Strand Consult: The 10 Parameters of Open RAN; AT&T memo to FCC
Strand Consult is attempting to determine if, when, and how Open RAN (TIP project) and O-RAN (Alliance) will replace conventional RAN on a 1:1 basis without compromising the network quality, security, energy efficiency, and other important factors. Mobile operators have little ability to raise price, so operators must compete on network quality, coverage, and other factors. Here are few things to keep in mind.
In general, mobile ARPU is falling. In many countries, operators are trying to shift the focus away from price by competing on innovation and quality. For example, US mobile operators compete on the quality and coverage of their 4G and 5G networks. Mobile operators are focused on rolling out technology quickly, maintaining customer satisfaction, and ensuring quality of experience and other key performance indicators (KPIs). Chief technology officers, network managers, and other technical staff are laser focused on these KPIs and are loath to make changes to which would negatively impact these indicators.
In general, Strand Consult observes that what public affairs officials say about OpenRAN differs significantly from what network managers say.
Strand Consult’s 10 parameters to evaluate OpenRAN:
Strand Consult’s investigation has been guided by 10 parameters or questions to determine the value of OpenRAN. Here is what we’ve learned.
- Whether OpenRAN is a technical standard. The O-RAN Alliance is a private organization that develops technical specifications for OpenRAN. It should not be confused with the OpenRAN Policy Coalition which is a public affairs organization. The O-RAN Alliance is not a standards development organization (SDO), but rather an industrial collaboration that builds solutions on top of 3GPP specifications. While industrial cooperation is important, there can be no mobile networks the 3rd Generation Partnership Project (3GPP), an umbrella term for many standards organizations which develop protocols for mobile telecommunications and define the technological inputs for cellular networks. Companies like Rakuten develop their own corporate and proprietary concepts for OpenRAN. These concepts that do not necessarily follow a particular standard (3GPP) or O-RAN Alliance specification.
- Whether OpenRAN can replace Chinese equipment. Some mobile operators have suggested that OpenRAN is the way to avoid Huawei and ZTE in mobile networks. However other Chinese companies are deeply involved with OpenRAN technical specifications, product roadmap, and strategy. One founding member of the O-RAN Alliance is China Mobile, a state-owned company and the world’s largest mobile operator with 950 million subscribers and 450,000 employees. The O-RAN Alliance has more than 40 Chinese member companies, many of which government-owned and military aligned (See Strand Consult’s research note 44 Chinese companies have joined the OpenRAN effort, a strategy to reduce Huawei’s presence in 5G). The Chinese members of 0-RAN Alliance outnumber the Europeans. China Mobile’s Chih-Lin has veto power over the organization. China Mobile leads or co lead 8 of the 9 O-RAN Alliance’s working groups either as chairman or vice-chairman.
- OpenRAN and 5G innovation. OpenRAN proponents claim it will have a revolutionary impact on 5G, however reports suggest that large scale deployment of OpenRAN won’t happen until 2025. This means that OpenRAN cannot replace existing RAN on a 1:1 basis today. The 5G networks rolled out today use the standards from 3GPP Release 15 with increased functionality forthcoming in Releases 16 and 17 within two years. There are more than 144 3GPP-5G commercial networks deployed but only one proprietary OpenRAN 5G network (Rakuten). If OpenRAN is to increase 5G innovation, it must evolve faster than 5G itself. Presently it is not on par with the 5G standards defined years ago. It is difficult to see how OpenRAN can catch up when the significant resources already supporting the 3GPP standards timeline.
- OpenRAN and 5G deployment. Today mobile operators are rolling out 5G at a faster than 4G and even fast than 3G. In practical terms, 5G networks already built in 2019 and 2020 and those to be built in 2021 and 2022 already have the standards roadmap in place. If OpenRAN can’t catch up by 2025, operators have only two choices, delay 5G until 2025 (when 6G will start to take root) or replace their 5G equipment in 4 to 5 years. OpenRAN may be too little, too late for 5G operators.
