At a live event held by The Mobile Network on 8 February, speakers from Rakuten Mobile, Rethink Research, Dense Air and Picocom addressed Open RAN topics such as:
- Deployments numbers and types to date
- Brownfield progress, or lack of it
- The requirements of deployers of Open RAN networks
- What the ecosystem needs to do next
- How open is Open RAN?
The aim was to give a real world view of how Open RAN is developing, the needs of different types of deployers, and the capabilities of those providing solutions to the market.
The view from Rakuten Mobile and Symphony
The event kicked off with a 1-2-1 interview between TMN editor Keith Dyer and Tareq Amin, CEO of Rakuten Mobile, which built a new network using Open RAN principles, and Rakuten Symphony, the telco platform group formed by Rakuten to commercialise cloud-native and Open RAN solutions.
The interview is cut into chapters, meaning you can move to the chapters you are interested in. The key takeways from Amin’s session were these:
- Rakuten Mobile’s network now has 60,000 macro base stations, 12,000 5G base stations, and a “massive” number of small cells deployed, giving 98% population coverage. All of this is managed by a network team of just 220 network operations personnel, due to the high level of cloud network automation
- Amin is “really, really disappointed” in the lack of brownfield progress to date, but he also thinks we will see a big brownfield deployment this year
- However, he is bullish on the prospects for Open RAN ecosystem, and is adamant that this whole thing it not just about backing one new large player into the market
- Related to that, he sees the need for new hardware solutions, underpinned by new chip technology and architecture, as critical. He referenced the new next gen Intel Xeon-based DU that Rakuten launched late last year
- He also defended the open status both of Rakuten Mobile’s own network and of Symphony’s offering. The external importance of this is that oepnness vs tigher integration is an important factor for telcos currently, as we also heard later on from analyst Caroline Gabriel
- Finally, Amin sees Open RAN within the wider context of the move to a cloud native, automated network, that can genuinely change the way telcos operate
The State of Open RAN: Caroline Gabriel
Click for full size
Rethink Research’s Caroline Gabriel presented “The State of Open RAN: 2023-2027”, putting some numbers on current deployments, and also backing herself to make some predictions.
Gabriel said that there are different motivations driving Open RAN. Politicians see it as a way to further regional self-sufficiency. Operators may not want to deal with dozens of suppliers but they do want optionality of vendors, and swapability. Rural coverage and indoor environments can challenge current TCO models – Open RAN may give a great diversity of price points. Finally, Open RAN may be an inflection point to move to vRAN, which may be easier and less risky with access to an open innovation base.
The upside of these multiple motivations is that there are lots of stakeholders supporting it, but it also risks a fragmented agenda and an unrealistic burden of expectation being placed on it.
Deployments and numbers
Click for full size
When it came to analysing deployments to date, Gabriel highlighted that an at-scale market will rely on adoption in brownfield networks, meaning a longer process of trials and migration. So far, major deployments that have started have had to rely on significant inhouse integration and have often used semi-customized equipment and software.
The biggest technical challenges are to achieve equivalency of performance in a virtualised Open RAN compared to an integrated RAN, in a heavily loaded urban 5G macrosite. That work is ongoing and most operators believe it will be market-viable around 2025-2027.
- The challenges in brownfield networks mean that Open RAN will account for just 3% of macro network radios deployed, and 5% of capex, in 2023.
- The bulk of trials and roll-outs, based on more common platforms, have been in rural areas or enterprise environments
- Small cells, mainly for enterprise networks, dominate the picture and remain the largest category throughout the forecast period to 2027.
- Most standards are not fully specified yet, and many early deployers define Open RAN as supporting Open Fronthaul alone. As the full platform becomes defined, there will be rising adoption of the whole Open RAN set of specs – in 2027, deployments of the whole platform, including RAN Intelligent Controller, will account for more than half the total for the first time
- These trends will lead to the growth rates seen on the left – 92% CAGR from 2022 to 2027 in small cells and 109% in macro.
Challenges to overcome
Click for full size
In Gabriel’s view, Open RAN does face challenges. If it can overcome the bulk of them, then by 2027 Open RAN could reach:
- 26% new macro and micro sites deployed in that year
- 60% new base station capex in that year (inc single-vendor Open RAN)
- Of that, 32% could be multivendor. Whether Open RAN networks will be single-vendor, end-to-end with open interfaces; or best-of-breed, perhaps unified by integrated platforms, is still an open question.
