The last mile has always been the most expensive part of the service provider network. Network infrastructure must be ready to serve each individual customer home or business, and construction of that network forms the largest part of the capital budget.

Could the high cost of building a fiber connection to each home and business be avoided by instead using fixed wireless technology? Can it deliver comparable performance?

Let's take a look at the current state-of-the-art in fixed wireless and see if it measures up to fiber.

Fiber can carry signals dozens of miles without any active electronics in the field. It is unaffected by weather, objects in the path or signal interference. It requires no FCC licensing or spectrum allocations.

Fixed wireless is subject to all these concerns; repeaters are required for paths longer than a few miles (depending on the frequency used), weather (fog or rain) can attenuate signals, foliage can impair transmission and licensing is required for some spectrum.

Optical fiber can transmit virtually unlimited bandwidth. Current access technologies enable transmission of at least 10 Gbit/s in a PON, and using point-to-point technologies can extend to terabits per second.

Fixed wireless access is currently limited to 100 Mbit/s to 1 Gbit/s, depending upon the frequency and technology chosen. Higher frequency (millimeter wave) spectrum can transmit much faster. However, the range is shorter and more sensitive to path impairments. Millimeter wave has recently been demonstrated to provide gigabit connections without direct line-of-site. Look deeper and you will find that the connections were made via building reflections in an urban environment. This would not help at all in a rural setting.

Fiber network infrastructure includes the outside plant connection from the optical line terminal (OLT) to splitter cabinets and from the splitter cabinets to fiber ports ("taps" or "PDOs") provided for every potential customer location. Wireless network infrastructure includes tower locations and the backhaul network connecting them.

Both a fiber-served and wireless-served home or business have the same router, voice or video adapter, and Wi-Fi access point CPE requirements. The differences are in the network-connecting CPE. Fiber locations require a drop cable and an optical network terminal (ONT), and wireless locations require an antenna, antenna cable and radio.

Fiber infrastructure has a big disadvantage compared to fixed wireless: significant cost is incurred on a network-wide (per home passed) basis, as well as a cost for each customer served. Fixed wireless, on the other hand, has a network cost associated with each tower, but other costs are presented only per customer served.

If we limit our comparison to sub-gigabit speeds, the cost of equipment for a radio is somewhat higher for a fixed wireless home than for a fiber home. GPON ONTs can be obtained for far less than $100 each, while a radio, cable and antenna will likely range from $100 to $250 each. Installation costs for fixed wireless home and fiber home are similar.

Where does fiber make more sense?

Because of the shared infrastructure in a fiber network, the larger the number of customers per mile of network, the more "bang for the buck" the network will have. Terrain that limits the capabilities of a wireless network will mean that more tower locations will be required, making a wireless network far more expensive than it would be in wide-open flat areas. An aerial fiber network passing more than 20 homes per mile could be cost-effective versus a wireless network covering the same area.

Where does fixed wireless make more sense?

In open terrain, with few homes per mile, fixed wireless may be more cost-effective than fiber. Wireless can also work over water crossings or difficult terrain, or other locations where it is hard to get permits for fiber construction.

What about urban settings? Linking buildings in a city via high-speed millimeter wave could be attractive because of the short distances involved, and the ability to make use of reflections. These networks might compete with wired options that already exist in a city but have the same high costs of internally wiring buildings to consider. Lower-speed fixed wireless will not offer speeds that are attractive to customers in cities where much higher-speed wired products are available.

Do you need fiber anyway?

Someone once wrote, "all wireless networks are wired networks." This is true because the tower locations all must tie back to the operator's Internet connections. Backhaul from the towers (or groups of towers connected to each other wirelessly) is almost always via fiber, either owned by the operator or leased from someone else.

Here is where I provide a non-controversial opinion: Fixed wireless makes sense in rural areas. If connectivity at 50+ Mbit/s is not available from a telco or cable operator, fixed wireless (from a WISP) is probably the best option.

And a controversial one: The problem is that next-generation satellite broadband services, such as Starlink and OneWeb, also serve this same potential customer base and will not require construction of towers and their backhaul paths to do so. If satellite broadband works as promised by its purveyors, and if it can obtain the same government subsidies to serve rural customers as fiber and/or fixed wireless, it could easily become a formidable competitor anywhere wireline broadband does not go.

When customers require connectivity approaching 1 Gbit/s or higher, fixed wireless alternatives to fiber or HFC are not yet viable. Spectrum allocations that restrict the large chunks of bandwidth to millimeter wave limit the range, and therefore make it too difficult to use these technologies in rural applications.

Nothing beats fiber for access networks where there is a sufficient density of customers. In rural areas lacking that density, fixed wireless or satellite broadband will pick up the slack with speeds up to 50 or 100 Mbit/s. Rural connectivity at this speed via fixed wireless technologies can be cost-effective, with or without government subsidies.

Finally, satellite broadband services are not a threat to fiber, or even modern cable HFC wired networks. However, they may provide direct and formidable competition to rural fixed wireless services.

— Jack Burton, Principal, Broadband Success Partners

Jon Pelson's new book, Wireless Wars: China's Dangerous Domination of 5G and How We're Fighting Back, dives into the meteoric rise of Huawei on the global telecommunications stage.

Relying on yesterday's cost structures, routing silicon, operational paradigms, tools and two decades of accumulated complexity has proven to fall short in meeting insatiable bandwidth requirements consistently and at sustainable power consumption.

It would seem that we can continue the graph for Neptune's Law for the moment, and that will give our engineers of the present and future time to create the next generation of submarine cable innovation.

It's clear that bandwidth demand will continue to increase exponentially. Optical transport capacity must scale to meet this demand, while at the same time the cost-per- bit must decrease, and optical transport equipment must become more space- and power-efficient.

Four tips for overcoming involuntary churn and improving customer lifetime value.

© 2022 Light Reading, part of Informa Tech, a division of Informa PLC.