What are technical and operational challenges in the midband spectrum for companies operating networks in the United States?
In every spectrum band, there are some sort of challenges that wireless systems have to deal with: Limited channel bandwidth, passive intermodulation, propagation or in-building penetration issues, to name a few. The midband is no exception, and it has some unique characteristics and usage rules that will shape how and where it is deployed over time. Let’s break down some of them.
The amount of bandwidth is a boon – but some strings are attached in using it. Nick Karter, director of product strategy for Analog Devices, points out that the midband provides the opportunity for large swaths of contiguous spectrum and that requirements are being discussed to provide channels that are 200 to 400 megahertz — the first time this is available outside of mmWave frequencies. That presents new challenges for filters and other RF components and requirements that have to be met by both mobile devices and base stations. Rex Chen, director for 5G business development at test equipment company LitePoint, says that his company is seeing instances where chipsets can support, say, 200 megahertz of bandwidth – but in actual deployments, carriers (used to five, 10 or 20-megahertz channels in previous Gs) aren’t using that entire capability because they’re still testing to make sure it works for across-the-board deployment.
Another thing that is changing in 5G, according to Chen, is the way that data is being consumed: The typical emphasis on the downlink is evolving. The increase in remote work, video conferencing and so on, means that uplink capabilities are becoming more important. This in turn, he says, is leading carriers and chipset makers to re-think how they mix and match spectrum resources: “One of the ideas is that, now that I have more uplink requirements, if my low-band is very congested, maybe I have the downlink on the midband and the uplink is on the low-band, or vice versa. We’ve also seen from the chipset side, as they talk about carrier aggregation, besides contiguous and non-contiguous bands, they’ve also proposed that maybe we can carrier-aggregate millimeter-wave and sub-6. Or again, mix and match.” While LitePoint hasn’t noted any particular behavioral quirks to the midband itself, he says, there is inherent variety in the use cases for 5G that may also end up making deployment challenging: Do you choose to build for coverage (important for smartphones and mobility) or high-capacity data (necessary for FWA)? “We know that as a whole, 5G isn’t just addressing the smartphone market,” Chen says. “Smartphones are definitely an important market, but we’re seeing fixed wireless access [and] IoT modules in this space. With the different devices that are coming out, the requirements are also quite different: Some, it’s coverage; some, it’s, you need a lot of data. … I think the variety of these device requirements has made the carriers understand that we need this midband to be part of our 5G strategy moving forward.”
Variation of power levels in the band. While the C-Band and 3.45 GHz spectrum rules allow high-power operations – which is the preference of mobile network operators – the 150 megahertz of shared CBRS spectrum has vastly different, and lower, power limitations. That doesn’t mean it won’t be used or can’t be aggregated, but it does mean that CBRS has limitations in a macro-cell context.
This spectrum sits squarely in the middle of the 3-4 GHz range and has been designated for full-power 5G transmission in many countries; the tiered incumbent-sharing system, which the U.S. pioneered, hasn’t really caught on broadly around the globe. However, the CBRS PALs auction was arguably successful simply from the number and diversity of bidders that it brought in, with more than 228 winners. CBRS is serving an important role as the U.S.’ play to provide non-carrier businesses with the opportunity to use desirable spectrum for private networks. Spectrum regulators in other countries, such as Germany and France, have instead reserved small portions of their midband spectrum allocations to be licensed to enterprises.
However, there are some practical concerns about the significant differences in power levels across the band. Some are worried that once the 3.45-3.55 GHz band can be used, the higher power levels allowed in that band will cause interference problems for the lower channels of the CBRS band.
Dish Network (which bought the most CBRS licenses) argued in a March 2021 filing to the FCC that the power level imbalance across the band “[leaves]a relatively encumbered CBRS band sandwiched between the 3.45 GHz Band and the 3.7 GHz Band, both of which … have services rules optimized for large-scale, wide-channel 5G service offerings. This is akin to connecting two cities with a new 8-lane high-speed roadway, but constructing a stretch of single lane road in the middle. The current U.S. approach undermines the overall usefulness of the entire 3 GHz band, and places the United States at odds with our major global competitors, especially China.” In particular, Dish fretted that CBRS’ far lower-power operations would encounter “blocking interference” from 3.45 GHz (both of them are TDD bands) and asked the FCC – as have a number of other companies – to create new, higher power limit categories for CBRS. While “careful network planning and coordination among spectrum users” may limit the harms to CBRS, the most efficient and logical solution is to increase the ceiling for CBRS power limits to place it in parity with its neighbor,” the company said. Dish isn’t the first and probably won’t be the last to ask the FCC to consider higher CBRS power limits, but thus far, the agency has not indicated that it plans to change the rules.
