Mallinson: 5G to eclipse but not eliminate fiber

Keith Mallinson

Prevailing opinion on whether wireless is complementary or competitive to wireline access in telecommunications has swung back and forth over recent decades as performance demand and supply capabilities of wireless and wireline technologies have increased at varying rates. Networks have been constructed accordingly, including large investments in various types of access technology over copper, wireless and fiber connections. Wireless access appears to be on the ascendancy, once again.

This could be a defining moment for 5G with potential technical and economic advantages in wireless over fiber for fixed as well as mobile access. Verizon and Alphabet have recently indicated that wireless can match the performance of fiber for connecting homes at significantly lower cost. As I predicted here a year ago, existing fixed access networks might be significantly marginalized by 5G networks, while network densification, speed and capacity increases with 5G will require lots more fiber deployment regardless.

30 years of access

When consumer telecommunications was all about voice, copper wireline access was originally the only game in town; however, by the mid 1990s 2G wireless was decidedly cheaper to deploy for boosting teledensity in developing regions such Eastern Europe and subsequently in the BRIC nations and their neighbors. The only snag was that by the end of that decade a large and increasing proportion of consumers everywhere wanted Internet access. Dial-up modems on landlines brought email and Internet access to the masses' PCs while circuit-switched data and then GPRS from around the turn of the millennium was not up to muster for that. And then ADSL and cable modems soon made fixed broadband access with megabit-per-second speeds available to a large proportion of households. Cellular remained significantly behind, with "best efforts" of no more than few hundred kilobits per second at most until mobile broadband with CDMA2000 EV-DO and then HDSPA came along in the mid 2000s. This revolutionized use of small devices while on the move but had minimal impact on fixed usage.

Wireless was progressing but continued to be outpaced by improvements in wireline access technologies while consumers demanded ever-bandwidth-hungrier applications and services at home. Mobile broadband started with niche demand for PC data cards and dongles and first received widespread consumer uptake with the iPhone 3G and Android devices from 2008. Nevertheless performance expensively paled in comparison to that of wired access until quite recently bar a few exceptions; such as in rural areas a long way from central offices, or in the most innovative and over-competitive cellular markets around the world. Clearwire's fixed-wireless service proposition with WiMAX around 2010 was economically unsustainable versus competing DSL and cable modem-based services. The company failed with fixed and mobile users suffering poor service quality from the network overload including significant PC usage. Throttling down to hundreds of kilobits per second was unsatisfactory for fixed users in particular while DSL and cable modem services were rarely, if ever, restricted in that way.  Copper, and in a small proportion of places fiber, continues almost invariably to be used to connect homes. That has been required to satisfy requirements for high-speed and heavy use including SD and HD video streaming to TVs, PCs and tablets as well as smartphones, even though Wi-Fi is commonly used to connect within the home.

However, cellular wireless is imminently catching up with fiber-speed capabilities and the demands on it for services used on fixed and well as mobile devices. As I indicated here last time, ever-faster mobile broadband is the most significant leg of development in the path from 4G to 5G. Fiber has more speed and capacity headroom, but for the time being, consensus appears to be that around a gigabit per second per household is adequate. That is within reach for 5G. In the last few weeks, both Verizon and Alphabet have indicated that wireless connections to the curb can provide the performance and the economics required for continuing rollout of the highest-speed capabilities to the home.  Verizon's CEO Lowell McAdam has stated that fixed deployment of 5G technologies 'gives you all the return on capital that you need.' Similarly, Google's parent Alphabet is looking to wirelessly connect homes on the basis that Chairman Eric Schmidt says wireless technologies are less expensive than laying cables with various possibilities being investigated.

Fiber buried but not dead

Fiber manufacturers and trench-digging equipment vendors need not worry about these developments. Fiber will continue to get ever-closer to most users, but in most cases it will just stop short of making the final connection from the curb to the home. While it is becoming less likely that yards will be dug up to replace copper with fiber, there will nevertheless be an increasing demand for ducts and poles to carry the fiber to the wireless antennas and provide the positioning for the latter so they can deliver these gigabit services to our homes. Propagation constraints with the centimeter and millimeter wavelength spectrum envisaged for 5G, as well as increasing user demands for data, mean that network antennas need to be ever more numerous and closer to users.

Keith Mallinson is a leading industry expert, analyst and consultant. Solving business problems in wireless and mobile communications, he founded consulting firm WiseHarbor in 2007. Find WiseHarbor on Twitter @WiseHarbor.

Mallinson: 5G to eclipse but not eliminate fiber

Suggested Articles

Crown Castle could be poised to narrow the performance gap in coming years between itself and tower REIT peers, according to MoffettNathanson.

Michael Thelander, president of Signals Research Group, said that he expected Chinese operators to be the first to announce standalone NR services.

The likelihood that either side in the T-Mobile/Sprint case decide to appeal? Low, according to New Street Research, in part due to spectrum.