Interoperability and better industry standards – that’s the premise and promise of Open RAN, apparently. But are we closer to solving cartelization, over-engineering, customer pain points, and de-innovation? Or are we feeding these beasts with a new hand?
By Pratima Harigunani
It sounds impossible. But there is actually something called the world’s longest-lasting light bulb. It is in a Fire Department in Livermore, California. It is claimed to have been installed in 1901. It has been turned off only a handful of times.
But what sounds more incredible is the flip side of this marvel. If it is, indeed, possible for a bulb to stay chuffed and alight for years and years; then why don’t companies make one for the everyday customer? Well, that’s where we start talking about industry cartels. Like the Phoebus cartel. Where top industry players got together to test and apply standards so that bulbs did not exceed the permitted life of 1000 hours. But why did they not get together and make standards for extending the life and quality of a bulb?
After all, a good industry could do that and ‘should’ do that. Because as contradictory as it may sound, joining hands can actually improve margins, enhance business and promote innovation.
Wait! Did someone cough – ‘yeah, on paper!’ ‘In your dreams!’
We have a good case on the table to find out whether it is idealism or realism. Open RAN is in progress. Maybe it will light up the world for Telcos and their end customers. Maybe not.
Let’s see why.
Inside the Bulb - What is RAN and why to ‘open’ it?
RAN or Radio Access Network is what connects individual devices to other parts of a network through radio connections. It is a type of network infrastructure that wirelessly connects user equipment to a core network. It encompasses a baseband unit, radio unit or remote radio unit, antennas, and software interfaces. The way it typically works is this - Data from a user gets to the network core after being received by a radio unit and is translated into a digital format by a baseband unit. So far different generations of mobile networks use different variations of RANs. That’s where Open RAN steps in.
Simply put, Open RAN is a concept and an industry-wide endeavor (to some extent) for standards for RAN interfaces so that interoperation between vendors’ equipment can be supported. The core drivers are flexibility, multi-vendor choice, better visibility, disaggregation of hardware and software, and open interfaces. It should also lead to open space for new players and innovations to enter the market – and easily.
Wait - is there a demand for Open RAN? Stefan Pongratz, VP and analyst for RAN and Telecom Capex, Dell’Oro Group, Inc. points out that 80 percent of the top 20 wireless service providers in the world are investing, deploying, or exploring Open RAN. The drivers, he reckons, will vary depending on who we speak with. “But when it comes to the broader movement, one of the leading drivers from a carrier perspective is the degree of competition in the RAN market.
Opening up the interface between the baseband and the radio should, in theory, assuming everything else remains constant, reduce the staying power of the incumbents. While there are some early indicators that the ongoing efforts by the US government to curb the rise of Huawei are starting to show in the numbers outside of China, regulators, governments, and operators remain concerned about the competitive dynamics and the implications of limited supply chain diversification can have on prices, innovation, and security.” And not all Open RAN is the same (many flavors of Open RAN), he reminds. “But operators are clearly warming up to the idea or the expectation that this interface will open up.”
John Baker, Senior Vice President, Business Development at Mavenir spells out the ‘need’ part on two specific areas. “Open RAN will enable vendor interoperability and vendor diversification. By opening the market and introducing competition amongst industry players, Open RAN sets up a rift between traditional incumbents, that have kept the market closed with proprietary interfaces, and forward-looking new entrants. Open RAN will likely force incumbent vendors to shift their business models away from hardware to a more software-centric approach, introducing new business and competitive risks through the transition.”
Interoperable interfaces allow smaller vendors to quickly introduce their own products and services, he explains. “They also enable Mobile Network Operators (MNOs) to adopt multi-vendor deployments and to customize their networks to suit their own unique needs. MNOs will be free to choose the products and technologies that they want to utilize in their networks, regardless of the vendor. As a result, MNOs will have the opportunity to build more robust and cost-effective networks leveraging innovation from multiple sources.”