Cost of deployment. Midband spectrum costs alone are stacking up: $4.58 billion raised in the CBRS PALs auction; a whopping $81.2 billion in the C-Band auction and a reserve price of about $14.8 billion for the 3.45 GHz auction. Those are substantial costs for operators, and on top of them come the costs of actual build-out. Luke Getto, director of product management at passive and power components company VoltServer, says that midband deployments are likely to be less expensive than mmWave for multiple reasons: The components are less expensive, operators don’t have to deploy as many sites because midband propagation is better and also, that vendors have been more forward-looking in their support roadmaps as carriers have made it clear that want to see plug-and-play readiness for things like CBRS. He expects C-Band to be treated similarly, and for it to be deployed on existing infrastructure first, in order to reduce the upfront deployment costs. While 5G in mmWave required a complete overlay, he said that CBRS is being deployed in such a way that the radios get put up, plugged in and then function as best they can in the existing system. Carriers, he went on, “only have so much money. They’ve got to look for economical ways of deploying it,” he continued. Midband coverage will probably be good enough that they can, much like in LTE, put up radios on existing infrastructure first and then decide later if they need to densify with additional sites. “They’re going to get both the coverage and capacity they need, but they’re not going to look to overlay an entire new radio architecture, in my opinion,” Getto said.
Coexisting with incumbents. While this aspect of network planning is most obvious in the CBRS band, with its three-tiered sharing levels and coastal sensor network that picks up use by itinerant naval radar systems for which other users must vacate the band, it’s not absent from either C-Band or the upcoming 3.45 GHz band. Satellite operators are busy repacking their operations to move into the top 200 megahertz of the band (4.0-4.2 GHz), protected by a 20-megahertz guard band, which require modification of existing equipment including filtering and re-orientation to mitigate interference. But C-Band licensees also have to do their own due diligence to check for registered Earth stations (there are about 16,000 of them across the U.S.) and make sure that their terrestrial network designs don’t generate out-of-band emissions that will interfere with those stations’ ability to operate. In the 3.45 GHz band, there are nearly three dozen areas around the country with federal incumbent systems where network operators will have to coordinate their use of the spectrum and may be restricted or at least, not entitled to interference protection from those systems. Those areas, according to NTIA and the FCC, are “limited in size and scope and include military training facilities, test sites, Navy home ports, and shipyards.” The Department of Defense has been using the 3.45 GHz band for high- and low-powered radar systems, including fixed, mobile, shipborne and airborne systems, as well as testing and training related to those systems. The places with federal incumbent systems have been dubbed Cooperative Planning Areas (CPAs, of which there are 33) or Periodic Use Area (PUAs; 23 of them overlap with CPAs), where coordination with federal incumbents will be required. In all but two of them, winners of 3.45 GHz licenses have to coordinate operations across all 100 megahertz of the band. NTIA has published federal incumbents’ transition plans and a workbook to guide coordination within the band.
One thing that does appear to be clear: The table stakes spectrum for successfully providing 5G services involve a far higher amount of spectrum and at least some of it needs to be midband. As AT&T put it in its FCC filing: “Ten years ago, providers such as MetroPCS and Leap could provide competitive, high-performing mainstream wireless services with 20 megahertz of paired spectrum in just a few markets. Today … providers will need 80-100 megahertz of contiguous mid-band spectrum to provide the full benefits of 5G to their customers. And demand for more such spectrum will only escalate as the new 5G ecosystem generates more bandwidth-intensive applications.”
Looking for more insights on the midband spectrum landscape? Check out the RCR Wireless News webinar featuring Aurora Insight, Analog Devices and LitePoint.