The wireless industry has been talking about CBRS, LTE, and 5G for over the last couple of years. To better understand how the technology and have market has developed over that time, DoubleRadius decided to host a panel of leading solution providers to discuss the trends they see today. This article provides the informative answers to the key questions we posed during that discussion.
Nearly one year into full CBRS operation, are we seeing any change in deployments?
Before part 96 rules it was all part 90, so it's mostly fixed wireless in the U.S market in the CBRS band. A lot of WISPs deployed it, some utilities. It was mostly fixed wireless private networks that were being used with this spectrum.
Since the launch of CBRS officially last year, we’ve definitely seen an uptick in a lot of new use cases which we're very excited about. In the public space, the carriers have been actively deploying mostly in urban spaces to add more bandwidth. Private networks are longer just for fixed wireless. We're definitely seeing a lot of mobility-type projects, warehousing being one of the biggest. Situations where mobility is very important, along with security. Those are the key ingredients for LTE within CBRS.
The spectrum is now available to everyone with CBRS. Whereas a license was previously required, there's a tremendous number of use cases popping up in IoT and mobility. A lot of companies are building infrastructure and using CBRS to do neutral-host type use cases, where they'll work with the carriers to offer some kind of roaming agreements.
It's no longer exclusively fixed wireless, although fixed usage has grown, especially in the education space. The COVID situation made it obvious that not everyone has access to broadband. Even if it’s available, it doesn't mean that all students have internet access. So school districts have been building their own networks, and they need a kind of nomadic network to support Chromebooks and other WiFi devices. This allows students to remain connected both on and off campus, as networks are being extended to low income areas.
Educational adoption of LTE has also been regional. California, for example, is a much more progressive with their funding for schools building networks. Some schools had already deployed in the Education Broadband Service (EBS) 2.5 GHz band in certain regions. When CBRS launched soon after the pandemic hit, schools that had been thinking about building LTE networks but weren’t able to previously, now had the access to spectrum. So we weren't surprised by schools building networks, but the CBRS band accelerated this beyond what we expected.
There number of CBSDs in the market is now nearly 170K. It has a very healthy ecosystem after one and a half years, and we expect much more growth in the near future. In addition to all the current use cases, we’re moving on to new use cases that will be enabled by 5G.
Have fixed wireless operators been hesitant to get onto the SAS?
Of course there is hesitancy with a new system especially for the Part 90 operators. But as soon as they get familiar with and understand its simplicity, that hesitancy goes away. CBRS is capable of providing a significant amount of spectrum, especially in rural areas for fixed wireless.
We’ve seen that a significant number of PAL bidders have been fixed operators, and this shows that they’re gradually having trust in part 96. They're not going to lose spectrum. Not only can they continue operating in the upper 50 MHz, but it also opens the door for the lower 100 MHz. So hesitancy has been there, but as soon as operators get educated this is not going to be a major concern.
What kind of activity have you seen with Utilities holding PAL licences?
What we've seen has been really interesting. We really didn't anticipate the amount of utilities and oil companies getting involved with the CBRS spectrum. That really opened
up some additional opportunities for us as OEMs to go after a new segment.
Utilities are very slow to change a lot of their network architectures. However, the unfortunate security issues with network hacking, plus the congestion in free spectrum, has created a lower point of entry for them. They can create much more secure wireless access LTE networks.
The amount of spectrum needed for a sensory network is only 5-10 MHz, but some utilities are also offering broad service to customers. PAL licenses can come into play for utilities securing their sensory networks, and then they can layer in their non-PAL licenses for capacity to also serve broadband to their customers. (For an example of a dual-band antenna supporting both the Anterix 900 MHz and 3.5 GHz, see our Utility Broadband blog.) Today, several other bands can be combined with CBRS as carrier aggregation, using LTE, 4G or 5G.
To echo what was previously mentioned, we’ve never seen so many school districts interested in CBRS. Municipalities as well. Some are covering their whole city with a sensory network, and providing access for their employees. WISPs can also sublet PAL licensees from larger carriers, in order to serve rural communities more quickly. Otherwise, these carriers might not get to those remote areas for years. Generally speaking, FCC rules promote using the license or letting it be used.
The sub-leasing process currently takes a few weeks using form 608, but the FCC will soon be allowing SAS providers “light-touch” management. This will allow a SAS to process a sub-lease transaction and make it operational within minutes! This should be coming within the next one to two months.
One year later, has the FCC accomplished their goal of opening up CBRS to a larger market successfully?
I believe they have. We see, from an OEM perspective, people using antenna products they wouldn’t traditionally use. A lot of that is attributed to the amount of lobbying with the FCC. Federated Wireless was a huge spoon feeding them. Their CEO, previously from Sprint, knew how to help the ecosystem work well. Google and everybody else behind this effort was also trying to make it as flexible as possible. The agency was willing to listen and figure out how to use that spectrum that was really of no value outside of the Navy using it.
I think the way the SAS platform was designed and developed gives a lot of really good flexibility in how to manage interference. It really has opened it up, to the point where some companies are wanting to deploy thousands of nodes.
There's been a lot of improvements right from day one right to where we're at today, on the SAS and vendor side, when it comes to the features and software. We’re also trying to make it more seamless as far as getting CBSDs onboarded. The only areas that we see some challenges today are within the Dynamic Protection Areas (DPA’s), but it's not like before. Most SAS providers now have same-day grants, so you can usually get your devices up and running.
