The Voice of 5G & LTE for the Americas

Select a Language: Spanish | Portuguese

An Update on Open RAN

Feb. 2023 – Chris Pearson, President, 5G Americas

As the president of a wireless trade association, I occasionally get a ring-side seat to fascinating conversations about technology and developments in the cellular industry. One of those topics is, of course, Open RAN. Open RAN (Open Radio Access Networks) has significant momentum and yet engenders robust and passionate conversations among the biggest names in the industry.  

But what is Open RAN? What new developments have taken place since we introduced you to the topic a couple years ago?

If you’re already familiar with Open RAN, 5G Americas has just released its update on the topic with our latest white paper “The Evolution of Open RAN”, which covers key areas of progress in the ongoing conversation.

However, if you need a quick refresher, our white paper “Transition to Open and Interoperable Networks” is a very good introduction to the topic. Briefly, Open RAN is a concept that focuses on interoperability and standardization of radio access network (RAN) elements among different software and hardware providers. It is an ongoing shift in mobile network architectures that provides industry-wide standards that telecommunications suppliers can rely on when producing network equipment.

Basically, Open RAN principles include:

  1. openness of interfaces between different functions or logical nodes in a network
  2. virtualization of network functions to run on commercial off-the-shelf (COTS) hardware
  3. intelligent management of control, user and management plane (M-Plane) functions
  4. programmable targets for optimization and configuration

By standardizing every element of the radio access network and allowing sophisticated virtual network functions to run on interchangeable commercial off-the-shelf hardware, Open RAN hopes to create an ecosystem that is more resilient to supply chain disruptions and is quicker to innovate. It creates a sort of standard playbook that helps manufacturers and software developers more quickly create new improvements and enhancements to wireless networks, much in the same way that advances in the Android operating system may have helped to stimulate active growth in new applications and use cases.

As you can imagine, trying to pull together open standards for an entire industry is not always easy, but progress is being made. Open RAN standards are led by the O-RAN Alliance, Telecom Infra Project (TIP) and 3GPP, with each focusing on different aspects of the challenge.

The O-RAN Alliance is responsible for defining standardization of Open RAN systems, which involves several different workgroups focusing on different elements of architecture and interfaces between the RAN elements. The Telecom Infra Project (TIP) is focused on collaboration and development of new technologies for deploying global telecom network infrastructure and integrating it into existing legacy cellular networks. The Third Generation Partnership Project (3GPP) forms the core standards and specifications for 5G and other wireless cellular networks, while the Open RAN Policy Coalition takes on the role of advocating for policy changes in the United States to reflect Open RAN goals.

As work has proceeded at these standards bodies, deployment of various elements of Open RAN have been undertaken by some mobile network operators. This chart below shows a timeline of some of these advances:

Fig. 2-13: Open RAN milestones, “The Evolution of Open RAN

Since the last 5G Americas white paper, progress has been made in several key areas, such as:

  • services-based architecture for service and management orchestration, as well as for real-time and non-real-time RAN intelligent controller (RIC)
  • improvements to network disaggregation and functional splits
  • improvements in service models and definitions in several key RAN interfaces
  • addressing cloudification and orchestration in the O2 interface
  • hybrid and hierarchical architectures for M-plane (management plane)
  • standardized APIs for certain elements
  • new interfaces for RAN analytics
  • improved management of network slices

Additionally, since the last white paper, two areas have seen rapid growth and development for Open RAN: cloud computing and artificial intelligence & machine learning (AI/ML).

The cloudification of Open RAN provides opportunities in scalability of network functions for operators, as many of those improvements are listed in the bullet points above. These improvements are paving the way for continued integration with public, private and hybrid clouds, which each individual operator will need to assess based on their own requirements for TCO savings, performance considerations, and brownfield vs greenfield deployments. The figure below shows how logical clouds might begin to integrate into Open RAN networks.

Fig 2-10: Logical clouds, “The Evolution of Open RAN

The other area of progress in Open RAN involves the use of artificial intelligence and machine learning, as they provide new capabilities in network performance and use cases. The recent rise in AI/ML is having a broad impact across several areas of network management, but are specifically acute in the following:

  • Delivery and development of apps and microservices that can take advantage of AI/ML advances in Open RAN networks.
  • AI/ML analytics, framework functions, and life cycle management differ based on near real-time or non-real time RIC characteristics
  • Interface service models and application design differ based on near-RT or non-RT requirements
  • Advanced learning algorithms are being developed for xApp functions that control network operations in the RAN, as well as rApp functions that control functions in a non-RT environment.

The use of AI/ML in the RAN happens at a very deep level within the RAN Intelligent Controller (RIC). As microservices, xApps, and orchestration occur, the RIC works to predict and allocate the necessary resources to ensure delivery. Learning engines embedded in the RIC use deep learning neural nets to discover optimal resourcing based on learned models and data analysis.

Figure 3-5 : Reinforcement learning engine in the Near-RT RIC, “The Evolution of Open RAN”

Suffice it to say, things are getting complex very quickly with the integration of cloud computing and AI/ML in modern wireless cellular networks. All of this complication is requiring a greater degree of cooperation, communication, and flexibility among our wireless companies – as they each try to navigate their way through the next few years on Open RAN. It’s clear that all the key stakeholders from throughout this growing mobile wireless communications industry will need to continue to work closely together for the deployment of open, disaggregated RAN systems.

The mobile wireless industry is built on resilience, communication, standards and cooperation and will continue to find new ways to innovate for our customers throughout the world.  

-Chris

Find out more on the progress of Open RAN

Learn about recent advancements in Open Radio Access Network (RAN) standards and the evolution in Open RAN trials and deployments.

Share this post

Sign up to receive our announcements