions virtualization (SDN/NFV) have a major near-term impact. This will be particularly true when it comes to next-generation PON deployments, with Verizon's NG-PON2 and AT&T's XGS-PON efforts salient examples in the U.S. Central Office Re-architected as a Datacenter (CORD) will prove influential as this trend plays out, I've been told.
"2018 will be a watershed year for commercial validation of Central Office Re-architected as a Datacenter (CORD) reference architectures. Leading network operators could begin commercializing gigabit broadband access, Carrier Ethernet, and metro transport services delivery using carrier-hardened versions of these open, disaggregated, and highly scalable CORD-based systems," Kurt Raaflaub, head of strategic solutions marketing at ADTRAN, wrote to me in an email late last year. "Further to this, operators will validate the ability to absorb provider edge and metro transport functionality within existing CORD components – evaluating the ability for transport I/O and management, subscriber policy management, and hierarchal QoS features and functionality to run on spare CORD compute and switching resources, eliminating separate dedicated platforms and management."