This idea of open space can help operators have interoperability for RAN elements with a white box approach to both hardware and software. It can liberate them from proprietary lock-ins and chaos. And the current RAN landscape can use a lot of this opening up. Because only a few RAN vendors offer equipment and software and most of it is totally proprietary.
The chief (or shall we say the most popular) effort in this direction is the O-RAN Alliance which aims for a well-documented architecture for open RAN interfaces and elements. It has been claimed that the new O-RAN standards will enable a more competitive and vibrant RAN supplier ecosystem. This will help faster innovation to improve user experience, while O-RAN-based mobile networks can improve the efficiency of RAN deployments as well as operations by the mobile operators.
The word interoperability is all pixies, and rainbows and marshmallows to hear of. But is it really a common incentive for everyone in the industry to achieve it? Even if that’s possible, can we ever achieve true and on-ground/last-mile interoperability?
Rémy Pascal, Principal Analyst, Mobile infrastructure, Omdia points out that interoperability existed before open RAN, for example- interfaces between the user device and RAN and between the RAN and core were already opened and this is why, with rare exceptions, any smartphone from any manufacturer can connect to the network of an operator. “This is also why operators can select different RAN and core vendors or use multiple RAN vendors with a single core vendor.”
What is new here, he clarifies, is the improved interoperability within the RAN domain. “And compliance with the O-RAN Alliance front-haul interface, indeed, facilitates interoperability. But it is also worth mentioning that this can be implemented without, necessarily, complying with the O-RAN Alliance specifications. This can be achieved (though it’s relatively rare) via a bilateral agreement between two vendors and interoperability work, generally at the request of their client, the operator.”
Critics also say that this is nothing but just another close-knit huddle of like-minded industry players. Ask Remy Pascal, a Principal Analyst in the Service Provider Networks team, Omdia and who has more than 10 years of experience in the telecoms and ICT industries and he would break it down a bit. “There are several organizations involved in open RAN specifications and development including the 3GPP, the O-RAN Alliance, the Telecom Infra Project (TIP), the Small Cell Forum, the Open Networking Foundation (ONF) and probably others. 3GPP is the main mobile standards organization and its work covers the entire mobile system. The O-RAN Alliance develops additional specifications focusing on open RAN. The O-RAN Alliance specifications are intended to complement (not replace) the 3GPP standards.” he reasons.
But was there a need or relevance for something like this. Did we not already have 3GPP and GSMA around? Pascal from Omdia argues that Open RAN is initially a service-provider (rather than vendors) led initiative that aims to create open interfaces, and the key driver is to give operators more freedom to choose suppliers of the various sub-systems/components of the RAN system. So operators, or at least some of them, felt there was such need.
According to the O-RAN alliance, some practical aspects are really being addressed. O-RAN claims that it conducts worldwide plug-fests to demonstrate the functionality as well as the multi-vendor interoperability of open network equipment. It also states that the O-RAN specification effort builds on common standards and cares about alignment with other industry bodies to ensure compatibility and to avoid duplication of work.
At the last count (Around June 17, 2021), the O-RAN ALLIANCE published another 33 specification documents, and these included - Management interface, timing and synchronization for Transport, the baseline for Non-Real-Time Radio Intelligent Controller (Non-RT RIC), and Near-Real-Time RIC Architecture and interfaces (E2, A1, F1/X2/Xn). It also covered certification and badging guidelines and parts of end-to-end, interoperability, and conformance testing.
Incidentally, Network Software Provider Mavenir and Vodafone have completed the first data and Voice over LTE (VoLTE) call across a containerized 4G small cell Open RAN solution in a lab environment. As per a press statement, it was revealed that the completed tests are the latest steps forward to delivering an open and vendor-interoperable 4G connectivity solution for small to medium-sized office locations. The plug-and-play small cell equipment can ensure comprehensive mobile coverage in every corner of the office, as explained.
