FAQ: How Can Open RAN Drive Efficiency, Enable Agility and Enhance 5G Adoption?
In this article
The economics of 5G are forcing network operators to deploy virtualized, multi-vendor solutions that can scale to help optimize infrastructure investments. And this disaggregation has made its way all the way to operators' Radio Access Network.
Naturally, operators have questions. Can Open RAN really save opex? How can Open RAN be deployed from a brownfield and greenfield perspective? Will architectures be vertical or application specific?
WWT experts have participated in several industry events focused on Open RAN and such questions. To better inform the broader service provider industry, we thought it would be beneficial to share those questions along with our answers here.
If you have a question of your own regarding Open RAN or how service providers can modernize their networks to deliver 5G services to end customers, please contact us and we will be sure to add questions to this list as appropriate.
Q: What advantages does Open RAN present to operators?
A: Open RAN is a virtualized RAN, but on general-purpose hardware that replaces legacy, single-vendor RAN solutions with stable, scalable and 5G-ready architectures. Operators can also take advantage of upgrading aging proprietary boxes. Instead of updating a proprietary box that is at end of life with a new proprietary box, operators can opt for an Open RAN solution. Operators can then leverage their own engineers that know the details of their market well with the help of system integrators to facilitate their transition to open RAN.
Q: How do you envision Open RAN changing the role of the Communications Service Provider (CSP)?
A: Open RAN is operator driven and enables CSPs to drive their own network, including putting together best-of-breed networks. System integrators play a key role in validating and managing the multi-vendor solutions that will help CSPs optimize cost and shrink the time it takes to bring solutions to market.
Q: As the RAN is decomposed and disaggregated, do you foresee a multi-vendor RAN that is unique to the key applications or vertical it serves?
A: One of the benefits of Open RAN is achieving a best-of-breed multi-vendor network that enables operators to bring innovative solutions specific to different industry verticals from RAN vendors, third parties and their own R&D teams. So yes, Open RAN will usher in unique multi-vendor architectures tailored to meet the needs of specific applications of industries.
Q: What is the difference in scope and purpose between the Open RAN Policy Coalition and the O-RAN Alliance?
A: The Open RAN Policy Coalition was formed to advocate for government policies supporting the development and adoption of open and interoperable solutions in the Radio Access Network (RAN). WWT is uniquely positioned as the only system integrator that is part of the coalition. The O-RAN Alliance was founded by operators to clearly define requirements and help build a supply chain eco-system to realize the benefits of open RAN by bringing openness and intelligence to the radio network.
Q: What is the role of the Telecom Infra Project?
A: Telecom Infra Project — or TIP — is a community of companies and organizations dedicated to driving open, disaggregated and standards-based technology solutions, such as Open RAN.
Open RAN is just one area of its areas of focus. The group has several other product interests, all aligned to three parts of the network: Access, transport, and core and services.
Q: How can we ensure interoperability of the various vendor solutions? Will there be a certification or common labs for mobile operators?
A: System integrators like WWT provide a lab environment to validate multi-vendor solutions. WWT works with multiple vendors to certify such mix-and-match solutions and makes lab environments available directly for service providers. This enables service providers to validate and test their own custom configurations.
Q: There are some concerns the shift towards Open RAN with multiple RAN DU, CU, hardware and software vendors could result in higher integration costs. Is this a valid concern?
A: This is an area where system integrators like WWT can play a significant role. System integrators can drive interoperability testing and validation. Integrators that are experienced and focused on this can help manage integration costs and bring new innovations to market quicker.
Q: What is the relationship between open vRAN and general processing hardware such as CPU or GPU?
A: Open RAN aims to bring cloud-scale economics to the RAN. One of the enablers for this is being able to use off-the-shelf hardware with CPU and GPU that result in a network that meets industry reliability and availability standards in a cost-efficient manner.
Q: How can a service provider closely tied with OEMs and their own engineering take advantage of the reference architecture and adjust to align with their own unique requirements?
A: New technologies like 5G present a great opportunity for operators to consider and adopt Open RAN solutions that will help them bring new technologies to market faster and reduce capex and open costs.
Q: Would operators license their RAN software (DU and CU) from one vendor or from multiple vendors? If it is one vendor only, isn't the operator locked into this one vendor's roadmap?
A: The general principal is that "best of breed" vendors would be used in each function. Each function is fungible (i.e. any two vendors are mutually interchangeable). It's possible that one vendor could be advanced enough to be selected by an operator in multiple areas, however that operator can still surgically excise that vendor in any or all those functions in a short timeframe.
Q: How do you see collaboration on system release management working in practice?
