• Making something as complex as an Open RAN deployment work properly is a tough ask, especially when multiple radio access technologies (RATs) are involved 
• Once you disaggregate and open the RAN you need extra help to control, optimise and continually refine all the new functions that become available
• Fortunately help is on hand – the RAN intelligent controller (RIC) is deemed the way forward for Open RAN
• RIC developments were discussed during our recent Open RAN Summit

For mobile network operators considering the potential of Open RAN, the potential of the RAN intelligent controller (RIC), which is responsible for managing and optimising multiple functions in a multi-vendor disaggregated radio access network, is a critical consideration for the future. 

That’s why the second day of TelecomTV’s recent Open RAN Summit saw our panelists concentrate on the crucial importance of the two categories ‘non-real time’ and ‘near-real time’ during the session – see How can telcos maximise the value of the RIC for network efficiency gains.

To step back slightly, the much lamented rise in complexity apparent with 5G has much to do with its so-called multi-radio access technologies (RAT) characteristic. Having multiple RATs working together introduces a rat’s nest of complexity when added to the management complexity already being introduced by the disaggregation inherent in Open RAN itself. The RIC is key to overcoming this.

The non-real time (non-RT) controller assigns events and resources requiring response times of one second or more and is usually located centrally. Then there’s the near-real time (near-RT) controller, designed to respond to its resources in as little as 10 milliseconds and, therefore, more likely to operate at the very edge of the network.

In essence, the non-RT controller is responsible for jobs like orchestration across the whole network. Because of its one second plus response requirements, it’s able to gather info from all corners of its network to make orchestration decisions. The near-RT controller, on the other hand, must make near-instant decisions (a few milliseconds) governing the operation of the RAN at the cloud edge by hosting so-called xApps which can, for example, manage spectrum efficiency. 

So is the RIC hype justified? “Yes,” our Open RAN Summit industry expert speakers replied emphatically.

They said that the RICs were already making a huge difference (improvements are always needed, of course) to Open RAN manageability in tests and trials and their adoption was crucial for operators as they track towards AI-enabled automation.

According to Constantine Polychronopoulos, group vice president of 5G and telco cloud at Juniper Networks, the impact of the RIC will be huge. “If we think back, we might remember that we used to have [management] applications customised on different hardware platforms. Then Linux came along and created a unification layer. That meant that we no longer talked about the portability of applications or worried about who the RAN provider was. It’s the same with the RIC. We can upgrade the network in a seamless way to open it up to new business models.”

Patrick Kelly, founder, partner and principal analyst at Appledore Research, added, “The RIC is the first time we’ve had an open system for application developers. [It means] you’re able to control the performance of the radio access network (RAN) and some of the individual elements of the RIC. The near-real time RIC opens up a finer level of control for app developers,” he claimed.

Kelly maintained he’s already seeing new entrants coming in and innovating faster than the incumbent RAN vendors. That, in fact, is one of the expected bonuses for the industry flowing from the adoption of Open RAN, a sentiment echoed by Lucia De Miguel Alberto, senior Open RAN manager at Vodafone, who said the RIC is the enabler for innovation in the RAN.

The incorporation of the near-real time RIC into a multi-vendor Open RAN deployment should allow what the panel described as “all sorts of fine-grained stuff,” including slicing and the development of quality of service features as well as the prioritisation of emergency services and so on, thus enabling operators to refine their use of available spectrum and reduce their energy consumption.

Rimma Iontel, global chief solutions architect for telco, media and entertainment at Red Hat, pointed out that the “RIC allows you to dynamically turn off carriers, change radio frequencies and make other adjustments. You can use so-called rApps [which use information from the RIC] to track trends on the utilisation of the network and then, based on that, you can decide to turn off certain carriers or cells, change frequencies or force handoffs to balance the network and so on.”

Network operators claim RIC benefits can go far further than nitty-gritty RAN optimisation and control. It opens up new vistas and possibilities around business models and ecosystem partnerships. Broad innovation is the real goal, not just technical tinkering.

“For BT, the RIC is the key element in its supplier engagement process,” said BT’s distinguished engineer, Richard McKenzie. “It means we can pick and choose different components from the entire ecosystem and then combine them [using the RIC] to create the services we want for our customers. We’re not there yet but it’s so important,” he added.

“We really want the flexibility. When there’s a new service that we want to deploy or a new situation arises, we want to be able to reconfigure quickly and easily.”

When it comes to radio features, McKenzie pointed out that it’s difficult to ‘out-configure’ a RAN vendor’s own integrated radio. “Normally they’ll have a feature set that’s fully optimised and they’ve done a really good job, so it’s a tough task to outperform a well-optimised single RAN solution… but with the RIC we have access to any of the capabilities from anywhere in the ecosystem, so we get this much larger feature set.” 

He said this means he can piece features together to create a new service via both the real-time and non-real-time RICs. “Over time we can use the larger feature set to outperform the traditional RAN.”

The RIC’s ability to apply some fine-grained multi-vendor tinkering right across the network is also important, he noted. “Spectrum is still one of the largest expenses for operators. If we can use the RIC to improve spectral efficiency, even by just a little bit, then that has huge value. The same dynamic applies to the use of the RIC to optimise power consumption.“

All our panellists tempered their enthusiasm over the RIC’s development with occasional health warnings, such as “there’s some way to go still,” or “we’re not quite there yet”, but overall the summit demonstrated that the RIC was in good health and making progress.

Of particular concern at the Open RAN Summit generally was the perennial conflict between standards evolution and innovation. “Is it possible to innovate in an open environment with a view to your innovation becoming standard eventually rather than necessarily always waiting for a standard before you move forward?” asked Appledore Research’s Francis Haysom. 

The answer? “Yes, but you must be confident that you have the expertise to do well in a particular area. No company has the foresight, resources or scope to be leading across the board,” he added. 

And that observation really loops back to reinforce the rationale for engaging with a broad ecosystem to share expertise and specialisms: That’s surely one of the main reasons why Open RAN was envisaged in the first place.

- Ian Scales, Managing Editor, TelecomTV