Open RAN Progressing on Network Interoperability and Performance with AI/ML

Open Radio Access Networks (Open RAN) is a concept that involves the interoperability of open hardware, software, and interfaces for cellular wireless networks. As new Open RAN developments progress, 5G Americas, the voice of 5G and 4G LTE for the Americas, today announced the release of “The Evolution of Open RAN”, which provides an update to a 5G Americas’ 2020 report on Open RAN standards. The new white paper focuses on market updates and the latest developments including artificial intelligence and machine learning tools. Chris Pearson, President of 5G Americas said, “Open RAN is rapidly evolving, as efforts are consolidating globally with new momentum driving the evolution of this technology forward. This continued progress requires governments, vendors, academics, and operators working closely together.”

 

“The Evolution of Open RAN” provides a comprehensive update on work from leading Open RAN standards bodies with an emphasis on application and microservices development in modern wireless cellular networks. The paper also examines how cloudification of Open RAN is providing opportunities in scalability of network functions for operators. The paper examines how architectural considerations may include network disaggregation and functional splits, hybrid vs. hierarchical management plane (M-plane), RAN cloudification and virtualization, and services-based architecture for RIC functions. Additionally, it examines how operators will have different Open RAN needs based on their own requirements for TCO savings, performance considerations, and brownfield vs greenfield deployments.

Some key topics surrounding evolving Open RAN standards in this 5G Americas report include:
    •    Open RAN and goals
    •    Principles of O-RAN, architectural considerations
    •    Operator trials and deployments
    •    Operational considerations and integration challenges
    •    Advantages and challenges with Open RAN architectures
    •    O-RAN use-case realization using AI/ML
    •    Interface service models for AI/ML-enabled xApp design in Near-RT RIC
    •    Advanced learning algorithms for xApp and rApp development in the non-RT RIC

Dr. Rajarajan Sivaraj, Director of RIC Architecture and Standards, Mavenir said, “Open RAN offers potentially new capabilities in network performance, end-user experience and service level assurances for enterprises. Open RAN can provide the ability to harness leading artificial intelligence and machine learning tools, cloud native infrastructure and tools, open interfaces between network functions, and exposure of fine-grained user-level RAN state and performance information of the network to novel third-party optimization applications. He continued, “With focus on inter-operability, Open RAN could help MNOs to simplify complex network operations and deliver enhanced performance capabilities in terms of high bandwidth, low latency and high reliability to enrich the end user experience.”