As open radio access networks (RANs) aim for mass commercialisation, recent technological advancements are expected to strengthen global adoption. Mobile network operators (MNOs) and network infrastructure providers are leveraging next-generation technologies such as artificial intelligence (AI), machine learning (ML), cloud computing and big data to accelerate the automation of software components within the open RAN architecture. The integration of these solutions with open RAN presents numerous benefits, including increased operational efficiency, smarter and more scalable networks, improved time-to-market of new services and functions, and reduced capex and opex of telcos.
A look at how these new-age technologies are enabling the transition of telecom networks towards open RAN…
AI/ML
One of the key aspects of open RAN is the native integration of intelligence into the RAN. In this regard, AI and ML play a key role in the intelligent management and operation of networks. These technologies have a far-reaching and transformative impact on the end-to-end network in an open RAN environment.
The specifications of the O-RAN Alliance provide a framework for AI and ML to optimise radio resources in LTE and 5G networks. The framework heavily relies on AI/ML technologies to improve performance and operation automation through intelligent algorithms that continuously improve the system. The algorithms further enable operators to analyse enormous amounts of real-time traffic data and network load without compromising network capacity. Such technologies allow telecom service providers to adjust network conditions and ensure proper load balancing, facilitating seamless transfer of active call or data sessions between channels or networks. Some use cases of these technologies in the open RAN space are parameter forecasting for energy-saving purposes, anomaly detection in the system, traffic steering and prediction of failures. All of these functions are performed through applications hosted on the RAN intelligent controller (RIC) platform introduced by the O-RAN Alliance, which can be implemented for near-real-time control and non-real-time control. AI-enabled policies and ML-based models generate messages in non-real-time RIC and are conveyed to the near-real-time RIC.
CI/CD
Continuous integration and continuous delivery (CI/CD) is an approach designed to automate the development and delivery processes for new software versions with more frequent updates. The presence of various components from different vendors makes CI/CD essential in the open RAN space. With CI/CD automation, upgrades and downgrades can be performed in seconds or minutes with minimal manual intervention. This can lead to a substantial reduction in costs, time and errors compared to traditional site visits for testing and upgrades. These updates can be monitored to evaluate their impact on end users and their contribution towards achieving business goals. Leading open RAN vendors provide CI/CD frameworks to enhance agility and flexibility for MNOs to launch new features, fix bugs in the system and implement software upgrades.
The key benefits of the technology to MNOs include increased efficiency through zero-touch deployment and testing as well as automated management of increased traffic. It also reduces risks with up-to-date security with gradual and frequent updates instead of large upgrade projects. It also improves service agility by reducing time-to-market.
Cloud computing
Open RAN allows operators to decouple software from hardware, facilitating the migration to a cloud-native model. Cloud computing in open RAN allows the network to be realised in software by helping MNOs leverage cloud-native architectural principles and enabling the deployment of RAN functions as microservices in containers over bare metal servers, supported by orchestration technologies such as Kubernetes. The model enables ubiquitous, on-demand network access to a shared pool of computing resources including networks, servers, storage, applications and services, which can be rapidly provisioned and released with minimal manual management. The key functionality of the model is provided by containerised network functions – software that runs on virtually any off-the-shelf server. The cloud-native open RAN model enables workflow orchestration and network automation to deploy, scale and heal with minimal manual management. It enables automation by taking over infrastructure management and operations, and handling the management of computing, storage and network resources.
Transitioning to a cloud-native environment offers several advantages for MNOs, including gaining visibility into the network, enabling real-time analytics to enhance network performance and providing an improved user experience. The integration of open RAN in a cloud framework can facilitate a multitude of innovative use cases, with service providers, telecom operators and enterprises envisioning market opportunities for cloud-native open RAN technologies across various domains, including 5G and the network edge. Industry experts believe that the involvement of global public cloud providers such as Amazon, Google and Microsoft will be a key driver for the adoption of cloud-native open RAN technology, opening up new opportunities for stakeholders across the telecom value chain.
Virtualisation
Virtualised RAN (vRAN) employs the principles of network functions virtualisation to bring the benefits of cloud computing and a platform-centric approach by deploying RAN network functions on an agile, scalable and hybrid platform using common-off-the-shelf (COTS) hardware. This helps break down silos and minimises MNOs’ capex and opex. vRAN is particularly beneficial for 5G deployments as network virtualisation is essential to support the use cases and performance requirements of 5G.
vRAN can provide numerous benefits for MNOs. It enables a single, uniform hardware platform across the core network, RAN and edge, which can simplify the management of the entire network, reducing operations and maintenance costs. The technology enables interoperability among RAN components and enhances supply chain security. Furthermore, virtualisation offers improved operational efficiency by utilising industry-established components for common tasks, eliminating the need for vendor-specific solutions and allowing operators to focus on business-specific components. This also means that RAN network functions from multiple vendors can run on the same hardware, increasing flexibility for the service provider. In some cases, the hardware can even be shared among vendors. Moreover, a widely adopted open platform can lower barriers to cross-domain innovation, facilitating the development of new use cases and services.
Big data analytics
In the open RAN architecture, big data analytics is deployed as rApps in the non-real-time RIC. The technology provides an overview of the network conditions through visual representations of patterns or abnormalities in the network, which helps MNOs comprehend and improve network performance for a better user experience. It facilitates the review of data and reports generated by AI and ML tools in the network.
The 3GPP 5G data analytics standards are critical to facilitate the integration of 5G RAN system components from differnt vendors. For example, 5G network data analytics function (NWDAF) specifies standard data types, formats, data collection application programming interfaces (APIs) and data outputs for analytics processing. NWDAF represents the mobile industry’s first attempt to standardise the function of analytics in the mobile core network. It incorporates standard interfaces for collecting different types of data from certain 5G core network functions and applies the results of analytics processing to inform the operation of other network functions, leveraging machine intelligence for network automation and service orchestration. Experts suggest that operators that incorporate 3GPP-compliant data analytics into their networks can scale up and manage 5G deployments cost effectively.
Zero-touch provisioning
Zero-touch provisioning involves automating software installation and configurations. It significantly reduces complexity in provisioning, saves time, minimises errors and lowers the cost. The technology is considered safe for RAN installation. It helps in the deployment of hundreds of sites without any manual intervention. Although achieving automation and zero-touch deployment may take time to materialise, it is essential for the dense deployment of 5G networks, especially when a large number of sites need to be configured.
In sum
The adoption of open RAN is steadily gaining acceptance among telecom sector players, mainly due to the emergence of next-generation technologies. However, there are still technical challenges that need to be addressed, such as high integration complexity in multivendor environments, security concerns and integration issues with brownfield operators. Nevertheless, as 5G and private network use cases continue to mature, virtualisation and automation will become increasingly essential in the open RAN environment. The open standards promoted by the O-RAN Alliance will further help leverage these technologies to develop open architectures and accelerate their deployments. Going forward, 2023 has the potential to be a defining year for disaggregated, multivendor open RAN architectures, which are set to play a key role in the roll-out of 5G and future network generations.