- OpenRAN and vendor diversity. OpenRAN proponents claim that it will create more competition in the network equipment market. The 5G network equipment vendor market has many vendors and segments. Omdia details more than a dozen full-service providers with additional providers in segments for antennas, basebands, remote radios, small cells, macro cells, phase shifters and so on. This idea that there are not many vendors for 5G equipment was likely created by Huawei to deter the security reviews and subsequent restrictions imposed on the military-aligned company. If anything, the restrictions on Huawei have helped to open the door to new equipment vendors which could not compete because of Huawei’s predatory pricing and anti-competitive tactics. For example, Samsung has quickly gained market share and is supply 5G rollouts in the US, Australia, and other countries.
- OpenRAN and network equipment cost. OpenRAN proponents suggest that it can lower the cost of network equipment. The cost competitiveness of OpenRAN versus RAN is not yet known. It may be that some OpenRAN providers can offer equipment more cheaply on some parameters, but the cost advantage may not be significant when considering all the costs such as supply, availability, energy consumption, security, warranty, network integration, equipment matching, new contracts and service level agreements etc. However, operators frequently reduce their number of vendors so that they can enjoy lower unit costs with volume purchasing, the company BUYIN is a great case. For an operator’s perspective, check out the comments from Neil McRae, Managing Director and Chief Architect at BT (Scroll to minute 51 minute in the video). McRae explains that when he took his job, he inherited a network portfolio with 50 vendors. He subsequently reduced it to 4 vendors and saved £1 billion in 3 years. He observed that too many vendors not only increased cost, it increased complexity. He is wary of notions of “open architectures” which require managing portfolios of 5-50 vendors. He noted that vendor reduction increased shareholder value and that he would pursue the same strategy again.
- OpenRAN and security. OpenRAN proponents suggest that OpenRAN technology and the “unbundling” of 5G hardware and software is the means reduce reliance on Huawei and hence greater security. However, it is not clear how trading one known insecure Chinese vendor for 50 unknown Chinese vendors is the path to greater security. The issue of backdoors is omnipresent on all Chinese hardware given the country’s disposition and associated intelligence and surveillance policies. Moreover, it is not clear how security is improved when network owners must vet not one, but multiple new OpenRAN vendors. The time and cost to perform this review would seem to be multiplied by the number of vendors the operators takes on.
- OpenRAN and energy efficiency. Energy efficiency is an increasingly important issue for mobile operators which expect to compete on carbon reduction strategies. Naturally if OpenRAN could offer a greener solution, that would be an advantage. However, it is reported that many OpenRAN installation use the Intel X86 processor, which is less efficient than specialized RAN chips. If anything, energy consumption could increase if signals must traverse a multitude of mix and match components instead of a single end-to-end system designed with energy efficiency in mind. To reduce energy, Apple developed tits own processer as an alternative to Intel X86.
- OpenRAN and Rakuten. The media has promoted a supposed Rakuten success story with OpenRAN success. However, Rakuten is not offering open-source tools, but rather proprietary OpenRAN solution. It offers this through a freemium model in which free service is offered for a period, and operators pay down the line. Some companies have success with freemium and loss leader models, but typically they need scale.
- OpenRAN and the indigenous movement. OpenRAN has been promoted as a way to support domestic innovation like India or Brazil or what Germany’s Economic Minister calls for European-only actors in 5G. Curiously many of these calls are coupled with operator strategies to keep Huawei equipment in place because OpenRAN will not be ready for some years. Policymakers have also pursued subsidies and other financial incentives to support local OpenRAN startups which may design the equipment in their respective country but manufacture it in China. Unfortunately, production in China and with Chinese partners could compromise security, as the Supermicro case demonstrates.
The many problems that OpenRAN is purported to solve is impressive. In fact, I have to go back to 3G in 2000 to find the level of hype observed today with OpenRAN. Indeed, the Huawei problem is so serious that people are desperate for a solution. However, in the enthusiasm for the OpenRAN solution, too many want to look past the inconvenient reality that China is shaping much of the Open RAN future particularly through the O-RAN Alliance.
It is important to develop secure alternatives to Huawei, but this is not a reason to oversell OpenRAN. While it may be commendable to pursue the goals proffered by OpenRAN proponents, the actual impact of OpenRAN must be measured by real world facts and experience.
The questions remain how OpenRAN will affect the CAPEX and OPEX mobile operators in the short, medium and long term and whether operators will buy OpenRAN as a serious 1:1 alternative to standard RAN in Paris, London, Berlin, Madrid, New York, Sao Paulo and Copenhagen. It seems that OpenRAN is falling short of expectations.
