How to ensure a successful Open RAN deployment – Amdocs

Open RAN initiatives have come a long way. Since the inception of the xRAN Forum in 2016, and its subsequent merger with the C-RAN alliance two years later to form the O-RAN alliance, the industry and ecosystem have made significant progress in defining open RAN architecture and deploying initial use cases.

Starting with the adoption of the O-RAN alliance’s open fronthaul specifications (a.k.a. 7-2x split), the industry subsequently evolved its focus to making open Centralized Unit (CU), open Distributed Unit (DU), and RAN Intelligent Controller (RIC) a reality.

Today, many operators – both new and incumbent – based on their understanding that adopting a fully O-RAN compliant architecture would enable them to deploy new and innovative use cases cost-effectively, are in the process of rolling out open RAN technology and products.

The Telecom Infrastructure Project (TIP), adopting the O-RAN alliance’s specifications, has also been driving open RAN initiatives with a focus on enabling testing and real-world open RAN deployments by bringing operators, vendors, and systems integrators together.

The 3 stages of Open RAN deployment

Stage 1: [Where the industry is Now] – Opening the interface between the radio unit (RU) and distributed unit (DU). This interface is referred to as the fronthaul interface (7-2x split) in the O-RAN parlance. Supporting an open fronthaul interface requires RU and DU vendors to support this functionality. Stage 2: – Opening the interface between DU and central unit (CU) following the 3GPP higher layer split. This interface is referred to as open F1 in the O-RAN parlance. DU and CU vendors must support open versions of this interface to achieve multivendor interoperability. This phase also includes transitioning from proprietary hardware to GPP / COTS hardware platforms for DU, CU. Stage 3: – Disaggregating the CU further, into a control plane and user plane components. This change is augmented by the introduction of the RAN Intelligent Controller (RIC) to host real-time, analytics, SON, & RRM applications, while the CU and the RIC are deployed in edge clouds. Underpinning the Open RAN deployment evolution is the development of open management interfaces for all open RAN components. Here, open, standardized operations and management models are necessary to truly realize the operational benefits of Open RAN. With the potential to deploy extreme automation, this enables operators to drive unprecedented scalability and efficiency into their future Open RAN operations.

State of current open RAN deployments and trials

While some pioneer greenfield operators such as Rakuten in Japan and DISH network in the USA are fully embracing Open RAN, brownfield operators too are now beginning to trial and deploy the standard. The recent developments in North America (US FCC’s open RAN NOI, US Secure and Trusted Communications Networks Act), Europe (MoU among major operator groups), Latin America, and Asia highlight the growing interest in testing and deploying Open RAN.

Some brownfield operators are finding Open RAN solutions – especially multi-technology deployments – to be a compelling proposition for deploying RAN at an affordable cost. Yet, at the same time, for operators and vendors alike, there is an understanding that there is more work to be done in the areas of developing, testing, deploying, and commercializing fully O-RAN standards-compliant solutions.

Open RAN is also garnering interest for private networking needs. Private enterprises are seeking virtualized, cloud-native, future-proof solutions for their private networking needs. Open RAN architecture offers them a compelling option, allowing them to start with 4G and subsequently upgrade to 5G software without swapping any hardware. Specifically, the deployment of cloud-native RAN components on edge stacks (e.g., MSFT Azure, AWS, GCP) combined with core network software and IT applications make this deployment scenario an attractive proposition.

Key imperatives for operators

To accelerate the adoption of Open RAN and realize the benefits, operators must first develop a full understanding of the following:

• Interoperability and performance of a multi-vendor Open RAN
• The total cost of ownership of Open RAN
• The tradeoff between the benefits of disaggregation and systems integration costs
• Level of automation, including the use of ML and AI technologies, to drive performance and efficiency
• Range of use cases and applications, notably at the network edge, that benefit from disaggregated Open RAN
• How Open RAN can co-exist with traditional RAN

To gain these understandings, operators must undertake the following initiatives:

• Test and integrate various disaggregated sub-systems (RU, CU, DU, RIC, xApps, SMO, rApps)
• Evolve to multi-vendor, end-to-end system testing to create a completely deployable Open RAN solution
• Deploy and test architecture and open interfaces in the areas of cloud management, network orchestration, software management, zero-touch provisioning, CI/CD, FCAPS, and assurance
• Deploy and test architectures and interfaces to ML/AI to enhance analytics and closed-loop optimization to drive higher network performance and efficiency
• Conduct field trials incorporating all open ecosystem software components (e.g. ONF SD-RAN, O-RAN OSC, ONAP) with open RU, open CU, and open DU subsystems
• Develop benchmarks, a playbook, and business cases for various deployment use cases and scenarios

Leveraging the open network ecosystem

Amdocs Open Ecosystem Accelerator for Open RAN provides access to open ecosystem partners, as well as open-source-based solutions and services to help operators test, deploy and launch open RAN networks. Amdocs’ capabilities span all phases of the Open RAN lifecycle, including pre-deployment – testing & integration, deployment – orchestration & management, and Operations – intelligence & automation.

