If 5G is to succeed in supporting use cases such as augmented reality and smart cities, the industry needs access to an open disaggregated Radio Access Network (RAN) that gives operators the flexibility required to meet the network challenges of 5G cost effectively.
While much of the network has already benefited from a wave of disaggregation and virtualization in support of 4G/LTE, the RAN remains the last piece of the network to be disaggregated.
This blog will start out by providing the background that has led to the industry's push for the adoption of an open, disaggregated RAN.
Before it all started
In the not-too distant past, operators primarily used proprietary, closed systems from traditional vendors that provided expensive monolithic solutions and required that the solution be purchased as a whole package. The RAN was no exception operators were limited in their options as these solutions lacked hardware and software separation and interfaces were proprietary and non-interoperable. What worked with one vendor didn't always work with another.
This lack of choice led to inefficiencies in the network as interfaces were tightly coupled with proprietary solutions, giving operators little visibility into their RAN nodes and spectrum efficiency. Furthermore, operators were challenged to introduce new services as the hardcoded logic and algorithms in these solutions offered little or no flexibility.
The industry needed to adopt virtualization trends and reduce capex/opex in RAN deployments.
The industry begins to solve the problem
The first steps toward disaggregation started with the C-RAN (Cloud RAN or centralized RAN) initiative from IBM, Intel and China Mobile. C-RAN resulted in a deployment model where a baseband unit (that hosted all of the base station function except the radio) could be located in a data center or a far-edge location, while the radio Remote Radio Head (RRH) was connected to the baseband unit via a dedicated high-bandwidth connection.
This wasn't without its challenges. The C-RAN model required a new Fronthaul interface, and various industry standards such as the Common Public Radio Interface (CPRI) and the Next Generation Fronthaul Interface (NGFI) evolved to enable these new interfaces between the base stations and the RAN.
As a natural evolution of C-RAN, multiple companies and industry initiatives started looking into splitting/disaggregating RAN functionalities. There were multiple study reports published that looked at the merits and challenges of various options:
These standards organizations began to prove that these initial steps towards disaggregated architecture were technically sound. Base stations (LTE and 5G) began to evolve, and three important elements of a fully disaggregated RAN began to emerge: the Centralized Unit (CU), the Distributed Unit (DU), and the Radio Unit (RU).
One of the other trends fueling the push for disaggregated RAN adoption deployment is Multi-access Edge Computing (MEC). As cloud solutions grew and the 4th Industrial Revolution gained ground, industries and operators alike recognized the need for increased computing demands delivered with greater speed and lower latency, which could be achieved by pushing compute from the data center out to the network's edge. Operators realized that, given RAN functions, MEC can colocate at some edge locations, allowing them to provide increased network services where central office-based solutions could not.
Creating alphabet soup?
As industry initiatives demonstrated the viability of a disaggregated RAN, operators began looking for vendors capable of enabling a disaggregated RAN. This led to a number of vendor companies forming their own initiatives to develop their ecosystems and promote interoperability among their partners.
At the same time, industry initiatives were launching to solve interoperability challenges and to define nodes and interfaces in a disaggregated RAN. Unlike vendor-led initiatives, which tended to orient toward specific products, the Standard Development Organization initiatives sought to align and standardize reference architectures for deployment scenarios, define truly interoperable interfaces and accelerate deployment of solutions based on Open RAN philosophy. Three of the key SDO initiatives include:
- The OpenRAN Project Group: An initiative within TIP working to define and build 2G, 3G and 4G RAN solutions based on a general-purpose vendor-neutral hardware and software-defined technology.
- The O-RAN Alliance: An operator-founded organization defining standards for next-generation RAN architecture and open interfaces across all use cases with a heavy focus on operator deployment needs and to help build a supply chain ecosystem.
- The Small Cell Forum: This organization focuses on accelerating small cell deployments, removing commercial and technical barriers, gathering requirements globally from service providers and enterprises, and defining FAPI (functional application programming interface) and network FAPI (nFAPI) standards for 4G and 5G.
While the launch of so many initiatives demonstrated the industry's desire to break free from a traditional proprietary model, it also had some unintended side effects. It created some confusion since all of these initiatives were focused on the RAN issue, the abbreviations and names for the initiatives were all similar sounding – CRAN, VRAN, DRAN, O-RAN, OpenRAN, Open RAN, etc.
This also led to a perception that these standards are not aligned and do not interoperate well together.
How it all falls into place
To help clear up some of the potential alphabet-soup confusion, it's important to be clear on the terminology. The term "open RAN" is a broader term for RAN solutions built on open commercial off-the-shelf (COTS) hardware and open interface standards.
The open RAN standards initiatives accelerate deployment of solutions based on open RAN philosophy.
- The O-RAN Alliance bases its specifications on 3GPP specifications and liaises with 3GPP and other standards organizations as required. The O-RAN Alliance has a dedicated working group on white box hardware.
- Recently, TIP signed a liaison agreement with the O-RAN Alliance. This agreement enables TIP's OpenRAN projects to refer to the O-RAN Alliance specifications. TIP focuses on open hardware designs to enable quicker adoption and deployments.
- The Small Cell Forum has a formal liaison with the O-RAN Alliance to work on alignment for small cells and DU solutions based on open RAN.
Momentum for open RAN
All of the above is good news for operators looking for open RAN solutions and for vendors enabling this evolution. Because the specifications are focused on interoperability, operators can look for solutions that meet their needs while vendors are able to develop and innovate products within their area of expertise.
Over the last couple of years, there has been good progress and greater alignment with open RAN initiatives across the industry, resulting in many operators signaling their intent to deploy Open RAN solutions. There have been 4G open RAN deployments already in a few countries, and many operators are issuing RFPs with O-RAN compliance as a requirement for their 5G networks. Radisys has engaged with a number of operators to test disaggregated RAN solutions in the lab and then seen those projects move from proof-of-concept to field trials.
The new RAN vendor ecosystem
The emerging trend of new architectures and interfaces enabling disaggregated RAN solutions along with COTS hardware options, all aligned around standards and specifications, is fundamentally changing the RAN vendor ecosystem. Operators are no longer locked into inflexible one-size-fits-most solutions. Instead, there is now space for newer players specializing in hardware and software components that deliver high performance, flexibility and increased visibility into RAN nodes and spectrum efficiency at optimized price points.
Ganesh Shenbagaraman, Head of Integrated Products and Ecosystems, Radisys