Two events not entirely unrelated dominated midband spectrum news in the US last week with implications for the rest of the 5G world. One was the widely publicized conclusion of the C-band auction of spectrum in the 3.7-3.98 GHz range, being vacated by satellite operators and their space-to-ground operations. The second event, of less immediate note, was the decision by the CBRS Alliance to rebrand as the OnGo Alliance, as part of a drive to extend its role in managing shared spectrum initiatives beyond American shores to the rest of the world.
The US-specific CBRS band, which has been the Alliance’s focus so far, covers 3.55-3.7 GHz, referred to as the 3.5 GHz band, and so is adjacent to the C-band. This adjacency gives rise to interference issues primarily, which the OnGo Alliance as it now is has expressed concern over and produced recommendations for mitigation. Both bands will also be used for 5G services and are sought after because they lie in the sweet spectrum spot combining acceptable range with considerable capacity and performance. But otherwise there are clear distinctions, with the CBRS band being shared spectrum, while the C-band is subject to the same rules as other auctioned public spectrum.
The commercial history of the CBRS band dates back just over a decade to 2010, when the US National Telecommunications and Information Administration (NTIA) identified the possibility of creating a publicly available transmission zone in the 3.5 GHz band for shared federal and non-federal use. This was dubbed the Citizens Broadband Radio Service or CBRS (rather confusingly since commercial truckers called their 28 MHz service citizens band radio).
After various workshops and deliberations, the FCC finally – in January 2020 – authorized full use of the CBRS band for commercial wireless services without the previous restrictions to prevent interference with military use of the spectrum. This cleared the way for mobile operators to deploy services without having to acquire spectrum licenses, unlike the situation for the adjacent C-band spectrum.
The CBRS band is now being governed through three tiers, one applying to the top 50 MHz between 3.65 GHz and 3.7 GHz, the second 100 MHz between 3.55 GHz to 3.65 GHz, and the third spanning the whole 150 MHz band up to 3.7 GHz. Incumbent users, including US Navy radar operators, are in the first tier with protection from interference from the other two tiers. Secondly, Priority access operates in the bottom 100 MHz and is governed by competitive licenses in 10 MHz channel chunks within this assigned through the FCC’s auction, which concluded in August 2020, raising $4.58bn.
Finally there is the third lower tier called General Authorized Access tier, a kind of hybrid defined as license-by-rule, to allow open and flexible access to the band for as many users as possible. Such users are permitted access to any portion of the whole 3.55 GHz to 3.7 GHz band that has not assigned to one of the two higher tiers. This is like WiFi in the sense that equipment can be deployed by individual building owners, or end users occupying what would typically be a multi-tenanted property, but they would have to pay a fee for the band they were accessing, administered via a server.
Experience grappling with this rather convoluted arrangement has led the OnGo Alliance to believe it is well placed to help marshal apportioning of shared spectrum in similar bands in the rest of the world. Indeed, OnGo issued this statement: “The Alliance has developed knowhow on building an ecosystem, identifying and resolving policy and technical issues, stimulating wireless markets, and building industry consensus.”
The name has been taken from the OnGo-branded LTE and 5G offerings originally developed for CBRS spectrum in the USA. The Alliance has promised to draw on the experiences of its 200-some members in the USA, many of which are ready to drive private LTE and 5G in shared spectrum in other markets. It also said its technical work so far will help form the basis for a new set of specifications to encompass other bands, and to enable a consistent coexistence framework across networks in a range of frequency bands.
Meanwhile, the imminent arrival of services in the C-band, now that the auction is complete, has also been occupying the OnGo Alliance, with concerns over mutual interference. The Alliance published a briefing document on this for its members, or indeed anybody interested, in October 2020, highlighting the general risk posed in both directions when either a base station or mobile device on the network attempts to access a signal in the presence of a much stronger interfering signal on a nearby frequency.
The result can be reduction in data throughput from 100Mbps to say 95Mbps in a minor case, to 50Mbps or even as far down as 10Mbps in progressively more severe instances, or full interruption to service. The document pointed out that use of filters to mitigate interference would be of limited value because they cannot suddenly cut in right on the 3.7 GHz boundary between the two bands and require around 20 MHz separation, which would mean a significant part of the band would still be at risk of disruption.
Fortunately, mitigating factors will come into play as deployments unfold, the Alliance predicts. For example, many of the worst-case estimates of impact to CBRS assume that C-band base stations will run at maximum allowed power, that is hundreds of kilowatts. In practice, most macro cellular systems do not operate at anywhere near maximum power.
Furthermore, CBRS and C-band systems will typically have their antennas tilting downwards and certainly not pointing directly at one another, which will again reduce interference way below worst-case value. This mitigating factor will be amplified as 5G rolls out with dynamic beamforming antennas. The beams will typically not dwell on an interferer in receive mode, as the Alliance puts it, and devices will not constantly be in the transmitted beam.
Another factor is that propagation loss resulting from clutter in the signal path, such as trees and buildings, which is a problem in one sense, will reduce impact of interfering signals.
So while there will be limits on what operators can do individually to mitigate interference, in practice it will become less of a problem as deployments roll out and experience with mitigation accumulates.
On the C-band side, operators are licking their financial wounds after the auction, which ended raising double some analyst expectations at $80.92bn. Their ability to exploit this newly acquired prized spectrum will depend on them being to absorb these costs and invest sufficiently and quickly enough in subsequent deployment.
To this end, we have already seen reports from Bloomberg that T-Mobile USA and AT&T have borrowed heavily to fund spectrum they have just acquired – $3bn for the former and $14bn for the latter. This indicates just how critical this spectrum band is viewed for gaining or retaining substantial shares of the 5G market.
At this stage, the bid allocations have not been revealed, but it is clear that AT&T has spent heavily, even though Verizon was expected to have spent the most of the large US carriers. Comcast, Charter, Cox Communications, US Cellular and Dish Network are also assumed to have been significant bidders.
There were 5,684 licenses available for those frequencies vacated by satellite operators now migrating into the adjacent 4.0-4.2 GHz range, with a 20 MHz guard band between the two sets of users. These satellite operators will receive incentive payments from the FCC that could almost reach $10bn, reducing the deferral take somewhat.
Next comes the assignment phase when the winners’ names should be revealed and able to bid for frequency–specific licences.