The two companies now intend to focus on finalizing the packaging and automation of the solution before beginning trials with selected customers. As Andrea Dona, Chief Network Officer, Vodafone UK, said in the announcement; “Open RAN is opening doors to simplified and intuitive connectivity solutions. For our wider network deployment strategy, Open RAN is enabling us to work with a wider pool of suppliers and to avoid vendor lock-in scenarios that might prevent us from taking advantage of the latest innovations. The same could be said for enterprise connectivity solutions.
Stefano Cantarelli, Executive Vice President and Chief Marketing Officer, Mavenir, underlined Open vRAN as a very flexible architecture that can serve any type of segment and called this an opportunity to show that automated and AI-controlled systems will simplify life to business and industry.
A long way from the incandescent days
While progress and experiments are going on in full steam, there is another angle to this crusade. If we had stuck to the carbon materials as a bulb filament, no one would have innovated towards tungsten and LED. But why would any player do that when the industry fixed a 1000-hour standard?
Will the tenets of standardization and interoperability not be tough to balance on a see-saw that also has competitive innovation and patents that industry players are investing in? Especially when these areas are gaining remarkable weight now?
“This is an interesting question, but I don’t have a simple answer,” tells a candid Pascal. “Some argue that opening interfaces facilitate the entry of new vendors which, in turn, encourages more competition and not an only price-based competition but also technology/innovation-based competition. Others will say that this model requires vendors, or at least a majority of them, to agree on supporting a specific feature for it to be added to the specifications, and if that’s not the case then this could block or slow down the introduction of new innovative features. So there are opposing forces.”
Pascal maintains that this does not mean, however, that innovation will not take place in other areas which are not covered by the specifications. “Another thing to keep in mind is that open RAN does not mean the end of proprietary technology. The software code remains the property of vendors and the O-RAN Alliance also has an IPR policy in place and its members can make their intellectual property available to others under licenses on fair, reasonable, and non-discriminatory (FRAND) terms, which is a common practice in this industry.”
As to achieving true and on-ground/last-mile interoperability, Pongratz from Dell’Oro Group argues. “If it is a must or wants is a different question. The incumbent suppliers have done an astonishing job with the proprietary RAN architecture managing the asymmetry between data traffic growth and topline RAN revenue growth because at the end of the day – the RAN market has advanced at a 1 percent CAGR between 2000 and 2020 while mobile data traffic has grown nearly 1000x. At the same time, we are still in the early days when it comes to blanketing society with connectivity and clearly, some believe that the current market concentration could impact pricing and innovation in this next phase of this journey.”
True interoperability, as he poignantly underlines, will, of course, be challenging and take some time given the sheer size of the installed base and the fact that the O-RAN spec is not part of 3GPP.
Baker from Mavenir avers that without complete and standardized interfaces, interoperability is impossible. “Testing is, therefore, essential to ensure compliance with standards. Simply cobbling together a few instruments and running a few tests is not an adequate solution. Testing each section individually to the maximum of its capabilities is critical. Choosing and implementing the right equipment for your network requires proper testing with the right tools, methodologies, and strategies. Therefore, it is important to test each component of the RAN in isolation for conformance to the standards and testing combinations of components for interoperability.”
Due to more sizable budgets and experience, established industry OEMs may maintain more advanced or mature Intellectual Property (IP). “While likely an accurate statement at this juncture, Open RAN’s open and extensive ecosystem can ultimately be expected to drive innovation across the ecosystem and create valuable IP.” He hopes.
Ikea Bulbs - Adding V to RAN
Is it a coincidence that these ever-lasting bulbs are found in fire departments? IT has been the designated firefighter for businesses for a long time. And that’s why new fire-hoses like virtualization can help in keeping these bulbs well-lit.
Virtual RAN (vRAN) describes a type of RAN implementation where RAN hardware and software are disaggregated, and it is possible to run some of the RAN functions on generic servers instead of purpose-built hardware, explains Rémy Pascal, Principal Analyst, Mobile infrastructure, Omdia. “And open vRAN describes a RAN implementation where both open RAN and vRAN principles are applied.”
