The advent of the 5G era makes cloud-based architectures more critical than they were in previous generations, but also makes it far more demanding to implement them. The highly centralized and relatively simple Cloud-RAN designs that were developed for 4G are not adequate for 5G networks, whose bandwidth and latency requirements dictate very distributed topologies. That has led to a disaggregation of the RAN, breaking it into centralized, distributed and radio units, as well as abstracting software from hardware and deploying that software in the cloud.
The emergence of Open RAN potentially enables operators to combine these twin transformations with a shake-up of their supply chain, maximizing the commercial impact of vRAN by opening up a wider base of innovation. On the other hand, the complexities of integrating multivendor macro RANs may make Open RAN mainly suitable for small cell or greenfield environments for now, and it could add to the challenges of macro vRAN, or even delay the process entirely.
All these dilemmas are highlighted in Rethink’s latest RAN report, and in a survey of 78 operators worldwide, which was conducted in June to August 2021 and provides a key source for the analysis. The respondents were asked about their deployment plans, drivers, challenges and spending expectations for vRAN and Open RAN. Together with comprehensive ecosystem interviews, the survey results provide unique inputs to our forecasting model, which has resulted in predictions of vRAN and Open RAN deployment scenarios and spending to the end of 2026. The focus is mainly on 5G, which is driving much of the change, though not entirely so, as many early deployments, especially in enterprise or rural environments, will initially rely on 4G or SingleRAN.
The survey identifies the main stages in a migration from traditional, integrated RAN to the end game of a disaggregated, cloud-native and multivendor RAN. Most early implementations and trials have been run by greenfield operators or technology leaders such as NTT Docomo or AT&T. For the majority of operators, which want to minimize cost and risk, there are many dilemmas inherent in planning a roadmap to full vRAN.
This report is structured around the most important decision points that were identified in the research process, which relate to whether an operator will adopt a step-by-step progression that could last into the beginning of the 6G era, or choose a ‘big bang’ transformation that will involve disruption on multiple fronts, including the network structure, the cloud architecture and the supply chain. The decision points are:
- Should they open up the interfaces in the vRAN, using Open RAN specifications, in phase one or wait for those specs to be more mature first?
- Should they disaggregate and virtualize the network in a single migration? Or will it be less risky to disaggregate first, splitting the baseband from the radio head and then disaggregating it into distributed and centralized units, but without deploying the baseband in the cloud until a second stage?
- Should they introduce vRAN and Open RAN first in smaller networks such as enterprise/private, small cell and rural extension environments, before targeting even bigger transformations in the macro RAN?
- Should they introduce new suppliers and deploy a multivendor RAN, enabled by Open RAN interfaces, from day one, or work with a single vendor first before seeking to shake up the supply chain at a future stage?
There is no simple answer, and operators will take many different approaches, and even adopt multiple approaches and architectures for different scenarios (greenfield sites or localized networks versus the main macro RAN for instance).
It emerged that disaggregation and virtualization, though often assumed to be intertwined, will be implemented separately at first by up to 25% of operators.
Small cells will frequently be the initial focus of a vRAN deployment and this market may develop a distinctive platform and ecosystem because it has a different set of deployers and performance requirements from those of the public macro network. Certainly, the interfaces and architectures that take the lead in small cell and enterprise environments are currently looking distinct.
And while an open supply chain and multivendor macro RAN are aspirations for about three-quarters of operators, at least 40% initially expect to work with just one or two established vendors (even while mandating open interfaces to ensure future flexibility).
As operators make their decisions and plot their roadmaps, the market will see a 60% CAGR in vRAN as a whole, and 124% CAGR in the Open RAN subset of that sector, between 2020 and 2026. In terms of numbers of radio units deployed, the CAGR figures are predicted to be 34% and 184% respectively.