No, millimeter-wave 5G is not ‘dead’
When the standards defining 5G were first released, one of the most remarkable facts was that it would rely on millimeter-wave frequencies, a region of the spectrum where no wireless carrier, or for that matter few others, have ventured. It envisioned the use of frequencies at least an order of magnitude higher in the electromagnetic spectrum than those currently used for 4G, potentially as high as 100 GHz. Once the media and other interested individuals had digested this fact, it became obvious that this would be an extraordinarily difficult task.
In fact, recent articles claim that “millimeter wave is dead” for 5G, primarily because the short range achievable at these frequencies will require immense numbers of small base stations, resulting in a potential cost to wireless carriers of $65 billion, by some estimates.
So, are the naysayers right, or will millimeter-wave frequencies be the salvation of 5G? This article attempts to answer that question, but first it’s necessary to understand why this has become such a contentious issue.
No vacancies
Since the earliest days of radio, the federal government has allocated frequencies for services such as AM and FM radio, television broadcast, public safety and many others based on the technology available at the time. At first, operation was only technologically possible at lower frequencies, so that’s where the allocations began. As technologies advanced, it became possible to operate at higher frequencies, so the FCC began allocating services there, placing them based on the coverage they required. The result of this is a spectrum chart so densely packed it’s impossible to read without a magnifying glass.
For wireless carriers, the millimeter-wave region is the only part of the spectrum with enough available bandwidth to accommodate 5G's massive data.
When cellular technology emerged, the most viable frequencies from a propagation perspective were those beginning at about 600 MHz and later extending to about 3 GHz, where carriers operate today. However, beginning with 4G and the emergence of data services that allowed the use of streaming video and thus required more bandwidth, the wireless industry faced a problem: There was very little available low-frequency spectrum left to auction off to the carriers, and some carriers had “acquired” more low-frequency spectrum than others. For example, T-Mobile acquired lots of the spectrum while AT&T and Verizon had not, instead relying on their stock of millimeter-wave allocations. Since then, the FCC has been feverishly finding ways to free up more spectrum in this region as well as a “refarming” existing services to other frequencies, among other approaches.
Now, 5G has exacerbated this problem because while the latest iterations of 4G enable data rates of about 70 Mb/s or slightly higher, 5G promises speeds that exceed 1 Gb/s that will require much more bandwidth, a commodity that is simply unavailable at lower frequencies. The use of very high frequencies is designed to solve this dilemma initially using the immense bandwidth available at 24 GHz, 28 GHz, and even higher in the future as very few services operate there. The drawback is that because propagation at these wavelengths is limited to several hundred feet and signals can be attenuated by almost anything, advanced technologies will be required to make use of them, some of which weren’t even fully developed when the 5G standards were released. These factors and others provide a rich source of fodder for those who believe that it will be economically disastrous for wireless carriers to deploy enough millimeter-wave infrastructure. Thus, they believe it is not viable.
However, as with so many technologies before, those required for millimeter-wave operation are already being developed and have demonstrated that it is indeed possible to provide this amount of coverage, albeit at a cost. What’s really required is a new network design in which “smart” repeaters, advanced software and algorithms, digital beamforming, and other techniques allow millimeter-wave signals to permeate RF-restrictive environments without dramatically increasing the number of small-cell base stations.
That is, instead of the Johnny Appleseed approach of depositing small cells everywhere, “smart” networked repeaters with greater coverage will be used, providing coverage virtually anywhere indoors or outdoors and potentially reduce capital expenditures and total cost of ownership by something like 50%. They achieve this because while 4G and previous cellular generations have expanded capacity and coverage primarily by adding more base stations, the repeater-based approach uses fewer small cells more effectively, delivering signals even to places where RF energy is restricted.
These repeaters are small enough to be placed almost anywhere, from light poles to ceilings in office buildings and throughout stadiums and other large venues. Not surprisingly, this will require a collection of technologies including bespoke chipsets, mesh networking, massive MIMO antennas with hundreds of elements, and algorithms that can react instantly to changes in traffic patterns and varying propagation conditions.
So, even at this relatively early stage of millimeter-wave technology deployment, there is already one solution on the table for bringing operation at these frequencies to fruition. In the future, it’s very likely that other approaches will be introduced that either use this model or something similar to make millimeter-wave 5G operation feasible while also producing the amount of coverage required and reducing cost to tolerable levels.
That said, there’s a more basic reason why millimeter-wave technology will continue to play a major role in 5G: There is simply no other way to achieve what the fifth generation of what was formally called cellular is intended to deliver. Even if spectrum were available at lower frequencies, the bandwidth required for gigabit-per-second data rates would consume it instantly. What’s more, the antenna arrays that can be constructed at these frequencies are relatively small and can deliver high amounts of gain over very narrow beamwidths by combining the antenna elements in phase to direct signals to very specific areas on demand.
It’s also important to keep in mind that the development of receiver, amplifier, antenna and other technologies at a pace consistent with what is required to deliver a 5G performance is very new. There are already hundreds of research papers proposing advanced techniques. Many others will follow.
So, regardless of the challenges faced by operating at these wavelengths, which are indeed substantial, the industry is far from conceding defeat and not just because they’ve already expended an enormous amount of money. It’s because 5G must succeed, and millimeter-wave frequencies provide the only realistic solution.