Baker from Mavenir adds that the move to a software-centric approach due to Open RAN is creating momentum for several industry-led open virtualized RAN initiatives that seek to unite an ecosystem of supply chain partners and advance open RAN through the definition, development, and testing of standards and reference architectures. Beyond the standards defined by 3GPP, multiple industry groups like Telecom Infrastructure Project, O-RAN Alliance, Open RAN Policy Coalition, Open Networking Foundation are leading the Open RAN movement, each with a different purpose.
When RAN uses Virtual RAN principles and technologies; it leads to easy automation, less network complexity, more network malleability, enhanced, and reduced CapEx and opex costs.
Really? That depends on a lot of other questions too.
A Flickering Light – Hope or Stress
As to what challenges envelop this road now or ahead, the list is still in the making.
There is no shortage of challenges and opportunities when it comes to the RAN market. In the reckoning of Pongratz from Dell’Oro Group, one of the main points of this picture is to show that the suppliers have to be extremely nimble to keep up with the pace of change. “We have seen more architectural and technological advancements over the past three to five years than we did in the first 30 years rolling out 1G-3G. So one of the reasons we are more optimistic about the world today than we said in the 3G to 4G transition is simply the fact that there are so many more efforts taking place in parallel not only on the supply side but also on the demand side of things.”
He also spells out the 5G context and future implications here. “Yes 5G and MBB are still extremely important and they drive the lion’s share of the market still but at the same time, 5G is only one piece of the puzzle. So at a high level, one of the key challenges going forward will be for both incumbents and suppliers with weaker footprints to figure how to optimize finite resources to maximize the likelihood of success with all the various opportunities.”
For Pascal from Omdia, the aspect about challenges is a broad question but assuming we focus on open RAN here, there is still some work to do in terms of specifications and standards - but they are maturing and continuously evolving which is the normal process in this industry, so not, necessarily. something to be too concerned about.
“Then, besides that, they are many other considerations for operators when they design/build/modernize their mobile networks. Aspects such as performance, energy efficiency, reliability, security, flexibility, cost, and more also have to be taken into account.
Both the open model and the integrated model have pros and cons; there are trade-offs. Operators assess all these dimensions and then they select the solutions that best suit them.
Finally, it does not have to be black or white, operators can combine different approaches, models will co-exist.” Pascal augurs.
Plus, there are other implications to get ready for. The imminent arrival of 5G, Edge Computing, private LTE, can translate into a lot of roadwork for Open RAN. Pascal from Omdia points out that 5G is already largely happening. “There were over 160 commercial 5G mobile networks around the World (as of June 21) and most of them are not open RAN compliant (Rakuten would be one exception), or only very partially. So 5G is effectively happening before Open RAN. That being said, we also understand from our conversations with operators that they will increasingly mandate compliance to some or all principles of Open RAN in the future (generally following the O-RAN Alliance set of specifications), so open RAN principles will increasingly influence procurement.” As for private networks, he feels that it is one of the use cases that several operators are considering for open RAN.
Another angle worth pondering over before it is too late is this. How can these tenets of standardization and interoperability be balanced with competitive innovation and patents that industry players are investing in? Baker from Mavenir argues that Open RAN does not tell you how to build the product, which can then be built-in innovative and creative ways in both hardware, software, or both. “A balanced framework for standardization and interoperability is key to incentivize R&D and innovation and stimulate investments. The industry can set up teams and focus on innovation and engagements in Open RAN architecture, be it through opening up 3GPP interfaces, or utilizing O-RAN Alliance or Small Cell Forum common and open interfaces.”
The Open RAN effort is, indeed, battling a lot of criticisms and doubts, every now and then. Apart from the worry of it being a fragmented effort, there are also shadows cast on its lack of transparency (the veto power and privilege of defining specifications that some big members apparently enjoy). Add to that the insistence that it should be pushed into the 3GPP for getting more representatives from about 200 countries and expanding the gamut of collaboration with even more vendors and contributors from academia and others. There is also confusion about whether Open RAN would decrease costs or increase the open burden for an operator.
And, of course, if there are more vendors than one, how do we ensure water-tight security of the deployment, and how do customers find a precise throat to choke in case something conks off?