A: System integrators such as WWT have significant experience from IT cloud and telco cloud multi-vendor integration and are actively developing solutions in the Open RAN space.
Q: What are your thoughts about additional costs from system integrators and network operation, and maintenance when mixing different vendors?
A: As per the previous question, Telecom Infra Project has defined re-usable System Release Management policies. Early adopters can either integrate themselves or engage System Integrators such as WWT. At the end of the integration, a 'Product" is created which can be licensed from the TIP Exchange. i.e. Build once, deploy many times.
Q: Overall ownership of the network seems more complex. How much appetite do operators have to being the "risk owners"?
A: Some larger operators may integrate themselves or engage system integrators such as WWT. At the end of the integration, a "product" is created, which will be maintained in a similar way to OEMs today. So, a risk partnership will mitigate those concerns.
Q: Do you see all solutions in the Open RAN space being built around the O-RAN Alliance standards? Or do you see other standards competing going forward? Is the industry coming together around the O-RAN Alliance Standards?
A: The majority of Operators, vendors and system integrators are members of the O-RAN Alliance, Open RAN Policy Coalition and Telecom Infra Project, thereby ensuring broad consistency within them. Each plays a specific role in accelerating Open RAN.
Q: RAN vendor consolidation happened because of the market pressure and efficiency race forced by carriers. Why is this a problem? The aviation business is by far bigger and it has only two suppliers, and no one is proposing an open airplane. Is this a real market need or is this a geopolitical collateral effect?
A: It's hard to compare the two sectors since the cycle durations are very different. Mobile infrastructure is typically swapped every five to seven years, and historically two or three vendors were used per operator. For 3G, there were eight to 10 strong vendors. For 4G launch it was four or five vendors, and now two vendors are a step function of scale ahead of other available vendors. I'd also argue that the long-time separation of airframe and engine supply puts aerospace at the forefront of open interfaces. In the wider context, a combination of technology- and timescale-aggressive greenfield cloud-native operators has thrust Open RAN onto the big stage.
Q: How can I identify the number of bits for CU/DU and sector in Open RAN CU/DU split?
A: There is an excellent Small Cell Forum resource that shows the transport requirements for a typical radio with the related user and air-interface data rates.
Q: Most 5G deployments in Europe are distributed RAN (DRAN). As Open RAN targets the virtual RAN and distributed RAN, and 5G TDD requires coverage increases, should we see this as a big opportunity for Open RAN?
A: European operators are commencing their Open RAN deployments in rural areas. DRAN will be the initial deployment schema to allow them to baseline the performance. If clusters can be created then more centralized deployments will allow pooling/trunking gains as well as centralized scheduler and other capacity and performance gains. These may coincide with the expected growth of suburban and urban Open RAN deployments.
Q: Do Open RAN elements need a special synchronization clock signal? Or is it possible to use existing clock signals?
A: Open RAN is simply RAN with virtualized/cloud-native disaggregation and open interfaces. Standard 3GPP requirements will support in-band, out-of-band or GPS synchronization at the operator's choosing.
Q: Does time synchronization require special or customized hardware? Or will Open RAN rely on a fronthaul transport network to provide the clock to the DU and RU?
A: As above, standard disaggregation techniques apply. Four example synch topologies are shown in figure 11 of this NGMN document. Fronthaul synch is underpinned by Time Sensitive Networking, also linked in the references.
Q: When will Open RAN be able to deliver a competitive solution with carrier-grade performance?
A: Traditional RAN systems are no longer the benchmark, since they too are evolving to a cloud-native virtualized/containerized disaggregated architecture. We have heard for several years about the impending arrival of NW slicing and URLLC services delivering new value to operators, enterprises and consumers. These will require the same functional decomposition and COTS implementation as used by Open RAN, but will now lack the wide vendor access which Open RAN can tap into for orchestration and automation. Without those, it will be appliance RAN, which will see a higher opex rise in order to deliver the same capabilities. Regarding interoperability, existing vendor RAN supports a very narrow band of interoperability. Open RAN is designed for a wide spectrum of interoperability, and industry and governmental efforts are being made to demonstrate optimum interworking performance. For example, the Telecom Infra Project has adopted a validate once, share and deploy many stance, and UK Government's Vendor Diversification Strategy is currently planning investment in interworking Labs similar to WWT's Advanced Technology Center.
Q: One of the main issues facing the industry is the requirement of multiband radios and massive MIMO capabilities requirements in dense urban areas. When do you foresee this ecosystem taking shape?
