The journey to O-RAN
Rome wasn’t built in a day, as the saying goes. Nor was it the work of one person. Like any major project, it took time and the bringing together of many skills. The same is true of O-RANs, which are fundamentally different from any previous mobile generation.
We will start by looking at the evolution to date and the journey ahead of us, then discuss what we need to do to get there.
Telco networks have evolved from switches and boxes to being increasingly software defined (see Figure 1 below). But they now face a dauntingly steep climb as they migrate from virtualized RANs (vRANs) to O-RANs.
The journey started with disaggregating radio software and hardware. This allowed network functions to be ‘virtualized’ – ie controlled by software – usually run in the cloud – which in turn meant networks could use a wider variety of hardware, so long as it could communicate with the software. This brought network flexibility, better performance, coverage and quality of service.
However, despite the flexibility, networks still tended to be built upon proprietary interfaces from a single vendor, creating ‘vendor lock in’ for hardware and software, and a lack of competition.
Open RAN solves this, allowing operators to combine hardware and software from multiple vendors, choosing the best performing or best value solutions for their specific network needs. The O-RAN Alliance (a worldwide community of mobile operators, vendors, and research & academic institutions) has set definitions for open interfaces, making interoperable cloud-native RAN a reality.
But that is not the end of the story. Like all cloud-native workloads, networks require massive intelligent automation at scale to realize their full potential, whilst also minimizing TCO (Total Cost of Ownership). This step up requires a collaborative effort by vendors, operators, and systems integrators, because no single company has all the necessary technologies and expertise to do it alone.
So, what will it take to bring about O-RANs successfully? As an industry, we must work together to address six major focus areas in an integrated way:
A cloud-centric architecture for baseband software will enable the true benefits of a virtualized RAN in a cloud environment.
Cloud-native edge compute - edge compute is a type of network architecture that offers services at the edge of the network, increasing service quality and reducing latency. This cloud-native, real-time platform with carrier-grade resilience will provide reliable services for hosting virtualized or containerized RANs.
Disaggregated O-RAN and cloud RAN - RAN software operating separately - in the cloud - from the specialized hardware it used to run on, and instead running a variety of heterogeneous, interoperable hardware in an ‘open’ environment. Disaggregation will split the radio units from the baseband software. The software can be hosted on the network edge. Large scale automation, leveraging carrier-grade DevOps, will allow operators to manage radio networks at a reduced TCO.
A data-driven autonomous network - autonomy allows networks to respond to demands dynamically without human input - replacing manual tasks with intelligent machine learning that can learn to reconfigure the network better than a person can. This supports lower TCO, whilst allowing networks to be more agile and adaptable. Data collected from the open interfaces of the O-RAN/CloudRAN architecture can now be leveraged to develop machine learning models that will power autonomous RANs.
A sustainable cloud-native network - a system for operating the entire network and its services and functions in the cloud.
These focuses require broad industry collaboration to create the common standards, architecture and models that will be critical for test and automation.
The main challenge to delivering all these different areas of focus is interoperability – ie the successful integration of many multi-vendor components.
In a traditional closed RAN, one vendor is responsible for all of the hardware, which it designs to work together according to its own clear standards. But, what has been described as "subtle ambiguity’ in certain O-RAN standards can lead to variations in implementation between vendors that can cause compatibility problems. In the early stages at least, it is likely that the usual tasks; service orchestration, network management, and maintaining an inventory of spares, will be more challenging with multiple vendors.
Currently, there are three major approaches that will help us succeed with O-RAN.
An open approach to testing that spans all network domains. This approach would be automated and API-based, with standardized test benches and performance benchmarks. Collaboration initiatives, like the i14y Lab, could also provide test tools such as robust 3GPP- and O-RAN-compliant simulators that can emulate Layer 1, user equipment or base stations.
Standardized automation platforms for network operations. These would automate both the network and the alerts that it generates. Besides providing the network operations center (NOC) staff with greater visibility, automation also would maximize their productivity and make the network more predictable, which aligns with one of O-RAN’s major goals: a lower TCO than traditional networks can achieve. For example, automation can enable the use of digital twins for multiple RANs, where the NOC can access several configurations in real time, provide a simulated environment before deploying new configurations, reproduce faults and correlate events. Automation platforms require enormous amounts of data, so they can be trained to handle a wide variety of real-world scenarios. Industry collaboration labs (like i14y) can play an important role, by providing that data.
An open orchestration architecture. Driven by open APIs, this architecture would include common data platforms and models. The collected data also would be open, so developers can build use cases around it.
In 2022, the US National Telecommunications and Information Administration (NTIA) launched a 2-year ‘5G Challenge’ program to foster a vibrant 5G O-RAN vendor community, hosted by the US Institute for Telecommunication Sciences (ITS).
In its first year, NTIA and ITS required the contestants to successfully integrate hardware and/or software solutions for one or more 5G network subsystems: Central Unit (CU), Distributed Unit (DU) and Radio Unit (RU). The Capgemini team competed in the Central Unit category, winning all three challenge stages.
In the 2023 5G Challenge, NTIA selected contestants with high-performing 5G subsystems that showcased multi-vendor interoperability across RUs and combined CUs and DUs (CU+DU). CableLabs hosted the challenge and provided two separate 5G test and emulation systems. The NTIA awarded Capgemini first place for Multi-Vendor E2E Integration. NTIA specifically applauded Capgemini for achieving a 100% pass rate on all feature and performance tests. At the closing ceremony, Capgemini was presented with two prizes: Multi-Vendor E2E Integration and Wrap-around testing for the Open RAN CU and DU.
Figure 1: telco network evolution
Figure 2: a summary of OpenRAN's challenges