“The disaggregation of the RAN increases the number of individual elements and network connections. This has prompted concern from some that Open RAN could potentially be more insecure than its ‘closed’ predecessor due to a greater number of interfaces. In fact, Open RAN is more secure because you can test and see what is being delivered by third parties.” Baker from Mavenir confronts the security angle with a positive lens. But he is concerned about efficiency. “Running software-based virtualized Open RAN applications may not be as efficient as processing them on purpose-built hardware. Although this is indeed a legitimate concern, the ability of Open RAN components to scale, increase flexibility, and drive down costs through increased competition, innovation, and volume offset this limitation.”
Lights Out?
The purpose of something as ambitious, and disruptive, as this can be defeated if wide-scale adoption and acceptance of these standards does not happen – or happens in a random or sporadic way.
The current geo-political dynamics have often put a pause button on such initiatives. It is important that players and members have fewer moments of inertia, walk-outs, disagreements, and deadlocks.
In a recent September update though, the O-RAN ALLIANCE stated that it remains fully committed to its mission of delivering open, intelligent, virtualized, and fully interoperable RAN. It explained that the O-RAN ALLIANCE became aware of concerns regarding some participants that may be subject to U.S. export regulations, and has been working with O-RAN participants to address these concerns.
According to a media statement, the O-RAN Board has approved changes to O-RAN participation documents and procedures. While it is up to each O-RAN participant to make their own evaluation of these changes, O-RAN said that it is optimistic that the changes will address the concerns and facilitate O-RAN’s mission.
Titans of the industry manifested a positive thrust too. Andre Fuetsch, Chairman of the O-RAN ALLIANCE and Chief Technology Officer of AT&T said here that O-RAN is an open and collaborative global alliance operating in a way that promotes transparency and participation of our member companies in the development and adoption of global open specifications and standards. Similarly, Alex Jinsung Choi, Chief Operating Officer of the O-RAN ALLIANCE and SVP of Strategy and Technology Innovation, Deutsche Telekom added in the update, “We remain fully committed to working together in the alliance to achieve the goals and objectives of O-RAN as quickly as possible.
Open RAN is often viewed as a magic architecture that will level the playing field and result in changing supplier dynamics, quips Pongratz from Dell’Oro Group as he hastens to remind us that the reality is that it is more of a first step to open the door. “But this architecture does not change the fact that the carriers are operating in an environment with constrained topline growth and they ultimately need to optimize TCO/energy consumption/spectral efficiency and find partners that can help them tackle new opportunities while also supporting legacy networks.”
The growing torrent of data traffic and resulting push to build 5G and edge services are driving wider virtualization and cloudification of wireless networks adds Baker from Mavenir. “These push telcos to expand network footprint to meet the needs of enterprise and residential consumers, but also to accelerate the adoption of Open RAN to evolve to 5G. Over the next three years, Open RAN deployments will help create big new revenue opportunities for carriers and industry suppliers. For enterprises and consumers, this latest convergence of computing and communications will speed the delivery of transformative new services for everything from AI and IoT to autonomous vehicles, private mobile networks, and more. It’s a key link in bringing cloud power to the mobile edge.”
Well, amidst all the hope and hype, it is not enough to start something that adds a common denominator for the industry. It has to have life, traction, and consistency.
The secret to the few long-lasting light bulbs still present in this world is not as much in their filaments but in their being 'switched on.' They have been seldom turned off, and experts tell that a lot of wear and tear in incandescent light bulbs happens by the constant turning-on and off they go through. Because its filament gets heated and cooled again and again. That makes the material of the filament expand and contract a lot and, hence, creates stress cracks.
The filament has to break at some point and that means the light will burn out – when we switch it on and off again and again.
Whenever the industry gets together for a common cause of developing better, simpler, and universal standards – it has to stay together. Else, it’s a pointless huddle. Switching such efforts on and off again hurts the bulb. And the light. Inside and out.
pratimah@cybermedia.co.in