Open RAN end-to-end lab services – Services include managing the program, developing end-to-end network design (IP RAN, IP CORE), preparing the bill of material, liaising with equipment vendors, procuring all equipment, as well as integrating and installing the RAN, core, and transport elements. It also includes infrastructure automation, which enables the provision of lab infrastructure as a service for on-premise, hybrid, and cloud environments. Open RAN testing, automation, and integration services – Services include creating test object lists, assembling test plans, executing test plans, performance validation, benchmarking, reporting results, deploying automation for testing, providing lab infrastructure as a service, and reporting to augment CI/CD pipelines as per O-RAN alliance testing specifications and methodologies. Pre-integrated open RAN solution – Services include reselling partner products, deploying and supporting pre-integrated and tested open RAN solutions. Furthermore, Amdocs works with various Open RAN systems providers to ensure the integrated solution meets the operators’ requirements and is ready to be tested and deployed in the field. Open source-based RAN orchestration, management, and automation – This solution utilizes open source software (e.g. ONAP components) to support multi-vendor Open RAN to orchestrate, configure, deploy (support CI/CD), as well as manage PNFs, VNFs, and CNFs in a multi-cloud environment. Capabilities include service designer, O-cloud management, support FCAPs, non-RT RIC, and rApps. Supported use cases include network slicing and assurance. Open source-based RAN intelligence and automation – This solution utilizes open source software (e.g. ONF SD-RAN, O-RAN OSC) to develop and deploy AI/ML-driven analytics and optimization applications. Use cases include near real-time geo-analytics, coverage capacity optimization, massive MIMO/beamforming optimization, QoE optimization, and traffic steering.

O-RAN Interview with Karpura Suryadevara

Which industry organizations are playing a prominent role in developing Open RAN?

The main organizations driving the evolution of Open RAN include O-RAN Alliance, O-RAN OSC (Linux Foundation), ONF (SD-RAN project), and Telecom Infra Project (TIP). O-RAN Alliance focuses on use cases, open RAN architecture, open interfaces, and specifications including testing and integration, while O-RAN OSC and the ONF SD-RAN project provide a software base for some of the components such as RIC. Meanwhile, TIP plays a crucial role in aligning the use cases, deployment requirements, vendors and operators so that first, tests can be conducted in labs and field networks, and subsequently, for the acceleration of commercial deployments.

How is Amdocs involved with O-RAN Alliance?

We’ve been collaborating with the O-RAN Alliance since 2018. Our focus areas include testing & integration, service management & orchestration (including Non-RT RIC, rApps), and near RT RIC applications (xApps). Our contributions include white papers and defining use cases such as massive MIMO & network slice management. We have voting rights in WG1 (Use Cases and Overall Architecture workgroup) and WG2 (Non-real-time RIC and A1 Interface workgroup) and we’re contributors to the 2021 Minimum Viable Plan committee’s (MVP-C) activities.

What about your involvement with Telecom Infra Project (TIP)?

We’ve been collaborating with TIP groups since 2016 with activities spanning Open RAN, Open Core, and Open Transport groups. Within Open RAN specifically, we’re members of the RIA (RAN Intelligence & Automation) and ROMA (RAN Orchestration, Management & Automation) subgroups. Within the RIA subgroup, we collaborate with operators and technology partners to deploy machine learning (ML)-driven use cases on ONF SD-RAN’s near-RT RIC platform. Amdocs has also been selected as part of the preferred vendor list for the first round of use case trials. Within the recently-formed ROMA sub-group, we look forward to contributing to various automation use cases, enabling more holistic and unified Open RAN management.

What benefits have you seen from your collaboration within the O-RAN Alliance & TIP?

In TIP, RIA and ROMA initiatives have been key to driving future RAN automation use cases towards PoCs, trials, and deployments. The RIA subgroup’s operator and TIP technical leads have been particularly successful in defining use cases and aligning vendor capabilities with operator interests in order to build early PoCs. Thanks to this collaboration, we’ve been able to take new ML-based analytics and optimization use cases and develop plans to conduct early proofs of concept, as well as actual deployments. In the O-RAN Alliance, the overall architecture and new interface definitions, including the OAM work for O-RAN, are providing us with a foundation to develop open, multi-vendor automation, while the specifications are also contributing to our own work in developing multi-vendor analytics and automation for RAN.