There's still the risk for DPA activity suspensions. We work with those DPA regions, and they always have concern for how they can provide “five nines” reliability. On our end we're trying to add multiple, redundant grants to make it more robust. But from the launch of CBRS to now, there's definitely been a lot of improvements and stability.
If you're outside the DPAs, it's in a sense a kind of unlicensed because the operator still has flexibility on what channels they can pick and it's based on the spectrum and query message. Within our system we allow the operators to set their preferred channels and so they can rank them and then based on the responses from the SAS. Our software will pick the desired channels so they still have the RF channel planning capability.
There are a couple of coastal areas in which there is a lot of radar activity (ie Norfolk, San Diego), but outside of those we haven’t seen significant impact. Within the affected DPA’s, the upper 50 MHz remains available. Only the lower 100 MHz would be affected, but when a radar is activated, only one to two channels max are taken. So there is availability.
We are actively working with the Dept. of Defence, the Navy and the FCC to make spectrum allocation more efficient. From their incumbent perspective, we haven't heard any significant concern, or any interference objections or complaints. We communicate regularly with them, so we can study and identify the areas that we can make CBRS more efficient. We can provide the available spectrum through GAA and PAL coexistence, and the user can pick within that range. So I really don't see a major interruption or hiccup in the operation of the fixed operators.
The question will be when General Authorized Access (GAA) coexistence kicks in, how much of that goes away. But for the most part, if you're outside the DPA’s, there aren’t many issues or concerns right now for most operators.
How do you plan to support 5G from a product perspective over the next couple of years?
We're seeing the technology of the antennas evolving to meet the requirements for 5G. When you're sub 6 GHz and going to 5G, you can typically support it on your traditional telecom bands, low and mid. Now they're integrating into the CBRS band the 3.5 GHz, and they have set aside their spectrum just for this in urban markets.
The antennas can be beamforming, 8T8R, going up to 32 or 64. The amount of complexity in the tenant is increasing, and of course the size because we're in the sub 6 GHz band. Once you're increasing the amount of beam forming elements in the antenna it is going from a four port to an eight port, almost double in width unless you're sacrificing something.
So really the technology has to keep up with 5G. It has been, but in telecom and fixed wireless we’re still a number of years out before that's going to be adopted. Not everyone's going to need the ultra low latency and real time control benefits over LTE for their use cases right now. The bandwidths are just outstanding, but LTE seems to be sufficing right now. I think it will be a few years before we see 5G adoption move from urban areas, when we start seeing the use cases increase in industrial and suburban areas which are going to need these applications and features of 5G.
It will be a natural transition from LTE to a 5G network. You can build on that to do a non-standalone 5G system. That's just the natural progression. That’s what the carriers did and it’s what we're doing. So following on the LTE path is great because you're looking towards the future, in case you are going to make that switch over to 5G. You're on that right spectrum at the 3.5 GHz. The way it's regulated right now means you have some security on using that band.
LTE still has a long life ahead. Right now the ecosystem's very much growing. You can see the number of UE supported devices almost weekly increasing, so we definitely see that lasting many years. We’re seeing 5G rollouts in urban areas, and a lot of macro deployments or millimeter wave small cells. But the chipsets and the devices to support 5G are still in the early stages. So 5G in my mind is not commercially available to the wide scale audience yet, and there’s not a lot of great devices to even support it yet. We're expecting next year for more 5G products to start rolling out big time. There's nothing preventing 5G today within CBRS, so it's really just the product landscape
There are solutions out there that you could deploy. If it's a RU/BBU split architecture, generally those systems will allow for 5G by just swapping out the baseband. So there's usually an upgrade path there. And we have a product line that we're launching to support that here pretty soon. However, a lot of the small cells with baked in SoC’s don’t have much of an upgrade path because the file layer is limited to 4G. So some of the devices won't have a software upgrade. Another thing to think about is the core. The 4G core isn't going to work unless you're doing a non-standalone setup. Most of the private deployments we're seeing are doing standalone, so it's a separate 5G core as well.
If you’re looking at a raw data throughput comparison between 5G and LTE, your downlink isn't much better. It's pretty much the same bits per hertz. The uplink is what we're excited about, because the uplink actually doubles. In LTE you're generally only doing a SISO, one data stream on the up, which is generally the first bottleneck when people are doing fixed wireless. 5G will at least have two data streams up, so that would definitely benefit . But as of today, there are many options unless you're going to buy very expensive macro equipment. LTE is now in the plateau of productivity stage, so WISPs that are really looking for an affordable solution are going to realize much better cost structures with LTE.
Two weeks ago Apple™ released their new iPhone 13 which supports CBRS as 5G (n48). This means the chipset is available, and I'm hoping that that will motivate the market to consider using the CBRS band. There are many ways to combine CBRS as either 5G or LTE with other bands. Dynamic Spectrum Sharing (DSS), which flexibly changes the CBRS span between 4G and 5G depending on the number of users in the market, is also available.
The Federated Wireless SAS is capable today to register and grant 5G CBRS. However, we need to do some work to enhance the solution and use all the capabilities that 5G can provide. For example, ensuring that the channel location is more adaptable to 5G. We’re optimizing it, doing some work related to dynamic beamforming that is not yet developed. So we have some work to do for the enhancement, but the basic 5G operation as far as SAS is concerned is there today.
Need more info on CBRS, LTE & 5G?
Learn more about CBRS, LTE & 5G solutions from the companies represented on our panel: Alpha Wireless, Baicells Technologies, and Federated Wireless. And to hear the final thoughts from our panelists on the market, and where we’ll be over the next few years, be sure to watch the recorded discussion by clicking the button below.
