Rome wasn’t built in a day. The same can be said of cloud-native, AI-optimized 5G open radio access networks (O-RANs), which are fundamentally different from any previous mobile generation. They are not just evolutionary, but a step change in how radio networks are designed and built. More than a platform for connectivity, 5G O-RAN is a platform for innovation.
Traditionally, the RAN market has been dominated by a few organizations with only their own interests at heart, never designing their architecture to be interoperable. This has forced network operators into vendor lock-in situations that hamper innovation, but with O-RAN that’s no longer the case. How quickly the industry is able to come together and take full advantage of this, however, is another matter entirely.
This industry-wide collaboration is being actively facilitated and encouraged by the O-RAN Alliance, a worldwide community of mobile network operators, vendors and research institutions aiming to reshape the RAN industry into a more intelligent, open, virtualized and fully interoperable ecosystem.
The importance of industry collaboration
Like all cloud workloads, in order to realize the full economic potential and reduced total cost of ownership (TCO) of open networks, intelligent automation at scale will be key. This cannot happen without the participation of vendors, operators, and systems integrators, because no single company holds all of the necessary technology and expertise to do it alone. For instance, transformation of an operator’s business support system (BSS) might be necessary, or a deep understanding of cloud-native edge computing may be required. One company might specialize in cloud-centric architecture for vertical applications, while another might have specialist knowledge in the 5G standalone (SA) core.
The i14y Lab is just one example of a successful, industry-wide collaboration for O-RAN. The Berlin-based open lab tests disaggregated mobile and transport networking components for their interoperability and integration in order to deliver more choice and flexibility for players. Ultimately, it aims to bring together the brightest minds to help overcome the challenges facing O-RAN and accelerate the time to market for disaggregated network solutions.
True collaboration will require progress in three key areas:
First, an open approach to testing that spans all domains. This open approach should be automated and API-based, with standardized testbenches and performance benchmarks. Collaboration initiatives such as the i14y Lab could provide test tools such as robust, 3GPP- and O-RAN-compliant simulators that can emulate Layer 1, user equipment or base stations.
Second, standardized automation platforms for network operations. These platforms would automate both the network and the alerts that it generates. Besides providing the network operations center (NOC) with greater visibility, automation would also maximize their productivity and make the network more predictable, which aligns with one of O-RAN’s major goals – to achieve a lower TCO than traditional networks.
Finally, an open orchestration architecture. Driven by open APIs, this architecture would include common data platforms and models. The collected data would be open so developers can build use cases around it.
Achieving automation at scale for cloud-native O-RAN
Although the migration toward cloud-native, open networks includes the use of commercial off-the-shelf (COTS) IT hardware, it can’t simply be used as-is. To meet stringent telecom requirements such as five-nines reliability and RAN functionality, COTS gear requires additional components, such as hardware accelerators. The radio unit must also be managed through the same automation framework. In vRAN, many edge data centers will be utilized to host the baseband functionality of the RAN as a software on such COTS gear. This kind of robust and highly distributed infrastructure — supported with tools for monitoring the network and predicting failures — is key to capturing the interest of systems integrators and infrastructure and cloud vendors interested in the mobile market. Their participation ensures that systems integrators do not have to take on all the responsibilities and liabilities associated with implementing cloud-native, open networks.
Both these aspects can generate significant telemetry, which can be leveraged through machine learning operations (MLOps) at scale to develop and test network AI algorithms. This will ensure that the networks will work in a real-world environment. But like COTS IT equipment, the MLOps currently used for cloud workloads will need to be modified to meet telecom’s unique requirements.
This technology is highly complex, and the only way that cloud-native, open networks will become reality and live up to their potential is through broad industry collaboration. The good news is that many operators, vendors and other stakeholders have recognized that need and begun working together to pioneer the future of this important industry.