You can download Caroline Gabriel’s slides in full here
PANEL DEBATE
PAUL SENIOR, Founder and Chairman, Dense Air: Senior described how Dense Air is building neutral host, dense small cell-based networks that can support multi-operator services. As it does so, it has specific product requirements to build solutions that are difficult to source from mainstream network equipment providers economically. The truth, then, is that Open RAN innovation gives Dense Air to path to access and build the solutions it needs to deliver its multi-operator, dense networks.
We need to support many hundreds (100s) of MHz bandwidth from a virtualised small cell and Open RAN gives us the opportunity to access and build that capability for ourselves. That product capability sits outside what we can access economically from other vendors
Senior also said that the company had benefitted from involvement in Government-backed pilots and projects – showing how Government backing for Open RAN is helping to stimulate innovation. Through the UK’s Future RAN Competition Dense Air is taking part in the CoMP-O-RAN project to improve the performance of densified 5G NR outdoor small cell networks.
Nor did he agree, in answer to an audience question, that the major network equipment providers could squash Open RAN by dropping their own prices. The truth is, for deployers like Dense Air, the Open RAN ecosystem delivers more innovation in an affordlyable manner than the incumbents have on offer.
DOUG PULLEY, Chief Solutions Architect, Picocom: Pulley said that Open RAN is attracting real-world engagement from manufacturers who are developing solutions to meet a variety of use cases. To demonstrate that he shared some numbers and names:
- Picocom has more than 30 paying customers, and the same number again that we are engaging with in serious discussions
- Publicly there are eight customers who have committed to displaying their Picocom-based RAN products on Picocom’s stand at Mobile World Congress
- These include SOLiD for an O-RU, Accton for an O-RU/O-DU and most recently Antevia Networks, for Enterprise Private Networks
- Geographically, Picocom has customers in Japan, China, Taiwan, Korea, the UK and North America
As we heard from Tareq, Open RAN needs chips, and chip solutions that can support a diverse range of hardware platforms. Picocom supports all splits with its chip and software solutions, and continues to develop its openRAN solution portfolio. And we will be launching our new chip later on this year.
CAROLINE GABRIEL: An audience question asked why brownfield operators are more commonly choosing D-RAN (Distributed-RAN which co-locates the RU and BBU at the site) deployments rather than the C-RAN model that greenfield operators have deployed.
Gabriel said greenfields have often gone for full cloud-RAN as they do not have to worry about legacy so can be quite ‘purist’ in their architecture choices, whereas established operators instinctively see D-RAN as easier to integrate/migrate. They are also concerned about performance trade-offs if the vBBU is too distant from cell sites, given the current state of acceleration technology.
AUDIENCE SURVEY AND POLLS:
We asked our audience three questions. We wanted to know what the general sense about Open RAN is now. This, of couse, from an audience who are engaged or interested in some way in Open RAN.
1. First off, we wanted to know where our audience thinks things are happening at the moment. We asked all registrants where they think Open RAN activity is happening right now.
- 37% reckoned Open RAN is mostly small live pilots
- 23% said most activity is in field trials
- 23% said it was happening at scale in live production networks
2. Next, during the live event, we asked the audience. “Where do you think the majority of Open RAN deployments have been based so far?”
The top answer here was Enterprise and Private networks, with 39%. Rural macro was a close second (31%), followed by dense urban (17%) and urban macro (13%)
3. Finally, near the close of the event, we asked the audience when they think major commercial developments will take off. An overwhelming 69% said they think deployments will take off between 2024-2026. Only 26% thought it would be later than that. 4% were optimists, opting for this year. And nobody at all chose our final option of “Never”!
Conclusion
So that was The Truth About Open RAN, as it looked to the speakers and attendees at the event.
If you’d like to find out more about how Picocom is enabling the ecosystem to move forward with Open RAN solutions, and the partners it is engaging with, then please do head over to its site to find out more.
Finally, Picocom and The Mobile Network will be bringing you a second web event, following Mobile World Congress, explaining the ecosystem momentum that is developing in Open RAN, plus the role of Picocom and its partners in that ecosystem. Not one to miss!