A: Vodafone and TIP recently announced the results of their oRU RFI. They announced five leading vendors for Multi-band RRH (including Fujitsu, Comba and Mavenir) and five for Massive MIMO (including NEC and GigaTera). Availability for RFI assessment is Q1 2021, and the NEC and Fujitsu products are already visible through their Rakuten and Dish deployment.
Q: How exactly will Open RAN reduce operating expenses?
A: Networks are becoming increasingly complex. The self-driving nature of Open RAN will be able to leverage new learning-based technologies that automate operational network functions and reduce opex.
In combination with Open RAN's open interfaces, AI-optimized closed-loop automation will be achievable and will enable a new era for network operations. Open interfaces are essential to enabling small vendors or operators to quickly introduce their own services. They also enable operators to customize the network to suit their own unique needs.
All these aspects of O-RAN can reduce operating expenses.
Q: From an operational standpoint, who is envisaged as being responsible for delivering a similar level of network performance as currently in the mature macro networks from the operator's standpoint?
A: There are two likely owners to this assurance: Either the operator or a systems integrator. System Release Verification is a term used to describe this, releasing validated and supported multi-vendor products.
Q: How do you expect to augment dedicated network security when offering Open RAN to enterprises?
A: Open interfaces actually make it easier for service providers to monitor the security of their networks, and the generally Cloud-Native implementation of Open RAN brings containerization security benefits.
Q: Do you think innovation would be better in a multivendor ecosystem even though each of these vendors is smaller than the traditional vendors? How much time do you think it will take for innovation to improve the telecommunications market in a multivendor scenario?
A: Twenty years ago OEMs such as Ericsson and Nokia wanted to control the entire experience from handset to core router. Today they have acknowledged that such an end-to-end strategy creates internal prioritization of R&D funds, which yields weak "me-too" products tying up capital and filling warehouses. In recent years, this evolved into a series of paper roadmaps with R&D invested in products which were never manufactured, due to the catch 22 of, "I would order it, but cannot wait six to nine months for you to put it into production."
In Open RAN, the best-of-breed ethos encourages vendors to do one thing and do it well. A strategic focus and targeting precious R&D investment in winners delivers innovation and efficiency hand-in-hand.
Q: Do you see Open RAN evolving into RAN-as-a-service that utilizes industry-leading cloud platforms?
A: The RAN is evolving to a RAN-as-a-service model deployed on different kinds of cloud platforms. The Evolved Packet Core (EPC) and IP Multimedia Subsystem (IMS) are already being delivered as a cloud-based service. Different components of the RAN are being developed as cloud-native containerized services, which facilitates RAN-as-a-service deployments.
Q: For the O-RU commercial launch, what kind of performance issues have been observed from current available offers?
A: From this recent live network report on Rakuten, it appears that Open RAN performance is very high, it is the use of roaming partners while deployment of their own network continues Disruption with O-RAN: Rakuten's mobile experience so far. Anecdotally this has much in common with prior new market entrants such as 3 or France's Free Mobile who initially used rival network coverage.
Q: How do you see Open RAN evolving for in-building use? How can it enable localized neutral-host networks?
A: Open RAN is applicable to multiple use cases including in-building solutions. In the case of in-building or venue use cases, open RAN plays a key role in enabling neutral host solutions that are less cumbersome, less expensive and easier to manage. This is because openness makes sharing the multiple layers of the network by multiple operators easier.
Q: Tier 1 operators are evaluating Open RAN in low average revenue per user (ARPU) parts of the network. How does a brownfield operator enter ORAN?
A: Today's Open RAN networks deliver capacity, reliability and availability that meet industry standards. This means Open RAN can be evaluated and deployed in multiple parts of the network. Different operators will follow different adoption paths based on their business needs.
Q: Do you believe mobile network operators (MNO) will be willing to take on the risk and responsibility of deploying Open RAN and the operational complexity it could bring?
A: The traditional monolithic RAN solution approach no longer meets today's quickly evolving network requirements. These requirements are driven by the need to bring solutions to market quickly, cost efficiently and with minimal hardware swaps. MNO decision makers are increasingly looking at Open RAN to address these needs and stay competitive.
Q: Without fiber transport, what is the base of transport network (MH)? How does it keep transport network throughput and latency?
A: Alternative transport network options like Microwave can be used. The transport network will have to be dimensioned and configured to meet the end-to-end performance requirements.
Q: How will the profit pool be shared by players in the Open RAN ecosystem?
A: The Open RAN ecosystem creates partnerships between system integrators, OEMs and operators with a common goal of getting new technologies to market in a cost-efficient manner. This approach creates business opportunities to all stakeholders including small innovative startup companies.