Moving Toward Open Networks

This August, the Ministry of Economic Affairs (MOEA), in partnership with U.S. tech multinational Cisco Systems, announced the launch of a 5G “open lab.” The initiative creates a platform where Taiwanese suppliers can apply to develop a range of 5G-related hardware and software, including small-cell base stations, IoT devices, and network switches, via Cisco’s cloud-based mobile network.

The purpose of the open lab is to forge a local supply chain for 5G network equipment and ultimately present a more competitive, cost-efficient alternative to the current standard of expensive, proprietary systems supplied by a few major telecom equipment providers. Proponents of the open infrastructure – also known as O-RAN, or open radio access network – say that it simplifies operations and procurement processes, creates economies of scale, and reduces the total cost of ownership of mobile networks.

“Under the old model, Taiwanese companies had no clear opportunities to get involved in network buildouts,” says Cisco Systems Taiwan Regional Manager Jeffrey Wang. “But through this project with the MOEA, we can begin erecting our 5G packet core software and linking it with small-cell base stations being produced by Taiwan manufacturers.” He notes that several local firms have applied to participate in the lab, some of which have already been accepted.

Cisco already has experience with developing open infrastructure for 5G networks. In 2018, it partnered with Japanese conglomerate Rakuten’s mobile carrier subsidiary to develop the company’s greenfield mobile network in Japan. Cooperating with over a dozen other companies, Cisco provided hardware and software for the network’s data center architecture, as well as packet core software and cloud technology.

Another major telecom operator that has recently undertaken efforts to develop an O-RAN network is AT&T, a decision that has set it apart from other U.S. service providers. The U.S. Federal Communications Commission has expressed hope that this model may also fill the gap left by the Trump administration’s decision to ban Huawei network equipment.

Observers say that Taiwan’s existing strengths in the OEM/ODM manufacturing of ICT products make it a good environment to begin cultivating a local O-RAN industry.

“Taiwan has the ability to supply 5G user terminals, 5G base stations, transmission networks, 5G nuclear networks, and other hardware devices built for the 5G open network architecture,” says Su Wei-gan, an analyst at the semi-official Market Intelligence & Consulting Institute. “For example, Taiwan has competitive production capacity in complete machines or components for use in servers and networking equipment.”

Yet none of Taiwan’s telecom operators is considering an open architecture for their own 5G networks, at least not yet.

“I think that O-RAN is definitely the future of communications systems technology, but it hasn’t gotten to the point where its price-performance ratio is comparable with the existing systems to justify mass deployment,” says Taiwan Mobile President Jamie Lin. “If we adopted it now, our costs would be so much higher than our competitors. We’d be putting ourselves in an awkward position because we couldn’t charge more for our services – people wouldn’t pay for that.”

Nevertheless, Lin and others say they can see a change taking place within the next five to ten years, if not less. Chunghwa Telecom President Kuo Shui-yi says that companies will likely start reconsidering the overall composition of their networks over the next few years as standalone 5G networks – those not built on top of the existing 4G/LTE infrastructure – become the norm.

Wang of Cisco says that for the time being, telecom operators may opt to use O-RAN for their enterprise clients’ private networks, which are smaller in scale and can allow for a wider margin of error. “It’s kind of like training troops,” he says. “Once they’re done with the training and see that there’s no problem with this new system, they’ll be willing to start trying it with their public networks.”

Still, in Wang’s view the time to get started is now. “We cannot wait,” he says. “If we don’t do it at this moment, we won’t be ready when the time comes. What we need to do now is begin attracting software talent, accumulating experience, and developing necessary techniques and skills.”

To generate more enthusiasm for the project, Wang suggests providing incentives to companies willing to try the O-RAN model.

“For example, incentivize them to use a certain number of locally produced base stations within a three-year period,” he says. “The government has collected so much money from these service providers. Giving some of it back in the form of subsidies to jumpstart this transformation could be a very good move.”