Separately, AT&T told the FCC it plans to begin adding open RAN-compliant equipment into its network “within the next year.”
That puts AT&T on roughly the same timeframe as Verizon. Verizon’s SVP Adam Koeppe told Light Reading earlier this year that the operator’s 5G hardware vendors – Ericsson, Samsung and Nokia – will begin supplying open RAN-compliant equipment starting later this year. And he expects that the bulk of their equipment shipments to Verizon will comply with open RAN specifications by next year. AT&T told the FCC it expects to implement similar changes into its own network.
“The challenge for an operator shifting to any open network architecture, including but not limited to O-RAN, will be maintaining network reliability, integrity and performance for customers during the transition,” the operator wrote in a filing. “For our part, AT&T serves multiple customer groups, with varied and often complex, service requirements. As we introduce O-RAN into our network, our goal will be maintaining the same high level of performance at scale. We are actively working in this direction.”
AT&T to FCC: Promoting the Deployment of 5G Open Radio ) Access Networks – GN Docket No. 21-63
3 thoughts on “Strand Consult: The 10 Parameters of Open RAN; AT&T memo to FCC”
Insightful analysis of Open RAN. It will be interesting to see how it develops.
T-Mobile’s top networking executive has expressed serious reservations about Open RAN technology. “It’s not ready for prime time for us,” T-Mobile’s Neville Ray said at an investor event late last year. “Today, I buy a solution from an Ericsson or a Nokia or a Samsung; it’s warrantied. I have one neck to choke. If something goes wrong, I know where to go. In an O-RAN environment, you have to do a lot more heavy lifting as the operator.”
Last week, in new comments to the FCC, both operators outlined their concerns about a technology that has so far been embraced by new market entrants like Japan’s Rakuten and Dish Network in the US. “While implementation of open RAN may ultimately be feasible in ‘greenfield’ deployments, deployment of open RAN in existing networks will be challenging,” T-Mobile explained in its new FCC filing. The operator continued: “Open RAN is not ready for full commercial deployment for several reasons. For example, methods must be developed for implementing regulatory mandates, like 911, across a multi-vendor environment.”
Security concerns, technical challenges T-Mobile specifically pointed to open RAN security, a topic that has generated plenty of debate among companies ranging from Ericsson to Dish Network. “In addition to recognizing the potential benefits of open RAN to network security, the commission should recognize that open RAN may, in fact, create security risks,” T-Mobile argued. “Open networks are more prone to cyber security threats because they introduce additional interfaces, additional functions, and functional splits, which expand the surface for potential threats. The fact that open RAN relies on open source software increases the number of potential entry points for security breached.”
Continued T-Mobile: “Moreover, there is no overall security assessment or requirement – whatever security mechanism exists are distributed and may not effectively flow through a network.” In its own filing, Verizon served up open RAN concerns as well.
“Access to Standards Essential Patents on Fair, Reasonable and Non-Discriminatory (FRAND) terms is another important factor as it will limit the uncertainty around intellectual property right infringement challenges that delay the development of new technologies such as open RAN,” Verizon explained. Verizon also briefly touched on some technological issues facing open RAN.
“Technical challenges generally relate to 5G wireless network-specific constraints that do not exist for traditional virtualized software deployments, such as extremely low latency requirements (potentially requiring workload-specific hardware accelerators), support for 5G-specific communication protocols rather than HTTP, energy efficiency, and security,” the operator wrote.
Let open RAN run free of government interference Despite such concerns, both Verizon and T-Mobile expressed support for broader open RAN principles regarding vendor diversity and open interfaces. “As T-Mobile has observed elsewhere, there can be benefits to open network architectures,” the operator wrote. “They can, for instance, create a larger and more diverse product ecosystem, increase competition and reduce the over-reliance on any one vendor – particularly vendors that the US has found pose significant national security risks. They can also promote vendor diversity and potentially create price and feature competition.” Both operators also strongly urged regulators at the FCC to abstain from issuing mandates around the use of open RAN technology. The carriers strongly argued that the wireless market should be left to embrace or reject open RAN without government interference.
Comments are closed.